@inproceedings{
 title = {Understanding ML Driven HPC: Applications and Infrastructure},
 type = {inproceedings},
 year = {2020},
 pages = {421-427},
 websites = {http://arxiv.org/abs/1909.02363},
 month = {3},
 publisher = {Institute of Electrical and Electronics Engineers (IEEE)},
 day = {20},
 id = {d2edc482-37d9-3eba-83e0-3c88dac06d5f},
 created = {2020-04-21T20:10:55.142Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.662Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Jha2020},
 private_publication = {false},
 abstract = {We recently outlined the vision of "Learning Everywhere" which captures the possibility and impact of how learning methods and traditional HPC methods can be coupled together. A primary driver of such coupling is the promise that Machine Learning (ML) will give major performance improvements for traditional HPC simulations. Motivated by this potential, the ML around HPC class of integration is of particular significance. In a related follow-up paper, we provided an initial taxonomy for integrating learning around HPC methods. In this paper, which is part of the Learning Everywhere series, we discuss "how" learning methods and HPC simulations are being integrated to enhance effective performance of computations. This paper identifies several modes --- substitution, assimilation, and control, in which learning methods integrate with HPC simulations and provide representative applications in each mode. This paper discusses some open research questions and we hope will motivate and clear the ground for MLaroundHPC benchmarks.},
 bibtype = {inproceedings},
 author = {Jha, Shantenu and Fox, Geoffrey},
 doi = {10.1109/escience.2019.00054}
}

We recently outlined the vision of "Learning Everywhere" which captures the possibility and impact of how learning methods and traditional HPC methods can be coupled together. A primary driver of such coupling is the promise that Machine Learning (ML) will give major performance improvements for traditional HPC simulations. Motivated by this potential, the ML around HPC class of integration is of particular significance. In a related follow-up paper, we provided an initial taxonomy for integrating learning around HPC methods. In this paper, which is part of the Learning Everywhere series, we discuss "how" learning methods and HPC simulations are being integrated to enhance effective performance of computations. This paper identifies several modes --- substitution, assimilation, and control, in which learning methods integrate with HPC simulations and provide representative applications in each mode. This paper discusses some open research questions and we hope will motivate and clear the ground for MLaroundHPC benchmarks.

@inproceedings{
 title = {Learning Everywhere: A Taxonomy for the Integration of Machine Learning and Simulations},
 type = {inproceedings},
 year = {2020},
 pages = {439-448},
 websites = {http://arxiv.org/abs/1909.13340},
 month = {3},
 publisher = {Institute of Electrical and Electronics Engineers (IEEE)},
 day = {20},
 id = {fc4ff380-f9ff-3388-8b55-902ff974390c},
 created = {2020-04-21T20:10:55.219Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.838Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Fox2020},
 private_publication = {false},
 abstract = {We present a taxonomy of research on Machine Learning (ML) applied to enhance simulations together with a catalog of some activities. We cover eight patterns for the link of ML to the simulations or systems plus three algorithmic areas: particle dynamics, agent-based models and partial differential equations. The patterns are further divided into three action areas: Improving simulation with Configurations and Integration of Data, Learn Structure, Theory and Model for Simulation, and Learn to make Surrogates.},
 bibtype = {inproceedings},
 author = {Fox, Geoffrey and Jha, Shantenu},
 doi = {10.1109/escience.2019.00057}
}

We present a taxonomy of research on Machine Learning (ML) applied to enhance simulations together with a catalog of some activities. We cover eight patterns for the link of ML to the simulations or systems plus three algorithmic areas: particle dynamics, agent-based models and partial differential equations. The patterns are further divided into three action areas: Improving simulation with Configurations and Integration of Data, Learn Structure, Theory and Model for Simulation, and Learn to make Surrogates.

@article{
 title = {Object Classifications by Image Super-Resolution Preprocessing for Convolutional Neural Networks},
 type = {article},
 year = {2020},
 pages = {476-483},
 volume = {5},
 publisher = {ASTES Journal},
 id = {39add601-2e46-3c7e-837a-3e9f7f7cdf27},
 created = {2020-04-21T20:10:55.395Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.928Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Na2020},
 private_publication = {false},
 bibtype = {article},
 author = {Na, Bokyoon and Fox, Geoffrey C},
 doi = {10.25046/aj050261},
 journal = {Advances in Science, Technology and Engineering Systems Journal},
 number = {2}
}

@article{
 title = {Targeted High-Resolution Structure from Motion Observations over the M w 6.4 and 7.1 Ruptures of the Ridgecrest Earthquake Sequence},
 type = {article},
 year = {2020},
 pages = {1-9},
 websites = {https://pubs.geoscienceworld.org/ssa/srl/article-pdf/doi/10.1785/0220190274/4967856/srl-2019274.1.pdf},
 id = {64a06f32-b167-3e95-a540-3ea79486a86f},
 created = {2020-04-21T22:47:54.772Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.443Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Donnellan2020},
 private_publication = {false},
 abstract = {Cite this article as Donnellan, A., G. Lyzenga, A. Ansar, C. Goulet, J. Wang, and M. Pierce (2020). Targeted High-Resolution Structure from Motion Observations over the M w 6.4 and 7.1 We carried out six targeted structure from motion surveys using small uninhabited aerial systems over the M w 6.4 and 7.1 ruptures of the Ridgecrest earthquake sequence in the first three months after the events. The surveys cover approximately 500 × 500 m areas just south of Highway 178 with an average ground sample distance of 1.5 cm. The first survey took place five days after the M w 6.4 foreshock on 9 July 2019. The final survey took place on 27 September 2019. The time between surveys increased over time, with the first five surveys taking place in the first month after the earthquake. Comparison of imagery from before and after the M w 7.1 earthquake shows variation in slip on the main rupture and a small amount of distributed slip across the scene. Cracks can be observed and mapped in the high-resolution imagery, which show en echelon cracking, fault splays, and a northeast-striking conjugate fault at the M w 7.1 rupture south of Highway 178 and near the dirt road. Initial postseismic results show little fault afterslip, but possible subsidence in the first 7-10 days after the earthquake, followed by uplift.},
 bibtype = {article},
 author = {Donnellan, Andrea and Lyzenga, Gregory and Ansar, Adnan and Goulet, Christine and Wang, Jun and Pierce, Marlon},
 doi = {10.1785/0220190274},
 journal = {Seismological Research Letters}
}

Cite this article as Donnellan, A., G. Lyzenga, A. Ansar, C. Goulet, J. Wang, and M. Pierce (2020). Targeted High-Resolution Structure from Motion Observations over the M w 6.4 and 7.1 We carried out six targeted structure from motion surveys using small uninhabited aerial systems over the M w 6.4 and 7.1 ruptures of the Ridgecrest earthquake sequence in the first three months after the events. The surveys cover approximately 500 × 500 m areas just south of Highway 178 with an average ground sample distance of 1.5 cm. The first survey took place five days after the M w 6.4 foreshock on 9 July 2019. The final survey took place on 27 September 2019. The time between surveys increased over time, with the first five surveys taking place in the first month after the earthquake. Comparison of imagery from before and after the M w 7.1 earthquake shows variation in slip on the main rupture and a small amount of distributed slip across the scene. Cracks can be observed and mapped in the high-resolution imagery, which show en echelon cracking, fault splays, and a northeast-striking conjugate fault at the M w 7.1 rupture south of Highway 178 and near the dirt road. Initial postseismic results show little fault afterslip, but possible subsidence in the first 7-10 days after the earthquake, followed by uplift.

@article{
 title = {Ground Deformation Data from GEER Investigations of Ridgecrest Earthquake Sequence},
 type = {article},
 year = {2020},
 month = {2},
 publisher = {Seismological Society of America (SSA)},
 day = {19},
 id = {2a6492d1-c84b-31c0-ab5b-53df706d065d},
 created = {2020-04-21T22:47:55.075Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.281Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Brandenberg2020},
 private_publication = {false},
 abstract = {Following the Ridgecrest earthquake sequence, consisting of an M 6.4 foreshock and M 7.1 mainshock along with many other events, the Geotechnical Extreme Events Reconnaissance association deployed a team to gather perishable data. The team focused their efforts on documenting ground deformations including surface fault rupture south of the Naval Air Weapons Station China Lake, and liquefaction features in Trona and Argus. The team published a report within two weeks of the M 7.1 mainshock. This article presents data products gathered by the team, which are now published and publicly accessible. The data products presented herein include ground-based observations using Global Positioning System trackers, digital cameras, and hand-measuring devices, as well as unmanned aerial vehicle-based imaging products using Structure from Motion to create point clouds and digital surface models. The article describes the data products, as well as tools available for interacting with the products.},
 bibtype = {article},
 author = {Brandenberg, Scott J. and Stewart, Jonathan P. and Wang, Pengfei and Nweke, Chukwuebuka C. and Hudson, Kenneth and Goulet, Christine A. and Meng, Xiaofeng and Davis, Craig A. and Ahdi, Sean K. and Hudson, Martin B. and Donnellan, Andrea and Lyzenga, Gregory and Pierce, Marlon and Wang, Jun and Winters, Maria A. and Delisle, Marie-Pierre and Lucey, Joseph and Kim, Yeulwoo and Gallien, Timu W. and Lyda, Andrew and Yeung, J. Sean and Issa, Omar and Buckreis, Tristan and Yi, Zhengxiang},
 doi = {10.1785/0220190291},
 journal = {Seismological Research Letters}
}

Following the Ridgecrest earthquake sequence, consisting of an M 6.4 foreshock and M 7.1 mainshock along with many other events, the Geotechnical Extreme Events Reconnaissance association deployed a team to gather perishable data. The team focused their efforts on documenting ground deformations including surface fault rupture south of the Naval Air Weapons Station China Lake, and liquefaction features in Trona and Argus. The team published a report within two weeks of the M 7.1 mainshock. This article presents data products gathered by the team, which are now published and publicly accessible. The data products presented herein include ground-based observations using Global Positioning System trackers, digital cameras, and hand-measuring devices, as well as unmanned aerial vehicle-based imaging products using Structure from Motion to create point clouds and digital surface models. The article describes the data products, as well as tools available for interacting with the products.

@article{
 title = {Self‐regulated studying behavior, and the social norms that influence it},
 type = {article},
 year = {2020},
 pages = {10-21},
 volume = {50},
 websites = {https://onlinelibrary.wiley.com/doi/abs/10.1111/jasp.12637},
 month = {1},
 publisher = {Blackwell Publishing Ltd},
 day = {6},
 id = {7b894dc5-bd47-32f9-9795-5a7ef61b27a8},
 created = {2020-04-22T21:44:57.102Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.716Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Eyink2020},
 private_publication = {false},
 abstract = {Teachers commonly use injunctive norms when telling students what they should be doing. But researchers find that sometimes descriptive norms, information about what others are actually doing, are more powerful influencers of behavior. In the present work, we examine which norm is more effective at increasing self-regulated studying and performance in an online college course across two semesters. To do this, we randomly assigned 751 undergraduate Introductory Psychology students to receive email messages at the start of every content unit that either contained descriptive norms, injunctive norms, information about the course, or a no message control. We found that injunctive norms increased study behaviors aimed at fulfilling course requirements (completion of assigned activities), but did not improve learning outcomes. Descriptive norms increased behaviors aimed at improving knowledge (ungraded practice with activities after they were due), and improved performance. These results suggest that norms more effectively influence behavior when there is a match, or a sense of fit, between the goal of the behavior (fulfilling course requirements vs. learning) and the pull of a stated norm (social approval vs. efficacy). We discuss these implications with respect to students' motivations for self-regulated studying behavior in contemporary learning environments, and the overall goals of education.},
 bibtype = {article},
 author = {Eyink, Julie R. and Motz, Benjamin A. and Heltzel, Gordon and Liddell, Torrin M.},
 doi = {10.1111/jasp.12637},
 journal = {Journal of Applied Social Psychology},
 number = {1}
}

Teachers commonly use injunctive norms when telling students what they should be doing. But researchers find that sometimes descriptive norms, information about what others are actually doing, are more powerful influencers of behavior. In the present work, we examine which norm is more effective at increasing self-regulated studying and performance in an online college course across two semesters. To do this, we randomly assigned 751 undergraduate Introductory Psychology students to receive email messages at the start of every content unit that either contained descriptive norms, injunctive norms, information about the course, or a no message control. We found that injunctive norms increased study behaviors aimed at fulfilling course requirements (completion of assigned activities), but did not improve learning outcomes. Descriptive norms increased behaviors aimed at improving knowledge (ungraded practice with activities after they were due), and improved performance. These results suggest that norms more effectively influence behavior when there is a match, or a sense of fit, between the goal of the behavior (fulfilling course requirements vs. learning) and the pull of a stated norm (social approval vs. efficacy). We discuss these implications with respect to students' motivations for self-regulated studying behavior in contemporary learning environments, and the overall goals of education.

@inproceedings{
 title = {What college students say, and what they do: Aligning self-regulated learning theory with behavioral logs},
 type = {inproceedings},
 year = {2020},
 keywords = {LMS,Self-regulated learning,Self-reports,Trace data},
 pages = {534-543},
 websites = {https://dl.acm.org/doi/10.1145/3375462.3375516},
 month = {3},
 publisher = {Association for Computing Machinery},
 day = {23},
 city = {New York, NY, USA},
 id = {f05ffca7-8dd4-3d6d-aeac-fd49696e6958},
 created = {2020-04-22T21:44:57.119Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.770Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Quick2020},
 private_publication = {false},
 abstract = {A central concern in learning analytics specifically and educational research more generally is the alignment of robust, coherent measures to well-developed conceptual and theoretical frameworks. Capturing and representing processes of learning remains an ongoing challenge in all areas of educational inquiry and presents substantive considerations on the nature of learning, knowledge, and assessment & measurement that have been continuously refined in various areas of education and pedagogical practice. Learning analytics as a still developing method of inquiry has yet to substantively navigate the alignment of measurement, capture, and representation of learning to theoretical frameworks despite being used to identify various practical concerns such as at risk students. This study seeks to address these concerns by comparing behavioral measurements from learning management systems to established measurements of components of learning as understood through self-regulated learning frameworks. Using several prominent and robustly supported self-reported survey measures designed to identify dimensions of self-regulated learning, as well as typical behavioral features extracted from a learning management system, we conducted descriptive and exploratory analyses on the relational structures of these data. With the exception of learners' selfreported time management strategies and level of motivation, the current results indicate that behavioral measures were not well correlated with survey measurements. Possibilities and recommendations for learning analytics as measurements for selfregulated learning are discussed. © 2020 Association for Computing Machinery.},
 bibtype = {inproceedings},
 author = {Quick, Joshua and Motz, Benjamin and Israel, Jamie and Kaetzel, Jason},
 doi = {10.1145/3375462.3375516},
 booktitle = {International Conference on Learning Analytics & Knowledge}
}

A central concern in learning analytics specifically and educational research more generally is the alignment of robust, coherent measures to well-developed conceptual and theoretical frameworks. Capturing and representing processes of learning remains an ongoing challenge in all areas of educational inquiry and presents substantive considerations on the nature of learning, knowledge, and assessment & measurement that have been continuously refined in various areas of education and pedagogical practice. Learning analytics as a still developing method of inquiry has yet to substantively navigate the alignment of measurement, capture, and representation of learning to theoretical frameworks despite being used to identify various practical concerns such as at risk students. This study seeks to address these concerns by comparing behavioral measurements from learning management systems to established measurements of components of learning as understood through self-regulated learning frameworks. Using several prominent and robustly supported self-reported survey measures designed to identify dimensions of self-regulated learning, as well as typical behavioral features extracted from a learning management system, we conducted descriptive and exploratory analyses on the relational structures of these data. With the exception of learners' selfreported time management strategies and level of motivation, the current results indicate that behavioral measures were not well correlated with survey measurements. Possibilities and recommendations for learning analytics as measurements for selfregulated learning are discussed. © 2020 Association for Computing Machinery.

@inproceedings{
 title = {Scalable Quality Assurance for Neuroimaging (SQAN): automated quality control for medical imaging},
 type = {inproceedings},
 year = {2020},
 keywords = {angularjs,automated quality control,javascript portal,medical research imaging,mongodb,node.js,protocol compliance,vue.js},
 pages = {6},
 volume = {11318},
 websites = {https://www.spiedigitallibrary.org/conference-proceedings-of-spie/11318/2549722/Scalable-Quality-Assurance-for-Neuroimaging-SQAN--automated-quality-control/10.1117/12.2549722.full},
 month = {3},
 publisher = {SPIE},
 day = {2},
 id = {6097f12f-b885-33af-b080-a50a2d0c5bff},
 created = {2020-04-23T05:31:54.087Z},
 accessed = {2020-04-23},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2021-04-23T19:54:42.250Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Gopu2020},
 private_publication = {false},
 abstract = {Medical imaging, a key component in clinical diagnosis of and research on numerous medical conditions, is very costly and can generate massive datasets. For instance, a single scanned subject produces hundreds of thousands of images and millions of key-value metadata pairs that must be verified to ensure instrument and research protocol compliance. Many projects lack funds to reacquire images if data quality issues are detected later. Data quality assurance (QA) requires continuous involvement by all stakeholders and use of specific quality control (QC) methods to identify data issues likely to require post-processing correction or real-time re-acquisition. While many useful QC methods exist, they are often designed for specific use-cases with limited scope and documentation, making integration with other setups difficult. We present the Scalable Quality Assurance for Neuroimaging (SQAN), an open-source software suite developed by Indiana University for protocol quality control and instrumental validation on medical imaging data. SQAN includes a comprehensive QC Engine that ensures adherence to a research study’s protocol. A modern, intuitive web portal serves a wide range of users including researchers, scanner technologists and data scientists, each of whom approach QC with unique priorities, expertise, insights and expectations. Since Fall 2017, a fully operational SQAN instance has supported 50+ research projects, and has QC’d ∼3.5 million images and over 700 million metadata tags. SQAN is designed to scale to any imaging center’s QC needs, and to extend beyond protocol QC toward image-level QC and integration with pipeline and non-imaging database  systems.},
 bibtype = {inproceedings},
 author = {Gopu, Arvind and Young, Michael D. and Avena-Koenigsberger, Andrea and Perigo, Raymond W. and West, John D. and Paramasivam, Meenakshisundaram and Hayashi, Soichi and Henschel, Robert},
 editor = {Deserno, Thomas M. and Chen, Po-Hao},
 doi = {10.1117/12.2549722},
 booktitle = {Medical Imaging 2020: Imaging Informatics for Healthcare, Research, and Applications}
}

Medical imaging, a key component in clinical diagnosis of and research on numerous medical conditions, is very costly and can generate massive datasets. For instance, a single scanned subject produces hundreds of thousands of images and millions of key-value metadata pairs that must be verified to ensure instrument and research protocol compliance. Many projects lack funds to reacquire images if data quality issues are detected later. Data quality assurance (QA) requires continuous involvement by all stakeholders and use of specific quality control (QC) methods to identify data issues likely to require post-processing correction or real-time re-acquisition. While many useful QC methods exist, they are often designed for specific use-cases with limited scope and documentation, making integration with other setups difficult. We present the Scalable Quality Assurance for Neuroimaging (SQAN), an open-source software suite developed by Indiana University for protocol quality control and instrumental validation on medical imaging data. SQAN includes a comprehensive QC Engine that ensures adherence to a research study’s protocol. A modern, intuitive web portal serves a wide range of users including researchers, scanner technologists and data scientists, each of whom approach QC with unique priorities, expertise, insights and expectations. Since Fall 2017, a fully operational SQAN instance has supported 50+ research projects, and has QC’d ∼3.5 million images and over 700 million metadata tags. SQAN is designed to scale to any imaging center’s QC needs, and to extend beyond protocol QC toward image-level QC and integration with pipeline and non-imaging database systems.

@techreport{
 title = {Technical Report: XSEDE Return on Investment (Proxy) Data and Analysis Methods, July 2014 to August 2019},
 type = {techreport},
 year = {2020},
 websites = {https://scholarworks.iu.edu/dspace/handle/2022/25704},
 id = {ff6d89e2-4aca-3653-aaa6-b471e8872592},
 created = {2020-09-09T20:14:39.754Z},
 accessed = {2020-09-09},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.120Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wernert2020},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Wernert, Julie A.; and Costa, Claudia M.; McMullen, Donald F.; Stewart, Craig A.; and Blood, Phillip D.; and Sinkovits, Robert; Mehringer, Susan H.; Knepper, Richard}
}

@techreport{
 title = {2020 External Review of the Pervasive Technology Institute},
 type = {techreport},
 year = {2020},
 keywords = {External Review,Technical Report},
 websites = {http://creativecommons.org/licenses/by/4.0/.},
 month = {5},
 publisher = {PTI},
 day = {31},
 id = {4d88f50d-8db2-3cc4-b4b6-2296e351e228},
 created = {2020-09-09T20:34:09.114Z},
 accessed = {2020-09-09},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.365Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2020},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {Review process description, findings, comments, and supporting documents.},
 bibtype = {techreport},
 author = {Stewart, C A and Slavin, S and Ping, R}
}

Review process description, findings, comments, and supporting documents.

@techreport{
 title = {Indiana University Pervasive Technology Institute preproposal and proposal management, documentation, and templates},
 type = {techreport},
 year = {2020},
 keywords = {Technical Report},
 websites = {https://scholarworks.iu.edu/dspace/handle/2022/25580},
 id = {5f3f1ff9-6ead-32e0-8c1d-80347fee8add},
 created = {2020-09-09T20:39:18.028Z},
 accessed = {2020-09-09},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.265Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2020a},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {Technical report detailing IUPTI processes, with templates and styles.},
 bibtype = {techreport},
 author = {Stewart, Craig A. and Miller, Therese and Snapp-Childs, Winona and Jankowski, Harmony and Husk, Malinda}
}

Technical report detailing IUPTI processes, with templates and styles.

@inproceedings{
 title = {Experiences from scaling scale Science Gateway operations},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3333159},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {68c39bd2-0e8f-331a-a1c0-e16c7a74f587},
 created = {2019-10-01T17:20:12.176Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.625Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Marru2019},
 private_publication = {false},
 abstract = {Science gateways are distributed computing systems that provide science-centric, end-user environments that simplify and expand the use of scientific software and data on diverse scientific software on backend resources. In this poster we describe the experiences of using a common software platform to host "Software as a Service" Science Gateways.},
 bibtype = {inproceedings},
 author = {Marru, Suresh and Piece, Marlon and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Christie, Marcus and Wannipurage, Dimuthu},
 doi = {10.1145/3332186.3333159},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

Science gateways are distributed computing systems that provide science-centric, end-user environments that simplify and expand the use of scientific software and data on diverse scientific software on backend resources. In this poster we describe the experiences of using a common software platform to host "Software as a Service" Science Gateways.

@inproceedings{
 title = {Towards a science gateway reference architecture},
 type = {inproceedings},
 year = {2019},
 keywords = {Cyberinfrastructure,Science gateways},
 volume = {2357},
 publisher = {CEUR-WS},
 id = {a6f38fe8-3671-35cf-8bd8-c2c7db8f38b2},
 created = {2019-10-01T17:20:14.617Z},
 accessed = {2019-08-19},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.919Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pierce2019},
 private_publication = {false},
 abstract = {Science gateways have been developed over the last twenty years and have grown into a large community of practice, as evidenced by international workshops and conferences. Because of the diversity of approaches to creating science gateways and the always changing landscape of technologies, the community lacks a common definition for the term " science gateway " itself and common terminology for describing the common components of a gateway architecture. Instead, a wide range of definitions and understandings exist and are used in different communities; this is evident, for example, in discussions whether science gateways are the same as virtual research environments. This paper attempts to address these issues by focusing on how science gateways support scientific research and considering the consequences on cyberinfrastructure.},
 bibtype = {inproceedings},
 author = {Pierce, Marlon E. and Miller, Mark A. and Brookes, Emre H. and Wong, Mona and Afgan, Enis and Liu, Yan and Gesing, Sandra and Dahan, Maytal and Walker, Tony and Marru, Suresh},
 booktitle = {CEUR Workshop Proceedings}
}

Science gateways have been developed over the last twenty years and have grown into a large community of practice, as evidenced by international workshops and conferences. Because of the diversity of approaches to creating science gateways and the always changing landscape of technologies, the community lacks a common definition for the term " science gateway " itself and common terminology for describing the common components of a gateway architecture. Instead, a wide range of definitions and understandings exist and are used in different communities; this is evident, for example, in discussions whether science gateways are the same as virtual research environments. This paper attempts to address these issues by focusing on how science gateways support scientific research and considering the consequences on cyberinfrastructure.

@article{
 title = {Programmable Education Infrastructure: Cloud resources as HPC Education Environments},
 type = {article},
 year = {2019},
 pages = {107-107},
 volume = {10},
 websites = {http://www.jocse.org/articles/10/1/18/},
 month = {1},
 publisher = {The Shodor Education Foundation, Inc.},
 id = {34cc2dec-8fcd-3109-ada5-6e6bca28f5dd},
 created = {2019-10-01T17:20:18.496Z},
 accessed = {2019-09-24},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.386Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Coulter2019b},
 private_publication = {false},
 abstract = {Cloud computing is growing area for educating students and per- forming meaningful scientific research. The challenge for many educators and researchers is knowing how to use some of the unique aspects of computing in the cloud. One key feature is true elastic computing - resources on demand. The elasticity and programmability of cloud resources make them an excellent tool for educators who require access to a wide range of computing environments. In the field of HPC education, such environments are an absolute necessity, and getting access to them can create a large burden on the educators above and beyond designing content. While cloud resources won't replace traditional HPC environments for large research projects, they are an excellent option for providing both user and administrator education on HPC environments. The highly configurable nature of cloud environments allows educators to tailor the educational resource to the needs of their attendees, and provide a wide range of hands-on experiences. In this demo, we'll show how the Jetstream cloud environment can be used to provide training for both new HPC administrators and users, by showing a ground-up build of a simple HPC system. While this approach uses the Jetstream cloud, it is generalizable across any cloud provider. We will show how this allows an educator to tackle everything from basic command-line concepts and scheduler use to advanced cluster-management concepts such as elasticity and management of scientific software.},
 bibtype = {article},
 author = {Coulter, Eric and Knepper, Richard and Fischer, Jeremy},
 doi = {10.22369/issn.2153-4136/10/1/18},
 journal = {The Journal of Computational Science Education},
 number = {1}
}

Cloud computing is growing area for educating students and per- forming meaningful scientific research. The challenge for many educators and researchers is knowing how to use some of the unique aspects of computing in the cloud. One key feature is true elastic computing - resources on demand. The elasticity and programmability of cloud resources make them an excellent tool for educators who require access to a wide range of computing environments. In the field of HPC education, such environments are an absolute necessity, and getting access to them can create a large burden on the educators above and beyond designing content. While cloud resources won't replace traditional HPC environments for large research projects, they are an excellent option for providing both user and administrator education on HPC environments. The highly configurable nature of cloud environments allows educators to tailor the educational resource to the needs of their attendees, and provide a wide range of hands-on experiences. In this demo, we'll show how the Jetstream cloud environment can be used to provide training for both new HPC administrators and users, by showing a ground-up build of a simple HPC system. While this approach uses the Jetstream cloud, it is generalizable across any cloud provider. We will show how this allows an educator to tackle everything from basic command-line concepts and scheduler use to advanced cluster-management concepts such as elasticity and management of scientific software.

@inproceedings{
 title = {How the Science Gateways Community Institute Supports Those Who Are Creating Websites to Access Shared Resources},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3333256},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {04ccd75d-dcf7-3d5a-8c28-74e9f362c813},
 created = {2019-10-01T17:20:19.903Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.012Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lawrence2019},
 private_publication = {false},
 abstract = {The Science Gateways Community Institute (SGCI) serves the gateway community with free, NSF-funded resources, services, experts, and ideas. Our community shares expertise, technologies, and best practices for creating and sustaining gateways. Gateways offer Web-based access to shared resources, including data, software, computing services, instruments, and educational materials. Gateways are also known by other names, including portals, virtual research environments or virtual labs, eScience, and research cyberinfrastructure.},
 bibtype = {inproceedings},
 author = {Lawrence, Katherine A. and Mullinix, Nayiri and Dahan, Maytal and Hayden, Linda and Pierce, Marlon and Wilkins-Diehr, Nancy and Zentner, Michael},
 doi = {10.1145/3332186.3333256},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

The Science Gateways Community Institute (SGCI) serves the gateway community with free, NSF-funded resources, services, experts, and ideas. Our community shares expertise, technologies, and best practices for creating and sustaining gateways. Gateways offer Web-based access to shared resources, including data, software, computing services, instruments, and educational materials. Gateways are also known by other names, including portals, virtual research environments or virtual labs, eScience, and research cyberinfrastructure.

@inproceedings{
 title = {The Distant Reader},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3333260},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {1d3ad6d9-9f9a-31f3-ba1e-0169bab01e7f},
 created = {2019-10-01T17:20:20.775Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.043Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Morgan2019},
 private_publication = {false},
 abstract = {The Distant Reader science gateway can be used to automatically create and analyze text corpora at a scale of thousands of user-supplied documents. These processing steps are deployed on a dynamic virtual cluster deployed on XSEDE's Jetstream academic cloud computing resource and are accessed through a Web interface. The science gateway uses Apache Airavata middleware to manage the interactions between the Web interface and the virtual clusters. The gateway leverages the Science Gateway Platform as a service (SciGaP) infrastructure at Indiana University, which provides user authentication, authorization, and identity management as well as access to the Distant Reader tools. The Distant Reader is designed to assist in the process of using & understanding corpora -- reading.},
 bibtype = {inproceedings},
 author = {Morgan, Eric Lease and Abeysinghe, Eroma and Pamidighantam, Sudhkar and Coulter, Eric and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/3332186.3333260},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

The Distant Reader science gateway can be used to automatically create and analyze text corpora at a scale of thousands of user-supplied documents. These processing steps are deployed on a dynamic virtual cluster deployed on XSEDE's Jetstream academic cloud computing resource and are accessed through a Web interface. The science gateway uses Apache Airavata middleware to manage the interactions between the Web interface and the virtual clusters. The gateway leverages the Science Gateway Platform as a service (SciGaP) infrastructure at Indiana University, which provides user authentication, authorization, and identity management as well as access to the Distant Reader tools. The Distant Reader is designed to assist in the process of using & understanding corpora -- reading.

@inproceedings{
 title = {Research Computing at a Business University},
 type = {inproceedings},
 year = {2019},
 pages = {1-5},
 publisher = {Association for Computing Machinery (ACM)},
 id = {2a034811-9aed-3559-a390-d384796f30d3},
 created = {2019-10-01T17:20:21.782Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.613Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wells2019},
 private_publication = {false},
 abstract = {Research Computing demands are exploding beyond traditional disciplines due to the proliferation of data in all walks of life. At Bentley University ("Bentley"), a business university in the Boston area, this expansion has been most readily seen in our Accounting, Economics, Mathematics, and Natural Sciences departments. The result has been a small effort to build a research computing capability at this small New England university. This poster will serve as an overview of the steps taken to build such an effort at a business university, the revelations we have had along the way, and our plans for the future.},
 bibtype = {inproceedings},
 author = {Wells, Jason and Coulter, J. Eric},
 doi = {10.1145/3332186.3333161},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

Research Computing demands are exploding beyond traditional disciplines due to the proliferation of data in all walks of life. At Bentley University ("Bentley"), a business university in the Boston area, this expansion has been most readily seen in our Accounting, Economics, Mathematics, and Natural Sciences departments. The result has been a small effort to build a research computing capability at this small New England university. This poster will serve as an overview of the steps taken to build such an effort at a business university, the revelations we have had along the way, and our plans for the future.

@inproceedings{
 title = {LSU Computational System Biology Gateway for Education},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3333259},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {b0ad8399-23fc-336e-8a7b-482c35a3ce1e},
 created = {2019-10-01T17:20:21.832Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.772Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Abeysinghe2019},
 private_publication = {false},
 abstract = {Science gateways are a mechanism for delivering scientific software as a service, especially when the software requires high performance computing (HPC) resources to run effectively. The existence of a science gateway eliminates the user's need to learn to work with HPC systems and to manage software installations and updates. With well-designed user interfaces, users can more quickly become effective users of scientific applications and can manage information needed for replicating, modifying, and sharing results. All of these efficiency gains enable users to focus more on their research. In addition, science gateways are being identified as an effective educational tool, a tool to be used in classroom environments as a method to get students quickly into research on domain specific questions. In the absence of a science gateway, students are likely to need a considerable time to learn to work with HPC systems, and any time spent on such will reduce their time on the actual science. This poster presents how the Louisiana State University (LSU) gateway for the Computational System Biology Group (CSBG) - (www.brylinski.org) was updated and improved to be a classroom teaching tool. This work makes extensive use of Apache Airavata's group management capabilities.},
 bibtype = {inproceedings},
 author = {Abeysinghe, Eroma and Brylinski, Michal and Christie, Marcus and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/3332186.3333259},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

Science gateways are a mechanism for delivering scientific software as a service, especially when the software requires high performance computing (HPC) resources to run effectively. The existence of a science gateway eliminates the user's need to learn to work with HPC systems and to manage software installations and updates. With well-designed user interfaces, users can more quickly become effective users of scientific applications and can manage information needed for replicating, modifying, and sharing results. All of these efficiency gains enable users to focus more on their research. In addition, science gateways are being identified as an effective educational tool, a tool to be used in classroom environments as a method to get students quickly into research on domain specific questions. In the absence of a science gateway, students are likely to need a considerable time to learn to work with HPC systems, and any time spent on such will reduce their time on the actual science. This poster presents how the Louisiana State University (LSU) gateway for the Computational System Biology Group (CSBG) - (www.brylinski.org) was updated and improved to be a classroom teaching tool. This work makes extensive use of Apache Airavata's group management capabilities.

@inproceedings{
 title = {Beyond Campus Bridging: A retrospective of Cyberinfrastructure Integration Efforts},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 publisher = {Association for Computing Machinery (ACM)},
 id = {ee621697-4739-316a-974f-6f86c1e752fb},
 created = {2019-10-01T17:20:25.087Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T23:36:28.632Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Coulter2019a},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Coulter, J. Eric and Knepper, Rich and Reynolds, Resa and Sprouse, Jodie and Bird, Stephen},
 doi = {10.1145/3332186.3333151},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

@inproceedings{
 title = {A Lightweight Framework for Research Data Management},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 publisher = {Association for Computing Machinery (ACM)},
 id = {5fc03501-5af9-326b-8c71-d2f8a757554f},
 created = {2019-10-01T17:20:27.382Z},
 accessed = {2019-08-27},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.431Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Nikolov2019},
 private_publication = {false},
 abstract = {We describe a framework for managing live research data involving two major components. First, a system for the scalable scheduling and execution of automated policies for moving, organizing, and archiving data. Second, a system for managing metadata to facilitate curation and discovery with minimal change to existing workflows. Our approach is guided by four main principles: 1) to be non-invasive and to allow for easy integration into existing workflows and computing environments; 2) to be built on established, cloud-aware, open-source tools; 3) to be easily extensible and configurable, and thus, adaptable to different academic disciplines; and 4) to integrate with and take advantage of infrastructure and services available on academic campuses and research computing environments. These principles give our solution a well-defined place along the spectrum of research data management software such as sophisticated electronic lab notebooks and science gate-ways. Our lightweight and flexible data management framework provides for curation and preservation of research data within a lab, department or university cyberinfrastructure.},
 bibtype = {inproceedings},
 author = {Nikolov, Dimitar and Tuna, Esen},
 doi = {10.1145/3332186.3333157},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) (PEARC '19)}
}

We describe a framework for managing live research data involving two major components. First, a system for the scalable scheduling and execution of automated policies for moving, organizing, and archiving data. Second, a system for managing metadata to facilitate curation and discovery with minimal change to existing workflows. Our approach is guided by four main principles: 1) to be non-invasive and to allow for easy integration into existing workflows and computing environments; 2) to be built on established, cloud-aware, open-source tools; 3) to be easily extensible and configurable, and thus, adaptable to different academic disciplines; and 4) to integrate with and take advantage of infrastructure and services available on academic campuses and research computing environments. These principles give our solution a well-defined place along the spectrum of research data management software such as sophisticated electronic lab notebooks and science gate-ways. Our lightweight and flexible data management framework provides for curation and preservation of research data within a lab, department or university cyberinfrastructure.

@inproceedings{
 title = {E-RPID PEARC 2019 - The Digital Object Architecture and Enhanced Robust Persistent Identification of Data},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3333255},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {9f784acb-28da-3071-861d-e21ab88f29df},
 created = {2019-10-01T17:20:29.475Z},
 accessed = {2019-09-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.172Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Quick2019},
 private_publication = {false},
 abstract = {The expansion of the research community's ability to collect and store data has grown much more rapidly than its ability to catalog, make accessible, and make use of data. Recent initiatives in Open Science and Open Data have attempted to address the problems of making data discoverable, accessible and reusable at internet scales. The Enhanced Robust Persistent Identification of Data (E-RPID) project's goal is to address these deficiencies and enable options for data interoperability and reusability in the current research data landscape by utilizing Persistent Identifiers (PIDs) and a kernel of state information available with PID resolution. To do this requires integrating a set of preexisting software systems along with a small set of newly developed software solutions. The combination of these software components and the core principles of making data FAIR (findable, accessible, interoperable and reusable) will allow us to use Persistent Identifiers to create an end-to-end fabric capable of realizing the Digital Object Architecture for researchers. This poster will acquaint the audience to the concepts of the Digital Object Architecture, describe the software service architecture necessary to enable this architecture, outline the existing E-RPID testbed that is available for experimental usage from the Jetstream cloud environment, and describe the diverse set of use cases already using E-RPID to enhance their data accessibility, interoperability and reusability. It will focus on how the Digital Object Architecture and E-RPID testbed would interact with XSEDE resources and how E-RPID could assist with interoperability, reusability and reproducibility of HPC workflows.},
 bibtype = {inproceedings},
 author = {Quick, Rob and Lannom, Larry and Krenz, Marina and Luo, Yu},
 doi = {10.1145/3332186.3333255},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

The expansion of the research community's ability to collect and store data has grown much more rapidly than its ability to catalog, make accessible, and make use of data. Recent initiatives in Open Science and Open Data have attempted to address the problems of making data discoverable, accessible and reusable at internet scales. The Enhanced Robust Persistent Identification of Data (E-RPID) project's goal is to address these deficiencies and enable options for data interoperability and reusability in the current research data landscape by utilizing Persistent Identifiers (PIDs) and a kernel of state information available with PID resolution. To do this requires integrating a set of preexisting software systems along with a small set of newly developed software solutions. The combination of these software components and the core principles of making data FAIR (findable, accessible, interoperable and reusable) will allow us to use Persistent Identifiers to create an end-to-end fabric capable of realizing the Digital Object Architecture for researchers. This poster will acquaint the audience to the concepts of the Digital Object Architecture, describe the software service architecture necessary to enable this architecture, outline the existing E-RPID testbed that is available for experimental usage from the Jetstream cloud environment, and describe the diverse set of use cases already using E-RPID to enhance their data accessibility, interoperability and reusability. It will focus on how the Digital Object Architecture and E-RPID testbed would interact with XSEDE resources and how E-RPID could assist with interoperability, reusability and reproducibility of HPC workflows.

@article{
 title = {Population Genetics of Paramecium Mitochondrial Genomes: Recombination, Mutation Spectrum, and Efficacy of Selection.},
 type = {article},
 year = {2019},
 keywords = {Paramecium,efficacy of purifying selection,mitochondria,mutation spectrum,recombination,telomeres},
 pages = {1398-1416},
 volume = {11},
 websites = {http://www.ncbi.nlm.nih.gov/pubmed/30980669,http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC6505448},
 day = {1},
 id = {6e3c462a-024d-3b54-87e7-53abc4bf605b},
 created = {2019-10-01T17:20:30.760Z},
 accessed = {2019-08-09},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.780Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Johri2019},
 private_publication = {false},
 abstract = {The evolution of mitochondrial genomes and their population-genetic environment among unicellular eukaryotes are understudied. Ciliate mitochondrial genomes exhibit a unique combination of characteristics, including a linear organization and the presence of multiple genes with no known function or detectable homologs in other eukaryotes. Here we study the variation of ciliate mitochondrial genomes both within and across 13 highly diverged Paramecium species, including multiple species from the P. aurelia species complex, with four outgroup species: P. caudatum, P. multimicronucleatum, and two strains that may represent novel related species. We observe extraordinary conservation of gene order and protein-coding content in Paramecium mitochondria across species. In contrast, significant differences are observed in tRNA content and copy number, which is highly conserved in species belonging to the P. aurelia complex but variable among and even within the other Paramecium species. There is an increase in GC content from ∼20% to ∼40% on the branch leading to the P. aurelia complex. Patterns of polymorphism in population-genomic data and mutation-accumulation experiments suggest that the increase in GC content is primarily due to changes in the mutation spectra in the P. aurelia species. Finally, we find no evidence of recombination in Paramecium mitochondria and find that the mitochondrial genome appears to experience either similar or stronger efficacy of purifying selection than the nucleus.},
 bibtype = {article},
 author = {Johri, Parul and Marinov, Georgi K and Doak, Thomas G and Lynch, Michael},
 doi = {10.1093/gbe/evz081},
 journal = {Genome Biology and Evolution},
 number = {5}
}

The evolution of mitochondrial genomes and their population-genetic environment among unicellular eukaryotes are understudied. Ciliate mitochondrial genomes exhibit a unique combination of characteristics, including a linear organization and the presence of multiple genes with no known function or detectable homologs in other eukaryotes. Here we study the variation of ciliate mitochondrial genomes both within and across 13 highly diverged Paramecium species, including multiple species from the P. aurelia species complex, with four outgroup species: P. caudatum, P. multimicronucleatum, and two strains that may represent novel related species. We observe extraordinary conservation of gene order and protein-coding content in Paramecium mitochondria across species. In contrast, significant differences are observed in tRNA content and copy number, which is highly conserved in species belonging to the P. aurelia complex but variable among and even within the other Paramecium species. There is an increase in GC content from ∼20% to ∼40% on the branch leading to the P. aurelia complex. Patterns of polymorphism in population-genomic data and mutation-accumulation experiments suggest that the increase in GC content is primarily due to changes in the mutation spectra in the P. aurelia species. Finally, we find no evidence of recombination in Paramecium mitochondria and find that the mitochondrial genome appears to experience either similar or stronger efficacy of purifying selection than the nucleus.

@techreport{
 title = {The Pervasive Technology Institute at 20: Two decades of success and counting},
 type = {techreport},
 year = {2019},
 websites = {http://hdl.handle.net/2022/22607},
 city = {Bloomington, IN},
 institution = {Pervasive Technology Institute},
 id = {f561e059-e5bc-3bcc-a741-ff457eb842ad},
 created = {2019-10-01T17:20:36.362Z},
 accessed = {2019-08-20},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.456Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2019d},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A.; and Welch, Von; and Doak, Thomas G.; and Miller, Therese; and Plale, Beth; and Walsh, John A.; and Link, Matthew R.; and Snapp-Childs, Winona},
 doi = {10.5967/QDF0-S837}
}

@inbook{
 type = {inbook},
 year = {2019},
 pages = {189-222},
 websites = {https://www.taylorfrancis.com/books/9781351036856/chapters/10.1201/9781351036863-8},
 month = {5},
 publisher = {CRC Press},
 day = {8},
 id = {a080099c-bf5b-3af6-88ef-9f49972b6644},
 created = {2019-10-01T17:20:36.811Z},
 accessed = {2019-08-14},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:59:22.878Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2019},
 private_publication = {false},
 abstract = {This chapter describes the implementation of Jetstream focusing on: architectures; implementation approaches; lessons learned; uses and applications that constitute interesting departures from the traditional National Science Foundation (NSF)-funded high performance computing (HPC) systems and their users. The Jetstream Project Execution Plan (PEP) included functionality tests for supporting Science Gateways, developed by agreement between the NSF and the Jetstream team. The Jetstream PEP also included functionality tests for supporting data movement, storage, and dissemination. OpenStack is the “cloud operating system” that powers Jetstream. It is a collection of projects that provide services to orchestrate the delivery of computational, networking, and storage resources. The OpenStack Application Programming Interfaces within Jetstream offer users the ability to leverage Jetstream in conjunction with other resources, applications, or science gateways via programmatic methods that are not available with most other NSF systems.},
 bibtype = {inbook},
 author = {Stewart, Craig A. and Hancock, David Y. and Miller, Therese and Fischer, Jeremy and Liming, R. Lee and Turner, George and Lowe, John Michael and Gregory, Steven and Skidmore, Edwin and Vaughn, Matthew and Stanzione, Dan and Merchant, Nirav and Foster, Ian and Taylor, James and Rad, Paul and Brendel, Volker and Afgan, Enis and Packard, Michael and Miller, Therese and Snapp-Childs, Winona},
 doi = {10.1201/9781351036863-8},
 chapter = {Jetstream: A Novel Cloud System for Science},
 title = {Contemporary High Performance Computing}
}

This chapter describes the implementation of Jetstream focusing on: architectures; implementation approaches; lessons learned; uses and applications that constitute interesting departures from the traditional National Science Foundation (NSF)-funded high performance computing (HPC) systems and their users. The Jetstream Project Execution Plan (PEP) included functionality tests for supporting Science Gateways, developed by agreement between the NSF and the Jetstream team. The Jetstream PEP also included functionality tests for supporting data movement, storage, and dissemination. OpenStack is the “cloud operating system” that powers Jetstream. It is a collection of projects that provide services to orchestrate the delivery of computational, networking, and storage resources. The OpenStack Application Programming Interfaces within Jetstream offer users the ability to leverage Jetstream in conjunction with other resources, applications, or science gateways via programmatic methods that are not available with most other NSF systems.

@article{
 title = {Training children aged 5–10 years in manual compliance control to improve drawing and handwriting},
 type = {article},
 year = {2019},
 pages = {42-50},
 volume = {65},
 websites = {https://www.sciencedirect.com/science/article/pii/S0167945717308552},
 month = {6},
 publisher = {North-Holland},
 day = {1},
 id = {99e328dd-d6be-31bc-8002-ef6ba1db379a},
 created = {2019-10-01T17:20:38.885Z},
 accessed = {2019-08-14},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:29.491Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Bingham2019},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {A large proportion of school-aged children exhibit poor drawing and handwriting. This prevalence limits the availability of therapy. We developed an automated method for training improved manual compliance control and relatedly, prospective control of a stylus. The approach included a difficult training task, while providing parametrically modifiable support that enables the children to perform successfully while developing good compliance control. The task was to use a stylus to push a bead along a 3D wire path. Support was provided by making the wire magnetically attractive to the stylus. Support was progressively reduced as 3D tracing performance improved. We report studies that (1) compared performance of Typically Developing (TD) children and children with Developmental Coordination Disorder (DCD), (2) tested training with active versus passive movement, (3) tested progressively reduced versus constant or no support during training, (4) tested children of different ages, (5) tested the transfer of training to a drawing task, (6) tested the specificity of training in respect to the size, shape and dimensionality of figures, and (7) investigated the relevance of the training task to the Beery VMI, an inventory used to diagnose DCD. The findings were as follows. (1) Pre-training performance of TD and DCD children was the same and good with high support but distinct and poor with low support. Support yielded good self-efficacy that motivated training. Post training performance with no support was improved and the same for TD and DCD children. (2) Actively controlled movements were required for improved performance. (3) Progressively reduced support was required for good performance during and after training. (4) Age differences in performance during pre-training were eliminated post-training. (5) Improvements transferred to drawing. (6) There was no evidence of specificity of training in transfer. (7) Disparate Beery scores were reflected in pre-training but not post-training performance. We conclude that the method improves manual compliance control, and more generally, prospective control of movements used in drawing performance.},
 bibtype = {article},
 author = {Bingham, Geoffrey P. and Snapp-Childs, Winona},
 doi = {10.1016/J.HUMOV.2018.04.002},
 journal = {Human Movement Science}
}

A large proportion of school-aged children exhibit poor drawing and handwriting. This prevalence limits the availability of therapy. We developed an automated method for training improved manual compliance control and relatedly, prospective control of a stylus. The approach included a difficult training task, while providing parametrically modifiable support that enables the children to perform successfully while developing good compliance control. The task was to use a stylus to push a bead along a 3D wire path. Support was provided by making the wire magnetically attractive to the stylus. Support was progressively reduced as 3D tracing performance improved. We report studies that (1) compared performance of Typically Developing (TD) children and children with Developmental Coordination Disorder (DCD), (2) tested training with active versus passive movement, (3) tested progressively reduced versus constant or no support during training, (4) tested children of different ages, (5) tested the transfer of training to a drawing task, (6) tested the specificity of training in respect to the size, shape and dimensionality of figures, and (7) investigated the relevance of the training task to the Beery VMI, an inventory used to diagnose DCD. The findings were as follows. (1) Pre-training performance of TD and DCD children was the same and good with high support but distinct and poor with low support. Support yielded good self-efficacy that motivated training. Post training performance with no support was improved and the same for TD and DCD children. (2) Actively controlled movements were required for improved performance. (3) Progressively reduced support was required for good performance during and after training. (4) Age differences in performance during pre-training were eliminated post-training. (5) Improvements transferred to drawing. (6) There was no evidence of specificity of training in transfer. (7) Disparate Beery scores were reflected in pre-training but not post-training performance. We conclude that the method improves manual compliance control, and more generally, prospective control of movements used in drawing performance.

@inproceedings{
 title = {Research Computing Desktops: Demystifying research computing for non-Linux users},
 type = {inproceedings},
 year = {2019},
 pages = {1-8},
 publisher = {Association for Computing Machinery (ACM)},
 id = {773a2970-8c29-3b82-8320-37e1dcac56a6},
 created = {2019-10-01T17:20:40.703Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.468Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Thota2019},
 private_publication = {false},
 abstract = {Many members of the current generation of students and researchers are accustomed to intuitive computing devices and never had to learn how to use command-line based systems, which comprise the majority of high-performance computing environments in use. In the 2013-14 time frame, both Indiana University and Purdue university separately launched virtual desktop front-ends for their high performance computing clusters with the aim of offering an easier on-ramp to new users. In the last five years we iterated on and refined these approaches, and we now have over two thousand annual active users combined. Over 75% of those users say that the desktop services are either moderately or extremely important for their ability to use HPC resources. In this paper, we share our experience bootstrapping this new service framework, bringing in the end-users, dealing with runaway success, and making this service a sustainable offering. This paper offers a comprehensive picture of the driving motivations for desktops at each institution, reasons users like desktops, and ways of getting started.},
 bibtype = {inproceedings},
 author = {Thota, Abhinav and Dietz, Daniel T. and Phillips, Christopher and Zhu, Xiao and Weakley, Le Mai and Fulton, Ben and Dennis, H. E. Cicada Brokaw and Huber, Laura and Michael, Scott and Snapp-Childs, Winona and Harrell, Stephen Lien and Younts, Alexander},
 doi = {10.1145/3332186.3332206},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) (PEARC '19)}
}

Many members of the current generation of students and researchers are accustomed to intuitive computing devices and never had to learn how to use command-line based systems, which comprise the majority of high-performance computing environments in use. In the 2013-14 time frame, both Indiana University and Purdue university separately launched virtual desktop front-ends for their high performance computing clusters with the aim of offering an easier on-ramp to new users. In the last five years we iterated on and refined these approaches, and we now have over two thousand annual active users combined. Over 75% of those users say that the desktop services are either moderately or extremely important for their ability to use HPC resources. In this paper, we share our experience bootstrapping this new service framework, bringing in the end-users, dealing with runaway success, and making this service a sustainable offering. This paper offers a comprehensive picture of the driving motivations for desktops at each institution, reasons users like desktops, and ways of getting started.

@article{
 title = {A High-Frequency Mobile Phone Data Collection Approach for Research in Social-Environmental Systems: Applications in Climate Variability and Food Security in Sub-Saharan Africa},
 type = {article},
 year = {2019},
 websites = {https://www.sciencedirect.com/science/article/pii/S1364815218303207?via%3Dihub},
 month = {5},
 publisher = {Elsevier},
 day = {20},
 id = {96f017df-dfbf-3174-a223-2d2a24414475},
 created = {2019-10-01T17:20:51.162Z},
 accessed = {2019-05-20},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.993Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Giroux2019},
 private_publication = {false},
 abstract = {Collecting high-frequency social-environmental data about farming practices in sub-Saharan Africa can provide new insight into environmental changes that farmers face and how they respond within smallholder agro-ecosystems. Traditional data collection methods such as agricultural censuses are costly and not useful for understanding intra-annual and real-time decisions. Short-message service (SMS) has the potential to transform the nature of data collection in coupled social-ecological systems. We present a system for collecting, managing, and synthesizing weekly data from farmers, including data infrastructure for management of big and heterogeneous datasets; probabilistic data quality assessment tools; and visualization and analysis tools such as mapping and regression techniques. We discuss limitations of collecting social-environmental data via SMS and data integration challenges that arise when linking these data with other social and environmental data. In combination with high-frequency environmental data, such data will help ameliorate issues of scale mismatch and build resilience in environmental systems.},
 bibtype = {article},
 author = {Giroux, Stacey A. and Kouper, Inna and Estes, Lyndon D. and Schumacher, Jacob and Waldman, Kurt and Greenshields, Joel T. and Dickinson, Stephanie L. and Caylor, Kelly K. and Evans, Tom P.},
 doi = {10.1016/J.ENVSOFT.2019.05.011},
 journal = {Environmental Modelling & Software}
}

Collecting high-frequency social-environmental data about farming practices in sub-Saharan Africa can provide new insight into environmental changes that farmers face and how they respond within smallholder agro-ecosystems. Traditional data collection methods such as agricultural censuses are costly and not useful for understanding intra-annual and real-time decisions. Short-message service (SMS) has the potential to transform the nature of data collection in coupled social-ecological systems. We present a system for collecting, managing, and synthesizing weekly data from farmers, including data infrastructure for management of big and heterogeneous datasets; probabilistic data quality assessment tools; and visualization and analysis tools such as mapping and regression techniques. We discuss limitations of collecting social-environmental data via SMS and data integration challenges that arise when linking these data with other social and environmental data. In combination with high-frequency environmental data, such data will help ameliorate issues of scale mismatch and build resilience in environmental systems.

@article{
 title = {Safe Open Science for Restricted Data},
 type = {article},
 year = {2019},
 pages = {50-60},
 volume = {3},
 websites = {https://content.sciendo.com/view/journals/dim/3/1/article-p50.xml},
 id = {1a52cd49-f051-32ed-9dce-2983b96f8bdf},
 created = {2019-10-01T17:20:51.328Z},
 accessed = {2019-06-16},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.950Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Plale2019},
 private_publication = {false},
 bibtype = {article},
 author = {Plale, Beth A and Dickson, Eleanor and Kouper, Inna and Liyanage, Samitha and Ma, Yu and McDonald, Robert H. and Walsh, John A and Withana, Sachith},
 doi = {10.2478/dim-2019-0005},
 journal = {Data and Information Management},
 number = {1}
}

@article{
 title = {Twister2: Design of a big data toolkit},
 type = {article},
 year = {2019},
 keywords = {big data,dataflow,event-driven computing,high performance computing},
 publisher = {John Wiley and Sons Ltd},
 id = {c887ff50-bb65-3ac4-b78e-0ed6f691138d},
 created = {2019-10-01T17:20:54.329Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.121Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Kamburugamuve2019},
 private_publication = {false},
 abstract = {Summary Data-driven applications are essential to handle the ever-increasing volume, velocity, and veracity of data generated by sources such as the Web and Internet of Things (IoT) devices. Simultaneously, an event-driven computational paradigm is emerging as the core of modern systems designed for database queries, data analytics, and on-demand applications. Modern big data processing runtimes and asynchronous many task (AMT) systems from high performance computing (HPC) community have adopted dataflow event-driven model. The services are increasingly moving to an event-driven model in the form of Function as a Service (FaaS) to compose services. An event-driven runtime designed for data processing consists of well-understood components such as communication, scheduling, and fault tolerance. Different design choices adopted by these components determine the type of applications a system can support efficiently. We find that modern systems are limited to specific sets of applications because they have been designed with fixed choices that cannot be changed easily. In this paper, we present a loosely coupled component-based design of a big data toolkit where each component can have different implementations to support various applications. Such a polymorphic design would allow services and data analytics to be integrated seamlessly and expand from edge to cloud to HPC environments.},
 bibtype = {article},
 author = {Kamburugamuve, Supun and Govindarajan, Kannan and Wickramasinghe, Pulasthi and Abeykoon, Vibhatha and Fox, Geoffrey},
 doi = {10.1002/cpe.5189},
 journal = {Concurrency Computation: Practice and Experience}
}

Summary Data-driven applications are essential to handle the ever-increasing volume, velocity, and veracity of data generated by sources such as the Web and Internet of Things (IoT) devices. Simultaneously, an event-driven computational paradigm is emerging as the core of modern systems designed for database queries, data analytics, and on-demand applications. Modern big data processing runtimes and asynchronous many task (AMT) systems from high performance computing (HPC) community have adopted dataflow event-driven model. The services are increasingly moving to an event-driven model in the form of Function as a Service (FaaS) to compose services. An event-driven runtime designed for data processing consists of well-understood components such as communication, scheduling, and fault tolerance. Different design choices adopted by these components determine the type of applications a system can support efficiently. We find that modern systems are limited to specific sets of applications because they have been designed with fixed choices that cannot be changed easily. In this paper, we present a loosely coupled component-based design of a big data toolkit where each component can have different implementations to support various applications. Such a polymorphic design would allow services and data analytics to be integrated seamlessly and expand from edge to cloud to HPC environments.

@article{
 title = {Regge phenomenology of the N∗ and Δ∗ poles},
 type = {article},
 year = {2019},
 volume = {99},
 month = {2},
 publisher = {American Physical Society},
 day = {1},
 id = {b9ee397f-29d2-3d33-96cb-b3400c1aceaa},
 created = {2019-10-01T17:20:54.517Z},
 accessed = {2019-09-03},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.234Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Silva-Castro2019},
 private_publication = {false},
 abstract = {We use Regge phenomenology to study the structure of the poles of the $N^*$ and $\mathrmΔ^*$ spectrum. We employ the available pole extractions from partial wave analysis of meson scattering and photoproduction data. We assess the importance of the imaginary part of the poles (widths) to o...},
 bibtype = {article},
 author = {Silva-Castro, J. A. and Fernández-Ramírez, C. and Albaladejo, M. and Danilkin, I. V. and Jackura, A. and Mathieu, V. and Nys, J. and Pilloni, A. and Szczepaniak, A. P. and Fox, G.},
 doi = {10.1103/PhysRevD.99.034003},
 journal = {Physical Review D},
 number = {3}
}

We use Regge phenomenology to study the structure of the poles of the $N^*$ and $\mathrmΔ^*$ spectrum. We employ the available pole extractions from partial wave analysis of meson scattering and photoproduction data. We assess the importance of the imaginary part of the poles (widths) to o...

@inproceedings{
 title = {Big Data Benchmarks of High-Performance Storage Systems on Commercial Bare Metal Clouds},
 type = {inproceedings},
 year = {2019},
 websites = {https://conferences.computer.org/serviceswp/2019/pdfs/CLOUD2019-5XaiJ82Ya5AyIoh2T1E5Bm/6ex2QOZUKAReRoHl2O8jiP/5Fkht00rYCKTzblzYTjaNc.pdf},
 month = {5},
 day = {27},
 id = {8d081792-2291-34a8-8183-c85bc3d92727},
 created = {2019-10-01T17:20:54.710Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.061Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lee2019},
 private_publication = {false},
 abstract = {Bare metal servers are widely available on public clouds to provide direct access to hardware and the system configuration with high performance storage and network devices are well suited for big data applications. Highly-optimized server with additional CPU core count and dense storage may lead to better performance in certain workloads and to ensure responsiveness of deployed services. Recent work on Hadoop ecosystems has addressed the performance improvement of scale-up machines configured with SSD storage and increased network bandwidth. The paper evaluates big data processing on dedicated clusters and provides the performance analysis of NVMe devices and SSD block storage options available on Amazon, Google, Microsoft, and Oracle Clouds. We show the benchmark results along with the system performance tests as we want to demonstrate the compute resource requirements for large-scale applications. The system capacity and limits for the underlying servers are described along with the cost analysis of scaling workloads on these platforms.},
 bibtype = {inproceedings},
 author = {Lee, Hyungro and Fox, Geoffrey},
 doi = {10.1109/CLOUD.2019.00014},
 booktitle = {IEEE International Conference on Cloud Computing (IEEE CLOUD 2019)}
}

Bare metal servers are widely available on public clouds to provide direct access to hardware and the system configuration with high performance storage and network devices are well suited for big data applications. Highly-optimized server with additional CPU core count and dense storage may lead to better performance in certain workloads and to ensure responsiveness of deployed services. Recent work on Hadoop ecosystems has addressed the performance improvement of scale-up machines configured with SSD storage and increased network bandwidth. The paper evaluates big data processing on dedicated clusters and provides the performance analysis of NVMe devices and SSD block storage options available on Amazon, Google, Microsoft, and Oracle Clouds. We show the benchmark results along with the system performance tests as we want to demonstrate the compute resource requirements for large-scale applications. The system capacity and limits for the underlying servers are described along with the cost analysis of scaling workloads on these platforms.

@techreport{
 title = {Learning Everywhere Resource for BDEC},
 type = {techreport},
 year = {2019},
 id = {90584fbe-d7f1-3ba4-af48-d1023715969b},
 created = {2019-10-01T17:20:54.968Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.102Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Fox2019},
 private_publication = {false},
 bibtype = {techreport},
 author = {Fox, Geoffrey and July, Updated}
}

@techreport{
 title = {Harnessing the Computing Continuum for Programming Our World},
 type = {techreport},
 year = {2019},
 pages = {1-11},
 websites = {https://www.researchgate.net/publication/332246123_Harnessing_the_Computing_Continuum_for_Programming_Our_World},
 id = {a159efb9-0f64-32a8-b05f-c7a0608b77a5},
 created = {2019-10-01T17:20:55.482Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.324Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Beckman2019},
 private_publication = {false},
 abstract = {This paper outlines a vision for how best to harness the continuum of interconnected sensors, actuators, instruments, and computing systems, from small numbers of very large devices to large numbers of very small devices. Our hypothesis is that only via a continuum perspective can we intentionally specify desired continuum actions and effectively manage outcomes and systemic properties—adaptability and homeostasis, temporal constraints and deadlines—and elevate the discourse from device programming to intellectual goals and outcomes.},
 bibtype = {techreport},
 author = {Beckman, Pete and Dongarra, Jack and Ferrier, Nicola and Fox, Geoffrey and Moore, Terry and Reed, Dan and Beck, Micah}
}

This paper outlines a vision for how best to harness the continuum of interconnected sensors, actuators, instruments, and computing systems, from small numbers of very large devices to large numbers of very small devices. Our hypothesis is that only via a continuum perspective can we intentionally specify desired continuum actions and effectively manage outcomes and systemic properties—adaptability and homeostasis, temporal constraints and deadlines—and elevate the discourse from device programming to intellectual goals and outcomes.

@inproceedings{
 title = {Perspectives on High-Performance Computing in a Big Data World},
 type = {inproceedings},
 year = {2019},
 pages = {145-145},
 publisher = {Association for Computing Machinery (ACM)},
 id = {b3fd52e4-5fee-3c48-870c-5907f55cdc3c},
 created = {2019-10-01T17:20:55.838Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.542Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2019b},
 private_publication = {false},
 abstract = {High-Performance Computing (HPC) and Cyberinfrastructure have played a leadership role in computational science even since the start of the NSF computing centers program. Thirty years ago parallel computing was a centerpiece of computer science research. Naively Big Data surely requires HPC to be processed, and transformational Big Data technology such as Hadoop and Spark exploit parallelism to success. Nevertheless, the HPC community does not appear to be thriving as a leader in Data Science while parallel computing is no longer a centerpiece. Some reasons for this are the dominant presence of Industry in technology futures and the universal fascination with Artificial Intelligence and Machine Learning. Maybe the pendulum will swing back a bit, but I expect the "AI first" philosophy to dominate in the foreseeable future. Thus I describe a future where HPC thrives in collaboration with Industry and AI. In particular, I discuss the promise of MLforHPC (AI for systems) and HPCforML (systems for AI).},
 bibtype = {inproceedings},
 author = {Fox, Geoffrey C.},
 doi = {10.1145/3307681.3325410},
 booktitle = {Proceedings of the 28th International Symposium on High-Performance Parallel and Distributed Computing (HPDC '19)}
}

High-Performance Computing (HPC) and Cyberinfrastructure have played a leadership role in computational science even since the start of the NSF computing centers program. Thirty years ago parallel computing was a centerpiece of computer science research. Naively Big Data surely requires HPC to be processed, and transformational Big Data technology such as Hadoop and Spark exploit parallelism to success. Nevertheless, the HPC community does not appear to be thriving as a leader in Data Science while parallel computing is no longer a centerpiece. Some reasons for this are the dominant presence of Industry in technology futures and the universal fascination with Artificial Intelligence and Machine Learning. Maybe the pendulum will swing back a bit, but I expect the "AI first" philosophy to dominate in the foreseeable future. Thus I describe a future where HPC thrives in collaboration with Industry and AI. In particular, I discuss the promise of MLforHPC (AI for systems) and HPCforML (systems for AI).

@inproceedings{
 title = {Learning Everywhere: Pervasive Machine Learning for Effective High-Performance Computation: Application Background},
 type = {inproceedings},
 year = {2019},
 pages = {33},
 websites = {http://dsc.soic.indiana.edu/publications/Learning_Everywhere.pdf,https://ieeexplore.ieee.org/document/8778333/},
 month = {5},
 publisher = {IEEE},
 id = {ecf9f27c-fa08-3caf-a779-1af03baaec95},
 created = {2019-10-01T17:20:57.114Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.632Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2019a},
 private_publication = {false},
 abstract = {The convergence of HPC and data-intensive methodologies provide a promising approach to major performance improvements. This paper provides a general description of the interaction between traditional HPC and ML approaches and motivates the Learning Everywhere paradigm for HPC. We introduce the concept of effective performance that one can achieve by combining learning methodologies with simulation-based approaches, and distinguish between traditional performance as measured by benchmark scores. To support the promise of integrating HPC and learning methods, this paper examines specific examples and opportunities across a series of domains. It concludes with a series of open computer science and cyberinfrastructure questions and challenges that the Learning Everywhere paradigm presents.},
 bibtype = {inproceedings},
 author = {Fox, Geoffrey and Glazier, James A and Kadupitiya, Jcs and Jadhao, Vikram and Kim, Minje and Qiu, Judy and Sluka, James P. and Somogy, Endre and Marathe, Madhav and Adiga, Abhijin and Chen, Jiangzhuo and Beckstein, Oliver and Jha, Shantenu and Somogyi, Endre and Marathe, Madhav and Adiga, Abhijin and Chen, Jiangzhuo and Beckstein, Oliver and Jha, Shantenu},
 doi = {10.1109/IPDPSW.2019.00081},
 booktitle = {2019 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)}
}

The convergence of HPC and data-intensive methodologies provide a promising approach to major performance improvements. This paper provides a general description of the interaction between traditional HPC and ML approaches and motivates the Learning Everywhere paradigm for HPC. We introduce the concept of effective performance that one can achieve by combining learning methodologies with simulation-based approaches, and distinguish between traditional performance as measured by benchmark scores. To support the promise of integrating HPC and learning methods, this paper examines specific examples and opportunities across a series of domains. It concludes with a series of open computer science and cyberinfrastructure questions and challenges that the Learning Everywhere paradigm presents.

@inproceedings{
 title = {Machine Learning for Performance Enhancement of Molecular Dynamics Simulations},
 type = {inproceedings},
 year = {2019},
 keywords = {Clouds,Machine learning,Molecular dynamics simulations,Parallel computing,Scientific computing},
 pages = {116-130},
 volume = {11537 LNCS},
 publisher = {Springer Verlag},
 id = {9646082a-6701-3077-a07c-8d7e8f4a328a},
 created = {2019-10-01T17:20:57.404Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.518Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kadupitiya2019},
 private_publication = {false},
 abstract = {We explore the idea of integrating machine learning with simulations to enhance the performance of the simulation and improve its usability for research and education. The idea is illustrated using hybrid OpenMP/MPI parallelized molecular dynamics simulations designed to extract the distribution of ions in nanoconfinement. We find that an artificial neural network based regression model successfully learns the desired features associated with the output ionic density profiles and rapidly generates predictions that are in excellent agreement with the results from explicit molecular dynamics simulations. The results demonstrate that the performance gains of parallel computing can be further enhanced by using machine learning.},
 bibtype = {inproceedings},
 author = {Kadupitiya, Jcs and Fox, Geoffrey C. and Jadhao, Vikram},
 doi = {10.1007/978-3-030-22741-8_9}
}

We explore the idea of integrating machine learning with simulations to enhance the performance of the simulation and improve its usability for research and education. The idea is illustrated using hybrid OpenMP/MPI parallelized molecular dynamics simulations designed to extract the distribution of ions in nanoconfinement. We find that an artificial neural network based regression model successfully learns the desired features associated with the output ionic density profiles and rapidly generates predictions that are in excellent agreement with the results from explicit molecular dynamics simulations. The results demonstrate that the performance gains of parallel computing can be further enhanced by using machine learning.

@article{
 title = {Object Classification by a Super-resolution Method and a Convolutional Neural Networks},
 type = {article},
 year = {2019},
 keywords = {CNN,convolution neural networks,deep learning,machine learning,object detection,super-resolution},
 pages = {16-23},
 volume = {1},
 websites = {http://ijdat.org/index.php/ijdat/article/view/4/4},
 id = {551a8527-fb8c-37fd-8a6a-03eeba531acb},
 created = {2019-10-01T17:21:01.059Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.780Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Na2019},
 private_publication = {false},
 abstract = {Recently with many blurless or slightly blurred images, convolutional neural networks classify objects with around 90 percent classification rates, even if there are variable sized images. However, small object regions or cropping of images make object detection or classification difficult and decreases the detection rates. In many methods related to convolutional neural network (CNN), Bilinear or Bicubic algorithms are popularly used to interpolate region of interests. To overcome the limitations of these algorithms, we introduce a super-resolution method applied to the cropped regions or candidates, and this leads to improve recognition rates for object detection and classification. Large object candidates comparable in size of the full image have good results for object detections using many popular conventional methods. However, for smaller region candidates, using our super-resolution preprocessing and region candidates, allows a CNN to outperform conventional methods in the number of detected objects when tested on the VOC2007 and MSO datasets},
 bibtype = {article},
 author = {Na, Bokyoon and Fox, Geoffrey C},
 journal = {International Journal of Data Mining Science},
 number = {1}
}

Recently with many blurless or slightly blurred images, convolutional neural networks classify objects with around 90 percent classification rates, even if there are variable sized images. However, small object regions or cropping of images make object detection or classification difficult and decreases the detection rates. In many methods related to convolutional neural network (CNN), Bilinear or Bicubic algorithms are popularly used to interpolate region of interests. To overcome the limitations of these algorithms, we introduce a super-resolution method applied to the cropped regions or candidates, and this leads to improve recognition rates for object detection and classification. Large object candidates comparable in size of the full image have good results for object detections using many popular conventional methods. However, for smaller region candidates, using our super-resolution preprocessing and region candidates, allows a CNN to outperform conventional methods in the number of detected objects when tested on the VOC2007 and MSO datasets

@article{
 title = {Advances in big data programming, system software and HPC convergence},
 type = {article},
 year = {2019},
 pages = {489-493},
 volume = {75},
 month = {2},
 publisher = {Springer New York LLC},
 day = {6},
 id = {acca6409-98dd-30a1-895f-ee8066ce01c8},
 created = {2019-10-01T17:21:01.218Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.876Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hsu2019},
 private_publication = {false},
 abstract = {In the era of big data, developing modern computing systems and system software that can scale to massive amounts of data becomes a key challenge to both researchers and practitioners. Scalability in distributed system usually means that the performance of a system should increase proportionally with the increase in resources. However, this is not sufficient in the big data era. The system should be designed in a way so that all the five Vs of big data can be tackled. Driven by this insight, this special issue aims at presenting the current state-of-the-art research and future trends on various aspects of big data programming and system software techniques for big data processing and attempts towards building highly adaptive big data systems that can automatically adapt their behaviours to the amount of available resources. The major subjects cover methodologies, modelling, analysis and newly introduced applications. Besides the latest research achievements, this special issue also covers innovative commercial data management systems, innovative commercial applications of big data technology and experience in applying recent research advances to real-world problems.},
 bibtype = {article},
 author = {Hsu, Ching Hsien and Fox, Geoffrey and Min, Geyong and Sharma, Sugam},
 doi = {10.1007/s11227-018-2706-x},
 journal = {Journal of Supercomputing},
 number = {2}
}

In the era of big data, developing modern computing systems and system software that can scale to massive amounts of data becomes a key challenge to both researchers and practitioners. Scalability in distributed system usually means that the performance of a system should increase proportionally with the increase in resources. However, this is not sufficient in the big data era. The system should be designed in a way so that all the five Vs of big data can be tackled. Driven by this insight, this special issue aims at presenting the current state-of-the-art research and future trends on various aspects of big data programming and system software techniques for big data processing and attempts towards building highly adaptive big data systems that can automatically adapt their behaviours to the amount of available resources. The major subjects cover methodologies, modelling, analysis and newly introduced applications. Besides the latest research achievements, this special issue also covers innovative commercial data management systems, innovative commercial applications of big data technology and experience in applying recent research advances to real-world problems.

@inproceedings{
 title = {Twister2: TSet high-performance iterative dataflow},
 type = {inproceedings},
 year = {2019},
 keywords = {Batch,Big data,Dataflow,Iterative,Mapreduce,Parallel programming,Stream},
 pages = {55-60},
 month = {5},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {1},
 id = {f11ba317-f161-3a42-aba4-bd4d585ce01e},
 created = {2019-10-01T17:21:01.592Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.134Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Wickramasinghe2019},
 private_publication = {false},
 abstract = {The dataflow model is slowly becoming the de facto standard for big data applications. While many popular frameworks are built around the dataflow model, very little research has been done on understanding the inner workings of the dataflow model. This has led to many inefficiencies in existing frameworks. It is important to note that understanding the relation between dataflow and HPC building blocks allows us to address and alleviate many of the fundamental inefficiencies in dataflow by learning from the extensive research literature in the HPC community. In this paper, we present TSet's, the dataflow abstraction of Twister2, which is a big data framework designed for high-performance dataflow and iterative computations. We discuss the dataflow model adopted by TSet's and the rationale behind implementing iteration handling at the worker level. Finally, we evaluate TSet's to show the performance of the framework.},
 bibtype = {inproceedings},
 author = {Wickramasinghe, Pulasthi and Kamburugamuve, Supun and Govindarajan, Kannan and Abeykoon, Vibhatha and Widanage, Chathura and Perera, Niranda and Uyar, Ahmet and Gunduz, Gurhan and Akkas, Selahattin and Fox, Geoffrey},
 doi = {10.1109/HPBDIS.2019.8735495},
 booktitle = {2019 International Conference on High Performance Big Data and Intelligent Systems, HPBD and IS 2019}
}

The dataflow model is slowly becoming the de facto standard for big data applications. While many popular frameworks are built around the dataflow model, very little research has been done on understanding the inner workings of the dataflow model. This has led to many inefficiencies in existing frameworks. It is important to note that understanding the relation between dataflow and HPC building blocks allows us to address and alleviate many of the fundamental inefficiencies in dataflow by learning from the extensive research literature in the HPC community. In this paper, we present TSet's, the dataflow abstraction of Twister2, which is a big data framework designed for high-performance dataflow and iterative computations. We discuss the dataflow model adopted by TSet's and the rationale behind implementing iteration handling at the worker level. Finally, we evaluate TSet's to show the performance of the framework.

@techreport{
 title = {Parallel Performance of Molecular Dynamics Trajectory Analysis},
 type = {techreport},
 year = {2019},
 keywords = {Big Data,Global Arrays,HDF5,HPC,MDAnalysis,MPI,MPI I/O,Molecular Dynamics,Python,Straggler,Trajectory Analysis},
 pages = {1-60},
 id = {1cd1748a-66c8-398f-8788-00d0b3a69597},
 created = {2019-10-01T17:21:01.838Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.902Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Khoshlessan2019},
 private_publication = {false},
 abstract = {The performance of biomolecular molecular dynamics (MD) simulations has steadily increased on modern high performance computing (HPC) resources but acceleration of the analysis of the output trajectories has lagged behind so that analyzing simulations is increasingly becoming a bottleneck. To close this gap, we studied the performance of parallel trajectory analysis with MPI and the Python MDAnalysis library on three different XSEDE supercomputers where trajectories were read from a Lustre parallel file system. We found that strong scaling performance was impeded by stragglers, MPI processes that were slower than the typical process and that therefore dominated the overall run time. Stragglers were less prevalent for compute-bound workloads, thus pointing to file reading as a crucial bottleneck for scaling. However, a more complicated picture emerged in which both the computation and the ingestion of data exhibited close to ideal strong scaling behavior whereas stragglers were primarily caused by either large MPI communication costs or long times to open the single shared trajectory file. We improved overall strong scaling performance by two different approaches to file access, namely subfiling (splitting the trajectory into as many trajectory segments as number of processes) and MPI-IO with Parallel HDF5 trajectory files. Applying these strategies, we obtained near ideal strong scaling on up to 384 cores (16 nodes). We summarize our lessons-learned in guidelines and strategies on how to take advantage of the available HPC resources to gain good scalability and potentially reduce trajectory analysis times by two orders of magnitude compared to the prevalent serial approach.},
 bibtype = {techreport},
 author = {Khoshlessan, Mahzad and Paraskevakos, Ioannis and Fox, Geoffrey C and Jha, Shantenu and Beckstein, Oliver}
}

The performance of biomolecular molecular dynamics (MD) simulations has steadily increased on modern high performance computing (HPC) resources but acceleration of the analysis of the output trajectories has lagged behind so that analyzing simulations is increasingly becoming a bottleneck. To close this gap, we studied the performance of parallel trajectory analysis with MPI and the Python MDAnalysis library on three different XSEDE supercomputers where trajectories were read from a Lustre parallel file system. We found that strong scaling performance was impeded by stragglers, MPI processes that were slower than the typical process and that therefore dominated the overall run time. Stragglers were less prevalent for compute-bound workloads, thus pointing to file reading as a crucial bottleneck for scaling. However, a more complicated picture emerged in which both the computation and the ingestion of data exhibited close to ideal strong scaling behavior whereas stragglers were primarily caused by either large MPI communication costs or long times to open the single shared trajectory file. We improved overall strong scaling performance by two different approaches to file access, namely subfiling (splitting the trajectory into as many trajectory segments as number of processes) and MPI-IO with Parallel HDF5 trajectory files. Applying these strategies, we obtained near ideal strong scaling on up to 384 cores (16 nodes). We summarize our lessons-learned in guidelines and strategies on how to take advantage of the available HPC resources to gain good scalability and potentially reduce trajectory analysis times by two orders of magnitude compared to the prevalent serial approach.

@inbook{
 type = {inbook},
 year = {2019},
 pages = {157-170},
 websites = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-5225-2255-3.ch092,http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-5225-7598-6.ch012},
 publisher = {IGI Global},
 city = {Hershey, PA, USA},
 id = {bce81e19-619f-361b-9476-891173448040},
 created = {2019-10-01T17:21:06.024Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:39.828Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2019a},
 source_type = {CHAP},
 private_publication = {false},
 abstract = {Computers accelerate our ability to achieve scientific breakthroughs. As technology evolves and new research needs come to light, the role for cyberinfrastructure as “knowledge” infrastructure continues to expand. In essence, cyberinfrastructure can be thought of as the integration of supercomputers, data resources, visualization, and people that extends the impact and utility of information technology. This article discusses cyberinfrastructure, the related topics of science gateways and campus bridging, and identifies future challenges and opportunities in cyberinfrastructure.},
 bibtype = {inbook},
 author = {Stewart, Craig A and Knepper, Richard and Link, Matthew R and Pierce, Marlon and Wernert, Eric and Wilkins-Diehr, Nancy},
 editor = {Mehdi Khosrow-Pour, D B A},
 doi = {10.4018/978-1-5225-7598-6.ch012},
 chapter = {Cyberinfrastructure, Cloud Computing, Science Gateways, Visualization, and Cyberinfrastructure Ease of Use},
 title = {Encyclopedia of Information Science and Technology, Fourth Edition}
}

Computers accelerate our ability to achieve scientific breakthroughs. As technology evolves and new research needs come to light, the role for cyberinfrastructure as “knowledge” infrastructure continues to expand. In essence, cyberinfrastructure can be thought of as the integration of supercomputers, data resources, visualization, and people that extends the impact and utility of information technology. This article discusses cyberinfrastructure, the related topics of science gateways and campus bridging, and identifies future challenges and opportunities in cyberinfrastructure.

@inproceedings{
 title = {Trusted CI Experiences in Cybersecurity and Service to Open Science},
 type = {inproceedings},
 year = {2019},
 pages = {1-8},
 publisher = {Association for Computing Machinery (ACM)},
 id = {d6670685-e93a-38b6-bc2d-8827dd13e994},
 created = {2019-10-01T17:21:14.222Z},
 accessed = {2019-09-11},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.724Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Adams2019},
 private_publication = {false},
 abstract = {This article describes experiences and lessons learned from the Trusted CI project, funded by the US National Science Foundation (NSF) to serve the community as the NSF Cybersecurity Center of Excellence (CCoE). Trusted CI is an effort to address cybersecurity for the open science community through a single organization that provides leadership, training, consulting, and knowledge to that community. The article describes the experiences and lessons learned of Trusted CI regarding both cybersecurity for open science and managing the process of providing centralized services to a broad and diverse community.},
 bibtype = {inproceedings},
 author = {Adams, Andrew and Jackson, Craig and Kiser, Ryan and Krenz, Mark and Marsteller, Jim and Miller, Barton P. and Peisert, Sean and Russell, Scott and Sons, Susan and Welch, Von and Zage, John and Avila, Kay and Basney, Jim and Brunson, Dana and Cowles, Robert and Dopheide, Jeannette and Fleury, Terry and Heymann, Elisa and Hudson, Florence},
 doi = {10.1145/3332186.3340601},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) (PEARC '19)}
}

This article describes experiences and lessons learned from the Trusted CI project, funded by the US National Science Foundation (NSF) to serve the community as the NSF Cybersecurity Center of Excellence (CCoE). Trusted CI is an effort to address cybersecurity for the open science community through a single organization that provides leadership, training, consulting, and knowledge to that community. The article describes the experiences and lessons learned of Trusted CI regarding both cybersecurity for open science and managing the process of providing centralized services to a broad and diverse community.

@inproceedings{
 title = {Integrity Protection for Scientific Workflow Data: Motivation and Initial Experiences},
 type = {inproceedings},
 year = {2019},
 pages = {1-8},
 publisher = {Association for Computing Machinery (ACM)},
 id = {6d46f628-0077-3408-b9f8-0c6201e5e7f8},
 created = {2019-10-01T17:21:14.292Z},
 accessed = {2019-09-11},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.750Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Rynge2019},
 private_publication = {false},
 abstract = {With the continued rise of scientific computing and the enormous increases in the size of data being processed, scientists must consider whether the processes for transmitting and storing data sufficiently assure the integrity of the scientific data. When integrity is not preserved, computations can fail and result in increased computational cost due to reruns, or worse, results can be corrupted in a manner not apparent to the scientist and produce invalid science results. Technologies such as TCP checksums, encrypted transfers, checksum validation, RAID and erasure coding provide integrity assurances at different levels, but they may not scale to large data sizes and may not cover a workflow from end-to-end, leaving gaps in which data corruption can occur undetected. In this paper we explore an approach of assuring data integrity - considering either malicious or accidental corruption - for workflow executions orchestrated by the Pegasus Workflow Management System. To validate our approach, we introduce Chaos Jungle - a toolkit providing an environment for validating integrity verification mechanisms by allowing researchers to introduce a variety of integrity errors during data transfers and storage. In addition to controlled experiments with Chaos Jungle, we provide analysis of integrity errors that we encountered when running production workflows.},
 bibtype = {inproceedings},
 author = {Rynge, Mats and Poehlman, William L. and Feltus, F. Alex and Vahi, Karan and Deelman, Ewa and Mandal, Anirban and Baldin, Ilya and Bhide, Omkar and Heiland, Randy and Welch, Von and Hill, Raquel},
 doi = {10.1145/3332186.3332222},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) (PEARC '19)}
}

With the continued rise of scientific computing and the enormous increases in the size of data being processed, scientists must consider whether the processes for transmitting and storing data sufficiently assure the integrity of the scientific data. When integrity is not preserved, computations can fail and result in increased computational cost due to reruns, or worse, results can be corrupted in a manner not apparent to the scientist and produce invalid science results. Technologies such as TCP checksums, encrypted transfers, checksum validation, RAID and erasure coding provide integrity assurances at different levels, but they may not scale to large data sizes and may not cover a workflow from end-to-end, leaving gaps in which data corruption can occur undetected. In this paper we explore an approach of assuring data integrity - considering either malicious or accidental corruption - for workflow executions orchestrated by the Pegasus Workflow Management System. To validate our approach, we introduce Chaos Jungle - a toolkit providing an environment for validating integrity verification mechanisms by allowing researchers to introduce a variety of integrity errors during data transfers and storage. In addition to controlled experiments with Chaos Jungle, we provide analysis of integrity errors that we encountered when running production workflows.

@article{
 title = {Intelligent systems for geosciences: An essential research agenda},
 type = {article},
 year = {2019},
 pages = {76-84},
 volume = {62},
 month = {1},
 publisher = {Association for Computing Machinery},
 day = {1},
 id = {5a1be824-dce6-3212-bb4c-cc507da1b092},
 created = {2019-10-01T17:21:16.980Z},
 accessed = {2019-08-15},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.709Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gil2019},
 private_publication = {false},
 abstract = {A research agenda for intelligent systems that will result in fundamental new capabilities for understanding the Earth system.},
 bibtype = {article},
 author = {Gil, Yolanda and Pierce, Suzanne A. and Babaie, Hassan and Banerjee, Arindam and Borne, Kirk and Bust, Gary and Cheatham, Michelle and Ebert-Uphoff, Imme and Gomes, Carla and Hill, Mary and Horel, John and Hsu, Leslie and Kinter, Jim and Knoblock, Craig and Krum, David and Kumar, Vipin and Lermusiaux, Pierre and Liu, Yan and North, Chris and Pankratius, Victor and Peters, Shanan and Plale, Beth and Pope, Allen and Ravela, Sai and Restrepo, Juan and Ridley, Aaron and Samet, Hanan and Shekhar, Shashi},
 doi = {10.1145/3192335},
 journal = {Communications of the ACM},
 number = {1}
}

A research agenda for intelligent systems that will result in fundamental new capabilities for understanding the Earth system.

@inproceedings{
 title = {Campus Champions: Building and sustaining a thriving community of practice around research computing and data},
 type = {inproceedings},
 year = {2019},
 pages = {1-7},
 publisher = {Association for Computing Machinery (ACM)},
 id = {d362d446-73e6-3178-9d0f-3a24d8d77e52},
 created = {2019-10-01T17:21:18.794Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.516Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Brazil2019},
 private_publication = {false},
 abstract = {The digital research landscape has changed dramatically over the years, as campuses across the nation have gained access to local research computing resources and services. The Campus Champions program, [1] founded in 2008, has also evolved with these changes to accommodate and support the diversity and growth of the research computing community that we have long supported. Our mission, to promote and facilitate the effective participation of a diverse national community of institutions in the application of advanced digital resources and services to accelerate scientific discovery and scholarly achievement, has been made possible by sustained funding from the National Science Foundation (NSF) via the eXtreme Science and Engineering Discovery Environment (XSEDE) [2] over the past ten years, and has enabled the Campus Champions program to support campus cyberinfrastructure and to foster a thriving community of practice [3] with nationwide impact. To facilitate the continued growth and sustainability of the Campus Champions program beyond XSEDE, and to provide a deeper understanding of the Champion culture and needs, the Champions Sustainability Working Group and the XSEDE Evaluation team conducted a climate study in 2018. The recommendations provided by the climate study will inform our Leadership Team and staff on how we may further our community outreach goals and plan for the future of the program. This paper will highlight and discuss the development, implementation, key findings, and recommendations from that study.},
 bibtype = {inproceedings},
 author = {Brazil, Marisa and Smith, Jack and Wernert, Julie and Brunson, Dana and Culich, Aaron and DeStefano, Lizanne and Jennewein, Douglas and Jolley, Tiffany and Middelkoop, Timothy and Neeman, Henry and Rivera, Lorna},
 doi = {10.1145/3332186.3332200},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

The digital research landscape has changed dramatically over the years, as campuses across the nation have gained access to local research computing resources and services. The Campus Champions program, [1] founded in 2008, has also evolved with these changes to accommodate and support the diversity and growth of the research computing community that we have long supported. Our mission, to promote and facilitate the effective participation of a diverse national community of institutions in the application of advanced digital resources and services to accelerate scientific discovery and scholarly achievement, has been made possible by sustained funding from the National Science Foundation (NSF) via the eXtreme Science and Engineering Discovery Environment (XSEDE) [2] over the past ten years, and has enabled the Campus Champions program to support campus cyberinfrastructure and to foster a thriving community of practice [3] with nationwide impact. To facilitate the continued growth and sustainability of the Campus Champions program beyond XSEDE, and to provide a deeper understanding of the Champion culture and needs, the Champions Sustainability Working Group and the XSEDE Evaluation team conducted a climate study in 2018. The recommendations provided by the climate study will inform our Leadership Team and staff on how we may further our community outreach goals and plan for the future of the program. This paper will highlight and discuss the development, implementation, key findings, and recommendations from that study.

@inproceedings{
 title = {Initial Thoughts on Cybersecurity And Reproducibility},
 type = {inproceedings},
 year = {2019},
 pages = {13-15},
 publisher = {Association for Computing Machinery (ACM)},
 id = {d64fa2e9-97ab-3daa-81f1-c30f312129d8},
 created = {2019-10-01T17:21:19.627Z},
 accessed = {2019-09-11},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.056Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Deelman2019},
 private_publication = {false},
 abstract = {Cybersecurity, which serves to protect computer systems and data from malicious and accidental abuse and changes, both supports and challenges the reproducibility of computational science. This position paper explores a research agenda by enumerating a set of two types of challenges that emerge at the intersection of cybersecurity and reproducibility: challenges that cybersecurity has in supporting the reproducibility of computational science, and challenges cybersecurity creates for reproducibility of computational science.},
 bibtype = {inproceedings},
 author = {Deelman, Ewa and Stodden, Victoria and Taufer, Michela and Welch, Von},
 doi = {10.1145/3322790.3330593},
 booktitle = {Proceedings of the 2nd International Workshop on Practical Reproducible Evaluation of Computer Systems (P-RECS '19)}
}

Cybersecurity, which serves to protect computer systems and data from malicious and accidental abuse and changes, both supports and challenges the reproducibility of computational science. This position paper explores a research agenda by enumerating a set of two types of challenges that emerge at the intersection of cybersecurity and reproducibility: challenges that cybersecurity has in supporting the reproducibility of computational science, and challenges cybersecurity creates for reproducibility of computational science.

@techreport{
 title = {Data Integrity Threat Model},
 type = {techreport},
 year = {2019},
 keywords = {Data Integrity,Non,OSCRP,Threat Model,malicious threat model},
 pages = {9},
 websites = {http://hdl.handle.net/2022/23225},
 month = {6},
 day = {27},
 city = {Bloomington, IN},
 institution = {ScholarWorks},
 id = {457f82a4-ca9a-3275-a86a-0d5339b92c18},
 created = {2019-10-01T17:21:20.524Z},
 accessed = {2019-08-20},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.072Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Welch2019},
 private_publication = {false},
 abstract = {This document uses OSCRP as a point of reference to construct a non-malicious Data Integrity Threat Model as a part of IRIS project. The goal of the project is to detect the source of unintentional integrity errors in the scientific workflow executions on distributed cyberinfrastructure.},
 bibtype = {techreport},
 author = {Welch, Von and Abhinit, Ishan}
}

This document uses OSCRP as a point of reference to construct a non-malicious Data Integrity Threat Model as a part of IRIS project. The goal of the project is to detect the source of unintentional integrity errors in the scientific workflow executions on distributed cyberinfrastructure.

@techreport{
 title = {A Guide for Software Assurance for SWIP},
 type = {techreport},
 year = {2019},
 pages = {1-37},
 websites = {https://pegasus.isi.edu/,http://hdl.handle.net/2022/23413},
 id = {d1dd9c75-c826-3608-9e07-32c01eea4b58},
 created = {2019-10-01T17:21:20.623Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.266Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Heiland2019},
 private_publication = {false},
 abstract = {The Scientific Workflow Integrity with Pegasus (SWIP) project adds data integrity checking to the Pegasus workflow management system (https://pegasus.isi.edu/). As part of SWIP, we performed software assurance (SwA) on the Pegasus software using the Software Assurance Marketplace (SWAMP, https://www.mir-swamp.org/). Initially, we planned to perform SwA only on the parts of the code base related to SWIP, i.e., only the code related to the data integrity checks. However, during the course of the SWIP project, a decision was made to perform SwA on the entire Pegasus code base. In addition, the project took on a research effort of trying to quantify differences in SwA results between Pegasus versions. We summarize our SwA process and results here. SwA results provide insight, but they are still subjective; developers of the software being assessed (Pegasus in this project) need to determine how those results need to be addressed.},
 bibtype = {techreport},
 author = {Heiland, Randy; and Rynge, Mats; and Vahi, Karan; and Deelman, Ewa; and Welch, Von}
}

The Scientific Workflow Integrity with Pegasus (SWIP) project adds data integrity checking to the Pegasus workflow management system (https://pegasus.isi.edu/). As part of SWIP, we performed software assurance (SwA) on the Pegasus software using the Software Assurance Marketplace (SWAMP, https://www.mir-swamp.org/). Initially, we planned to perform SwA only on the parts of the code base related to SWIP, i.e., only the code related to the data integrity checks. However, during the course of the SWIP project, a decision was made to perform SwA on the entire Pegasus code base. In addition, the project took on a research effort of trying to quantify differences in SwA results between Pegasus versions. We summarize our SwA process and results here. SwA results provide insight, but they are still subjective; developers of the software being assessed (Pegasus in this project) need to determine how those results need to be addressed.

@inproceedings{
 title = {Applicability Study of the PRIMAD Model to LIGO Gravitational Wave Search Workflows},
 type = {inproceedings},
 year = {2019},
 pages = {1-6},
 websites = {http://dl.acm.org/citation.cfm?doid=3322790.3330591},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {a8fc707b-a689-30c9-9a1f-4bfd41ec7f94},
 created = {2019-10-01T17:21:21.350Z},
 accessed = {2019-09-11},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.259Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Chapp2019},
 private_publication = {false},
 abstract = {The PRIMAD model with its six components (i.e., Platform, Research Objective, Implementation, Methods, Actors, and Data), provides an abstract taxonomy to represent computational experiments and enforce reproducibility by design. In this paper, we assess the model applicability to a set of Laser Interferometer Gravitational-Wave Observatory (LIGO) workflows from literature sources (i.e., published papers). Our work outlines potentials and limits of the model in terms of its abstraction levels and application process.},
 bibtype = {inproceedings},
 author = {Chapp, Dylan and Rorabaugh, Danny and Brown, Duncan A. and Deelman, Ewa and Vahi, Karan and Welch, Von and Taufer, Michela},
 doi = {10.1145/3322790.3330591},
 booktitle = {Proceedings of the 2nd International Workshop on Practical Reproducible Evaluation of Computer Systems - P-RECS '19}
}

The PRIMAD model with its six components (i.e., Platform, Research Objective, Implementation, Methods, Actors, and Data), provides an abstract taxonomy to represent computational experiments and enforce reproducibility by design. In this paper, we assess the model applicability to a set of Laser Interferometer Gravitational-Wave Observatory (LIGO) workflows from literature sources (i.e., published papers). Our work outlines potentials and limits of the model in terms of its abstraction levels and application process.

@inproceedings{
 title = {The USD Science Gateway},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3333254},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {979a5cd4-01d9-37c1-9720-f7cde99ec71b},
 created = {2019-10-01T17:21:22.601Z},
 accessed = {2019-08-15},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.454Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kleinsasser2019},
 private_publication = {false},
 abstract = {Science Gateways are virtual environments that accelerate scientific discovery by enabling scientific communities to more easily and effectively utilize distributed computing and data resources. Successful Science Gateways provide access to sophisticated and powerful resources, while shielding their users from the underlying complexities. Here we present updated work completed by the University of South Dakota (USD) Research Computing Group in conjunction with the Science Gateways Community Institute (SGCI) [1] and Science Gateways Research Center at Indiana University to set up a Science Gateway to access USD's high-performance computing resources. We also introduce improvements to the system since the previous presentation of our work. These resources are now available to both faculty and students and allow ease of access and use of USD's distributed computing and data resources. The implementation of this gateway project has been multifaceted and has included placement of federated user login, user facilitation and outreach, and integration of USD's cyberinfrastructure resources. We present this project as an example for other research computing groups so that they may learn from our successes and the challenges that we have overcome in providing this user resource. Additionally, this project serves to exemplify the importance of creating a broad user base of research computing infrastructure through the development of alternative user interfaces such as Science Gateways.},
 bibtype = {inproceedings},
 author = {Kleinsasser, Adison A. and Pamidighantam, Sudhakar and Jennewein, Douglas M. and Madison, Joseph D. and Christie, Marcus and Abeysinghe, Eroma and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/3332186.3333254},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

Science Gateways are virtual environments that accelerate scientific discovery by enabling scientific communities to more easily and effectively utilize distributed computing and data resources. Successful Science Gateways provide access to sophisticated and powerful resources, while shielding their users from the underlying complexities. Here we present updated work completed by the University of South Dakota (USD) Research Computing Group in conjunction with the Science Gateways Community Institute (SGCI) [1] and Science Gateways Research Center at Indiana University to set up a Science Gateway to access USD's high-performance computing resources. We also introduce improvements to the system since the previous presentation of our work. These resources are now available to both faculty and students and allow ease of access and use of USD's distributed computing and data resources. The implementation of this gateway project has been multifaceted and has included placement of federated user login, user facilitation and outreach, and integration of USD's cyberinfrastructure resources. We present this project as an example for other research computing groups so that they may learn from our successes and the challenges that we have overcome in providing this user resource. Additionally, this project serves to exemplify the importance of creating a broad user base of research computing infrastructure through the development of alternative user interfaces such as Science Gateways.

@article{
 title = {Community Organizations: Changing the Culture in Which Research Software Is Developed and Sustained},
 type = {article},
 year = {2019},
 keywords = {research software,scientific software,software community culture,software ecosystems,software productivity,software sustainability},
 pages = {8-24},
 volume = {21},
 websites = {https://ieeexplore.ieee.org/document/8565942/},
 month = {3},
 publisher = {IEEE Computer Society},
 day = {1},
 id = {825e2569-a861-3f98-8b0c-3fb1b105cbcd},
 created = {2019-10-01T17:21:24.571Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.746Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Katz2019},
 private_publication = {false},
 abstract = {Software is the key crosscutting technology that enables advances in mathematics, computer science, and domain-specific science and engineering to achieve robust simulations and analysis for science, engineering, and other research fields. However, software itself has not traditionally received focused attention from research communities; rather, software has evolved organically and inconsistently, with its development largely as by-products of other initiatives. Moreover, challenges in scientific software are expanding due to disruptive changes in computer hardware, increasing scale and complexity of data, and demands for more complex simulations involving multiphysics, multiscale modeling and outer-loop analysis. In recent years, community members have established a range of grass-roots organizations and projects to address these growing technical and social challenges in software productivity, quality, reproducibility, and sustainability. This article provides an overview of such groups and discusses opportunities to leverage their synergistic activities while nurturing work toward emerging software ecosystems.},
 bibtype = {article},
 author = {Katz, Daniel S. and McInnes, Lois Curfman and Bernholdt, David E. and Mayes, Abigail Cabunoc and Hong, Neil P. Chue and Duckles, Jonah and Gesing, Sandra and Heroux, Michael A. and Hettrick, Simon and Jimenez, Rafael C. and Pierce, Marlon and Weaver, Belinda and Wilkins-Diehr, Nancy},
 doi = {10.1109/MCSE.2018.2883051},
 journal = {Computing in Science & Engineering},
 number = {2}
}

Software is the key crosscutting technology that enables advances in mathematics, computer science, and domain-specific science and engineering to achieve robust simulations and analysis for science, engineering, and other research fields. However, software itself has not traditionally received focused attention from research communities; rather, software has evolved organically and inconsistently, with its development largely as by-products of other initiatives. Moreover, challenges in scientific software are expanding due to disruptive changes in computer hardware, increasing scale and complexity of data, and demands for more complex simulations involving multiphysics, multiscale modeling and outer-loop analysis. In recent years, community members have established a range of grass-roots organizations and projects to address these growing technical and social challenges in software productivity, quality, reproducibility, and sustainability. This article provides an overview of such groups and discusses opportunities to leverage their synergistic activities while nurturing work toward emerging software ecosystems.

@inproceedings{
 title = {Towards Run Time Estimation of the Gaussian Chemistry Code for SEAGrid Science Gateway},
 type = {inproceedings},
 year = {2019},
 pages = {1-8},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3338101},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {87b8c5f5-e8bd-3ad2-99bf-ae92f8d4eeac},
 created = {2019-10-01T17:21:25.957Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.266Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Beltre2019},
 private_publication = {false},
 abstract = {Accurate estimation of the run time of computational codes has a number of significant advantages for scientific computing. It is required information for optimal resource allocation, improving turnaround times and utilization of science gateways. Furthermore, it allows users to better plan and schedule their research, streamlining workflows and improving the overall productivity of cyberinfrastructure. Predicting run time is challenging, however. The inputs to scientific codes can be complex and high dimensional. Their relationship to the run time may be highly non-linear, and, in the most general case is completely arbitrary and thus unpredictable (i.e., simply a random mapping from inputs to run time). Most codes are not so arbitrary, however, and there has been significant prior research on predicting the run time of applications and workloads. Such predictions are generally application-specific, however. In this paper, we focus on the Gaussian computational chemistry code. We characterize a data set of runs from the SEAGrid science gateway with a number of different studies. We also explore a number of different potential regression methods and present promising future directions.},
 bibtype = {inproceedings},
 author = {Beltre, Angel and Zaman, Shehtab and Chiu, Kenneth and Pamidighantam, Sudhakar and Qiao, Xingye and Govindaraju, Madhusudhan},
 doi = {10.1145/3332186.3338101},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

Accurate estimation of the run time of computational codes has a number of significant advantages for scientific computing. It is required information for optimal resource allocation, improving turnaround times and utilization of science gateways. Furthermore, it allows users to better plan and schedule their research, streamlining workflows and improving the overall productivity of cyberinfrastructure. Predicting run time is challenging, however. The inputs to scientific codes can be complex and high dimensional. Their relationship to the run time may be highly non-linear, and, in the most general case is completely arbitrary and thus unpredictable (i.e., simply a random mapping from inputs to run time). Most codes are not so arbitrary, however, and there has been significant prior research on predicting the run time of applications and workloads. Such predictions are generally application-specific, however. In this paper, we focus on the Gaussian computational chemistry code. We characterize a data set of runs from the SEAGrid science gateway with a number of different studies. We also explore a number of different potential regression methods and present promising future directions.

@inproceedings{
 title = {Virtual Clusters in the Jetstream Cloud},
 type = {inproceedings},
 year = {2019},
 pages = {1-6},
 websites = {http://dl.acm.org/citation.cfm?doid=3355738.3355752},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {fa36eede-2175-3f71-883e-c98fc892ac11},
 created = {2019-10-01T17:21:26.261Z},
 accessed = {2019-09-24},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.812Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Coulter2019c},
 private_publication = {false},
 abstract = {We discuss our work providing resources for batch computing via the Jetstream cloud, in the form of SLURM clusters. While these are mainly used by science gateways, there have been a few used in the more traditional commandline manner. The flexible nature of these has also lent itself well to educational work, and has provided the basis for a very successful series of tutorials and workshops. This paper discusses the technical evolution of the Virtual Cluster product, and gives an overview of the science enabled. We discuss the challenges in supporting an ecosystem of these virtual clusters, and in supporting research on cloud resources in general.},
 bibtype = {inproceedings},
 author = {Coulter, J. Eric and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Pierce, Marlon},
 doi = {10.1145/3355738.3355752},
 booktitle = {Proceedings of the Humans in the Loop: Enabling and Facilitating Research on Cloud Computing - HARC '19}
}

We discuss our work providing resources for batch computing via the Jetstream cloud, in the form of SLURM clusters. While these are mainly used by science gateways, there have been a few used in the more traditional commandline manner. The flexible nature of these has also lent itself well to educational work, and has provided the basis for a very successful series of tutorials and workshops. This paper discusses the technical evolution of the Virtual Cluster product, and gives an overview of the science enabled. We discuss the challenges in supporting an ecosystem of these virtual clusters, and in supporting research on cloud resources in general.

@inproceedings{
 title = {Assessment of financial returns on investments in cyberinfrastructure facilities},
 type = {inproceedings},
 year = {2019},
 pages = {1-8},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3332228},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {40a7290c-9552-3036-aefd-d73576e5d06a},
 created = {2019-10-01T17:21:27.383Z},
 accessed = {2019-08-14},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.060Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2019b},
 private_publication = {false},
 abstract = {In recent years, considerable attention has been given to assessing the value of investments in cyberinfrastructure (CI). This paper includes a survey of current methods for the assessment of financial returns on investment (ROI) in CI. Applying the financial concept of ROI proves challenging with regard to a service that, in most academic environments, does not generate a "sold amount" such as one would find in the buying and selling of stocks. The paper concludes with a discussion of future research directions and challenges in the assessment of financial ROI in CI. This work is intended less as a definitive guide than as a starting point for further exploration in the assessment of CI's value for scientific research.},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Apon, Amy and Payne, Ron and Slavin, Shawn D. and Hancock, David Y. and Wernert, Julie and Furlani, Thomas and Lifka, David and Sill, Alan and Berente, Nicholas and McMullen, Donald F. and Cheatham, Thomas},
 doi = {10.1145/3332186.3332228},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning)  - PEARC '19}
}

In recent years, considerable attention has been given to assessing the value of investments in cyberinfrastructure (CI). This paper includes a survey of current methods for the assessment of financial returns on investment (ROI) in CI. Applying the financial concept of ROI proves challenging with regard to a service that, in most academic environments, does not generate a "sold amount" such as one would find in the buying and selling of stocks. The paper concludes with a discussion of future research directions and challenges in the assessment of financial ROI in CI. This work is intended less as a definitive guide than as a starting point for further exploration in the assessment of CI's value for scientific research.

@article{
 title = {The open diffusion data derivatives, brain data upcycling via integrated publishing of derivatives and reproducible open cloud services},
 type = {article},
 year = {2019},
 keywords = {Brain imaging,Cognitive neuroscience,Computational science,Magnetic resonance imaging,Network models},
 pages = {69},
 volume = {6},
 websites = {http://www.nature.com/articles/s41597-019-0073-y},
 month = {12},
 publisher = {Nature Publishing Group},
 day = {23},
 id = {65ec6c82-9364-3b53-b87f-1f2bff6c2d8e},
 created = {2019-10-01T17:21:28.091Z},
 accessed = {2019-08-14},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.404Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Avesani2019},
 private_publication = {false},
 abstract = {We describe the Open Diffusion Data Derivatives (O3D) repository: an integrated collection of preserved brain data derivatives and processing pipelines, published together using a single digital-object-identifier. The data derivatives were generated using modern diffusion-weighted magnetic resonance imaging data (dMRI) with diverse properties of resolution and signal-to-noise ratio. In addition to the data, we publish all processing pipelines (also referred to as open cloud services). The pipelines utilize modern methods for neuroimaging data processing (diffusion-signal modelling, fiber tracking, tractography evaluation, white matter segmentation, and structural connectome construction). The O3D open services can allow cognitive and clinical neuroscientists to run the connectome mapping algorithms on new, user-uploaded, data. Open source code implementing all O3D services is also provided to allow computational and computer scientists to reuse and extend the processing methods. Publishing both data-derivatives and integrated processing pipeline promotes practices for scientific reproducibility and data upcycling by providing open access to the research assets for utilization by multiple scientific communities.},
 bibtype = {article},
 author = {Avesani, Paolo and McPherson, Brent and Hayashi, Soichi and Caiafa, Cesar F. and Henschel, Robert and Garyfallidis, Eleftherios and Kitchell, Lindsey and Bullock, Daniel and Patterson, Andrew and Olivetti, Emanuele and Sporns, Olaf and Saykin, Andrew J. and Wang, Lei and Dinov, Ivo and Hancock, David and Caron, Bradley and Qian, Yiming and Pestilli, Franco},
 doi = {10.1038/s41597-019-0073-y},
 journal = {Scientific Data},
 number = {1}
}

We describe the Open Diffusion Data Derivatives (O3D) repository: an integrated collection of preserved brain data derivatives and processing pipelines, published together using a single digital-object-identifier. The data derivatives were generated using modern diffusion-weighted magnetic resonance imaging data (dMRI) with diverse properties of resolution and signal-to-noise ratio. In addition to the data, we publish all processing pipelines (also referred to as open cloud services). The pipelines utilize modern methods for neuroimaging data processing (diffusion-signal modelling, fiber tracking, tractography evaluation, white matter segmentation, and structural connectome construction). The O3D open services can allow cognitive and clinical neuroscientists to run the connectome mapping algorithms on new, user-uploaded, data. Open source code implementing all O3D services is also provided to allow computational and computer scientists to reuse and extend the processing methods. Publishing both data-derivatives and integrated processing pipeline promotes practices for scientific reproducibility and data upcycling by providing open access to the research assets for utilization by multiple scientific communities.

@inproceedings{
 title = {Assessment of non-financial returns on cyberinfrastructure},
 type = {inproceedings},
 year = {2019},
 pages = {1-10},
 websites = {http://dl.acm.org/citation.cfm?doid=3355738.3355749},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {77b5b68e-fcfe-3390-a0c7-859357595961},
 created = {2019-10-01T17:21:28.169Z},
 accessed = {2019-10-01},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.298Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2019c},
 private_publication = {false},
 abstract = {In recent years, considerable attention has been given to assessing the value of investments in cyberinfrastructure (CI). This paper focuses on assessment of value measured in ways other than financial benefits - what might well be termed impact or outcomes. This paper is a companion to a paper presented at the PEARC'19 conference, which focused on methods for assessing financial returns on investment. In this paper we focus on methods for assessing impacts such as effect on publication production, importance of publications, and assistance with major scientific accomplishments as signified by major awards. We in particular focus on the role of humans in the loop - humanware. This includes a brief description of the roles humans play in facilitating use of research cyberinfratructure - including clouds - and then a discussion of how those impacts have been assessed. Our conclusion overall is that there has been more progress in the past very few years in developing methods for the quantitative assessment of financial returns on investment than there has been in assessing non-quantitative impacts. There are a few clear actions that many research institutions could take to start better assessing the non-financial impacts of investment in cyberinfrastructure. However, there is a great need for assessment efforts to turn more attention to the assessment of non-financial benefits of investment in cyberinfrastructure, particularly the benefits of investing in humans and the benefits to humans who are involved in supporting and using cyberinfrastructure, including clouds.},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Apon, Amy and Hancock, David Y. and Furlani, Thomas and Sill, Alan and Wernert, Julie and Lifka, David and Berente, Nicholas and Cheatham, Thomas and Slavin, Shawn D.},
 doi = {10.1145/3355738.3355749},
 booktitle = {Proceedings of the Humans in the Loop: Enabling and Facilitating Research on Cloud Computing - HARC '19}
}

In recent years, considerable attention has been given to assessing the value of investments in cyberinfrastructure (CI). This paper focuses on assessment of value measured in ways other than financial benefits - what might well be termed impact or outcomes. This paper is a companion to a paper presented at the PEARC'19 conference, which focused on methods for assessing financial returns on investment. In this paper we focus on methods for assessing impacts such as effect on publication production, importance of publications, and assistance with major scientific accomplishments as signified by major awards. We in particular focus on the role of humans in the loop - humanware. This includes a brief description of the roles humans play in facilitating use of research cyberinfratructure - including clouds - and then a discussion of how those impacts have been assessed. Our conclusion overall is that there has been more progress in the past very few years in developing methods for the quantitative assessment of financial returns on investment than there has been in assessing non-quantitative impacts. There are a few clear actions that many research institutions could take to start better assessing the non-financial impacts of investment in cyberinfrastructure. However, there is a great need for assessment efforts to turn more attention to the assessment of non-financial benefits of investment in cyberinfrastructure, particularly the benefits of investing in humans and the benefits to humans who are involved in supporting and using cyberinfrastructure, including clouds.

@inproceedings{
 title = {Implementing a Flexible, Fault Tolerant Job Management System for Science Gateways},
 type = {inproceedings},
 year = {2019},
 pages = {1-8},
 websites = {http://dl.acm.org/citation.cfm?doid=3332186.3332233},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {81ea3925-59cb-30fb-8332-2af82619daec},
 created = {2019-10-01T17:21:29.424Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.062Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wannipurage2019},
 private_publication = {false},
 abstract = {This paper summarizes our experiences evaluating and deploying a new task execution management system within the open source Apache Airavata framework for science gateways. We base our choices on our operational requirements and experiences running Airavata software as a multi-tenanted production service for multiple gateway clients. Our considerations include integrating semi-independent components, making major upgrades to those components while retaining the system's overall functionality, and choosing between integrating third party and in-house developed components. While we focus on Apache Airavata as the platform for evaluation, our results should be of general interest. After considering the options of extensions to our previous, in-house job management system using Apache Kafka or replacing it with Kubernetes, we ultimately chose Apache Helix, primarily for its ability to execute multiple tasks coupled into directed acyclic graphs. We have integrated this approach into Apache Airavata and have tested extensively over several months with many thousands of jobs, both from our internal throughput testing and operational tests with early adopter science gateway clients. The new system has proven to be at least as reliable as the previous system with the advantages that we now have simplified maintenance, do not need to support an in-house system that required extensive developer training to modify, and can support more sophisticated job execution scenarios.},
 bibtype = {inproceedings},
 author = {Wannipurage, Dimuthu and Marru, Suresh and Piece, Marlon and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Christie, Marcus and Shenoy, Gourav and Dhamnaskar, Ajinkya and Jayathilaka, Lahiru},
 doi = {10.1145/3332186.3332233},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

This paper summarizes our experiences evaluating and deploying a new task execution management system within the open source Apache Airavata framework for science gateways. We base our choices on our operational requirements and experiences running Airavata software as a multi-tenanted production service for multiple gateway clients. Our considerations include integrating semi-independent components, making major upgrades to those components while retaining the system's overall functionality, and choosing between integrating third party and in-house developed components. While we focus on Apache Airavata as the platform for evaluation, our results should be of general interest. After considering the options of extensions to our previous, in-house job management system using Apache Kafka or replacing it with Kubernetes, we ultimately chose Apache Helix, primarily for its ability to execute multiple tasks coupled into directed acyclic graphs. We have integrated this approach into Apache Airavata and have tested extensively over several months with many thousands of jobs, both from our internal throughput testing and operational tests with early adopter science gateway clients. The new system has proven to be at least as reliable as the previous system with the advantages that we now have simplified maintenance, do not need to support an in-house system that required extensive developer training to modify, and can support more sophisticated job execution scenarios.

@inproceedings{
 title = {InterACTWEL Science Gateway for Adaptation Planning in Food-Energy-Water Sectors of Local Communities},
 type = {inproceedings},
 year = {2019},
 pages = {1-4},
 publisher = {Association for Computing Machinery (ACM)},
 id = {85a94c66-8600-38c2-9617-ba0ce96d7d38},
 created = {2019-10-01T17:21:29.637Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.125Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Babbar-Sebens2019},
 private_publication = {false},
 abstract = {Since their inception in mid 2000s, adoption of Science Gateways as interfaces and conduits for digital infrastructure needed in science and engineering research and education has significantly increased. This trend has also driven changes in the types of services and resources that are now being expected from the Science Gateways by a growing group of diverse end users. In this poster, we present a novel Science Gateway, InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land), which serves as a research cyberinfrastructure as well as an applied decision support system for adaptive natural resources management in interdependent food, energy, and water sectors. End users of this gateway include not only interdisciplinary technical and social science researchers, but also public and private sectoral stakeholders. The gateway is a collaboration between Oregon State University and Science Gateways Research Center, Pervasive Technology Institute at Indiana University, and is addressing challenges and solutions related to computational services, visualization techniques, advanced software applications, collaboration capabilities, cyber security and privacy, and data repositories unique to food-energy-water sectors and their stakeholders.},
 bibtype = {inproceedings},
 author = {Babbar-Sebens, Meghna and Rivera, Samuel and Abeysinghe, Eroma and Marru, Suresh and Pierce, Marlon and Coulter, Eric and Farahani, Majid and Wannipurage, Dimuthu and Christie, Marcus},
 doi = {10.1145/3332186.3333253},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (learning) - PEARC '19}
}

Since their inception in mid 2000s, adoption of Science Gateways as interfaces and conduits for digital infrastructure needed in science and engineering research and education has significantly increased. This trend has also driven changes in the types of services and resources that are now being expected from the Science Gateways by a growing group of diverse end users. In this poster, we present a novel Science Gateway, InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land), which serves as a research cyberinfrastructure as well as an applied decision support system for adaptive natural resources management in interdependent food, energy, and water sectors. End users of this gateway include not only interdisciplinary technical and social science researchers, but also public and private sectoral stakeholders. The gateway is a collaboration between Oregon State University and Science Gateways Research Center, Pervasive Technology Institute at Indiana University, and is addressing challenges and solutions related to computational services, visualization techniques, advanced software applications, collaboration capabilities, cyber security and privacy, and data repositories unique to food-energy-water sectors and their stakeholders.

@article{
 title = {Science gateways: Sustainability via on-campus teams},
 type = {article},
 year = {2019},
 keywords = {On-campus groups,Science gateways,Science gateways community institute,Sustainability},
 pages = {97-102},
 volume = {94},
 websites = {https://linkinghub.elsevier.com/retrieve/pii/S0167739X18315395},
 month = {5},
 publisher = {Elsevier B.V.},
 day = {1},
 id = {3ed3751c-51eb-3e95-ba4e-91948c1f58b2},
 created = {2019-10-01T17:21:29.947Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.858Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gesing2019},
 private_publication = {false},
 abstract = {The challenges for creators of specific science gateways are manifold, and the expertise needed for well-designed science gateways is very diverse. The sustainability of science gateways is crucial to serve communities effectively, efficiently and reliably. One measure to achieve greater sustainability of science gateways is establishing on-campus teams. Researchers are served more efficiently since the support by experienced developers reduces individual project investments, and a team can bring the diversity of required expertise for a well-designed science gateway. This paper goes into detail about the challenges and the benefits of on-campus groups and of sharing resources across a campus. We provide four successful cases, describe the services of the Science Gateways Community Institute (SGCI) to support the process in building such groups, and recommend strategies for using free campus resources.},
 bibtype = {article},
 author = {Gesing, Sandra and Lawrence, Katherine and Dahan, Maytal and Pierce, Marlon E. and Wilkins-Diehr, Nancy and Zentner, Michael},
 doi = {10.1016/j.future.2018.09.067},
 journal = {Future Generation Computer Systems}
}

The challenges for creators of specific science gateways are manifold, and the expertise needed for well-designed science gateways is very diverse. The sustainability of science gateways is crucial to serve communities effectively, efficiently and reliably. One measure to achieve greater sustainability of science gateways is establishing on-campus teams. Researchers are served more efficiently since the support by experienced developers reduces individual project investments, and a team can bring the diversity of required expertise for a well-designed science gateway. This paper goes into detail about the challenges and the benefits of on-campus groups and of sharing resources across a campus. We provide four successful cases, describe the services of the Science Gateways Community Institute (SGCI) to support the process in building such groups, and recommend strategies for using free campus resources.

@article{
 title = {Managing authentication and authorization in distributed science gateway middleware},
 type = {article},
 year = {2019},
 websites = {https://linkinghub.elsevier.com/retrieve/pii/S0167739X18314729},
 month = {10},
 id = {96738da1-39ae-3626-9d75-90effb06dc61},
 created = {2019-10-15T16:15:54.912Z},
 accessed = {2019-10-15},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.627Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Christie2019},
 private_publication = {false},
 abstract = {Establishing users’ identities and determining their permissions before they access research infrastructure resources are key features of science gateways. With many science gateways now relying on general purpose gateway platform services, the challenges of managing identity-derived features have expanded to include network-based authentication and authorization scenarios that connect science gateway tenants, science gateway platform middleware, and third party identity provider services, including campus identity management systems. This paper examines both architectural and implementation considerations for integrating these services. We provide a summary case study that further shows how end-to-end authentication and authorization can be provided between gateways, campus authentication systems, science gateway middleware, and campus computing resources. We conclude with observations on lifecycle management of third party components in science gateway platform services, which is an important consideration for both selection of new technologies and transitioning from older systems.},
 bibtype = {article},
 author = {Christie, Marcus A. and Bhandar, Anuj and Nakandala, Supun and Marru, Suresh and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Pierce, Marlon E.},
 doi = {10.1016/j.future.2019.07.018},
 journal = {Future Generation Computer Systems}
}

Establishing users’ identities and determining their permissions before they access research infrastructure resources are key features of science gateways. With many science gateways now relying on general purpose gateway platform services, the challenges of managing identity-derived features have expanded to include network-based authentication and authorization scenarios that connect science gateway tenants, science gateway platform middleware, and third party identity provider services, including campus identity management systems. This paper examines both architectural and implementation considerations for integrating these services. We provide a summary case study that further shows how end-to-end authentication and authorization can be provided between gateways, campus authentication systems, science gateway middleware, and campus computing resources. We conclude with observations on lifecycle management of third party components in science gateway platform services, which is an important consideration for both selection of new technologies and transitioning from older systems.

@inproceedings{
 title = {Scientific image restoration anywhere},
 type = {inproceedings},
 year = {2019},
 keywords = {Deep learning,Edge computing,Image restoration,Model quantization},
 pages = {8-13},
 websites = {http://arxiv.org/abs/1911.05878},
 month = {11},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {1},
 id = {2d2c9cdb-1b36-3c97-895e-6596834cd688},
 created = {2020-04-21T20:10:54.927Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.035Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Abeykoon2019a},
 private_publication = {false},
 abstract = {The use of deep learning models within scientific experimental facilities frequently requires low-latency inference, so that, for example, quality control operations can be performed while data are being collected. Edge computing devices can be useful in this context, as their low cost and compact form factor permit them to be co-located with the experimental apparatus. Can such devices, with their limited resources, can perform neural network feed-forward computations efficiently and effectively? We explore this question by evaluating the performance and accuracy of a scientific image restoration model, for which both model input and output are images, on edge computing devices. Specifically, we evaluate deployments of TomoGAN, an image-denoising model based on generative adversarial networks developed for low-dose x-ray imaging, on the Google Edge TPU and NVIDIA Jetson. We adapt TomoGAN for edge execution, evaluate model inference performance, and propose methods to address the accuracy drop caused by model quantization. We show that these edge computing devices can deliver accuracy comparable to that of a full-fledged CPU or GPU model, at speeds that are more than adequate for use in the intended deployments, denoising a 1024x1024 image in less than a second. Our experiments also show that the Edge TPU models can provide 3x faster inference response than a CPU-based model and 1.5x faster than an edge GPU-based model. This combination of high speed and low cost permits image restoration anywhere.},
 bibtype = {inproceedings},
 author = {Abeykoon, Vibhatha and Liu, Zhengchun and Kettimuthu, Rajkumar and Fox, Geoffrey and Foster, Ian},
 doi = {10.1109/XLOOP49562.2019.00007},
 booktitle = {Proceedings of XLOOP 2019: 1st Annual Workshop on Large-Scale Experiment-in-the-Loop Computing, Held in conjunction with SC 2019: The International Conference for High Performance Computing, Networking, Storage and Analysis}
}

The use of deep learning models within scientific experimental facilities frequently requires low-latency inference, so that, for example, quality control operations can be performed while data are being collected. Edge computing devices can be useful in this context, as their low cost and compact form factor permit them to be co-located with the experimental apparatus. Can such devices, with their limited resources, can perform neural network feed-forward computations efficiently and effectively? We explore this question by evaluating the performance and accuracy of a scientific image restoration model, for which both model input and output are images, on edge computing devices. Specifically, we evaluate deployments of TomoGAN, an image-denoising model based on generative adversarial networks developed for low-dose x-ray imaging, on the Google Edge TPU and NVIDIA Jetson. We adapt TomoGAN for edge execution, evaluate model inference performance, and propose methods to address the accuracy drop caused by model quantization. We show that these edge computing devices can deliver accuracy comparable to that of a full-fledged CPU or GPU model, at speeds that are more than adequate for use in the intended deployments, denoising a 1024x1024 image in less than a second. Our experiments also show that the Edge TPU models can provide 3x faster inference response than a CPU-based model and 1.5x faster than an edge GPU-based model. This combination of high speed and low cost permits image restoration anywhere.

@inproceedings{
 title = {Benchmarking Deep Learning for Time Series: Challenges and Directions},
 type = {inproceedings},
 year = {2019},
 keywords = {benchmark,deep learning,machine learning,performance,time series},
 pages = {5679-5682},
 month = {12},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {1},
 id = {c05351c4-ecbf-3608-bedc-3ca00c0d99fb},
 created = {2020-04-21T20:10:54.932Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.634Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Huang2019},
 private_publication = {false},
 abstract = {Deep learning for time series is an emerging area with close ties to industry, yet under represented in performance benchmarks for machine learning systems. In this paper, we present a landscape of deep learning applications applied to time series, and discuss the challenges and directions towards building a robust performance benchmark of deep learning workloads for time series data.},
 bibtype = {inproceedings},
 author = {Huang, Xinyuan and Fox, Geoffrey C. and Serebryakov, Sergey and Mohan, Ankur and Morkisz, Pawel and Dutta, Debojyoti},
 doi = {10.1109/BigData47090.2019.9005496},
 booktitle = {Proceedings - 2019 IEEE International Conference on Big Data, Big Data 2019}
}

Deep learning for time series is an emerging area with close ties to industry, yet under represented in performance benchmarks for machine learning systems. In this paper, we present a landscape of deep learning applications applied to time series, and discuss the challenges and directions towards building a robust performance benchmark of deep learning workloads for time series data.

@inproceedings{
 title = {Streaming Machine Learning Algorithms with Big Data Systems},
 type = {inproceedings},
 year = {2019},
 keywords = {Big-Data,Dataflow,Streaming Machine Learning},
 pages = {5661-5666},
 month = {12},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {1},
 id = {c96c8784-fa31-36a0-9404-ced31d1fcd89},
 created = {2020-04-21T20:10:55.147Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.636Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Abeykoon2019},
 private_publication = {false},
 abstract = {Designing low latency applications that can process large volumes data with higher efficiency is a challenging problem. With the limited time to process data, usage of online algorithms are becoming important in the big-data applications. Stream processing is a well-known area that has been studied for a long time. In this research, our objective is to use state of the art big-data analytic engines to implement online algorithms and compare the strengths and weaknesses in each system. We use a streaming version of Support Vector Machines (SVM) and KMeans to do the analysis. Apache Flink, Apache Storm and Twister2 streaming frameworks are used to implement these algorithms. Our study focuses on the efficiency of online training of these algorithms and the results show higher performance in Twister2 framework for these algorithms.},
 bibtype = {inproceedings},
 author = {Abeykoon, Vibhatha and Laszewski, Gregor Von and Kamburugamuve, Supun and Govindrarajan, Kannan and Wickramasinghe, Pulasthi and Widanage, Chathura and Perera, Niranda and Uyar, Ahmet and Gunduz, Gurhan and Akkas, Selahattin},
 doi = {10.1109/BigData47090.2019.9006337},
 booktitle = {Proceedings - 2019 IEEE International Conference on Big Data, Big Data 2019}
}

Designing low latency applications that can process large volumes data with higher efficiency is a challenging problem. With the limited time to process data, usage of online algorithms are becoming important in the big-data applications. Stream processing is a well-known area that has been studied for a long time. In this research, our objective is to use state of the art big-data analytic engines to implement online algorithms and compare the strengths and weaknesses in each system. We use a streaming version of Support Vector Machines (SVM) and KMeans to do the analysis. Apache Flink, Apache Storm and Twister2 streaming frameworks are used to implement these algorithms. Our study focuses on the efficiency of online training of these algorithms and the results show higher performance in Twister2 framework for these algorithms.

@article{
 title = {Automated Ice-Bottom Tracking of 2D and 3D Ice Radar Imagery Using Viterbi and TRW-S},
 type = {article},
 year = {2019},
 keywords = {Feature extraction,glaciology,ice thickness,ice tracking,radar tomography},
 pages = {3272-3285},
 volume = {12},
 month = {9},
 publisher = {Institute of Electrical and Electronics Engineers},
 day = {1},
 id = {a8fac9c7-a0b9-371f-b0cc-0cf9139884b1},
 created = {2020-04-21T20:10:55.220Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.740Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Berger2019},
 private_publication = {false},
 abstract = {Multichannel radar depth sounding systems are able to produce two-dimensional (2D) and three-dimensional (3D) imagery of the internal structure of polar ice sheets. Information such as ice thickness and surface elevation is extracted from these data and applied to research in ice flow modeling and ice mass balance calculations. Due to a large amount of data collected, we seek to automate the ice-bottom layer tracking and allow for efficient manual corrections when errors occur in the automated method. We present improvements made to previous implementations of the Viterbi and sequential tree-reweighted message passing (TRW-S) algorithms for ice-bottom extraction in 2D and 3D radar imagery. These improvements are in the form of novel cost functions that allow for the integration of further domain-specific knowledge into the cost calculations and provide additional evidence of the characteristics of the ice sheets surveyed. Along with an explanation of our modifications, we demonstrate the results obtained by our modified implementations of the two algorithms and by previously proposed solutions to this problem, when compared to manually corrected ground truth data. Furthermore, we perform a self-assessment of tracking results by analyzing differences in the estimated ice-bottom for surveyed locations where flight paths have crossed and, thus, two separate measurements have been made at the same location. Using our modified cost functions and preprocessing routines, we obtain significantly decreased mean error measurements from both algorithms, such as a 47% reduction in average tracking error in the case of 3D imagery between the original and our proposed implementation of TRW-S.},
 bibtype = {article},
 author = {Berger, Victor and Xu, Mingze and Al-Ibadi, Mohanad and Chu, Shane and Crandall, David and Paden, John and Fox, Geoffrey Charles},
 doi = {10.1109/JSTARS.2019.2930920},
 journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
 number = {9}
}

Multichannel radar depth sounding systems are able to produce two-dimensional (2D) and three-dimensional (3D) imagery of the internal structure of polar ice sheets. Information such as ice thickness and surface elevation is extracted from these data and applied to research in ice flow modeling and ice mass balance calculations. Due to a large amount of data collected, we seek to automate the ice-bottom layer tracking and allow for efficient manual corrections when errors occur in the automated method. We present improvements made to previous implementations of the Viterbi and sequential tree-reweighted message passing (TRW-S) algorithms for ice-bottom extraction in 2D and 3D radar imagery. These improvements are in the form of novel cost functions that allow for the integration of further domain-specific knowledge into the cost calculations and provide additional evidence of the characteristics of the ice sheets surveyed. Along with an explanation of our modifications, we demonstrate the results obtained by our modified implementations of the two algorithms and by previously proposed solutions to this problem, when compared to manually corrected ground truth data. Furthermore, we perform a self-assessment of tracking results by analyzing differences in the estimated ice-bottom for surveyed locations where flight paths have crossed and, thus, two separate measurements have been made at the same location. Using our modified cost functions and preprocessing routines, we obtain significantly decreased mean error measurements from both algorithms, such as a 47% reduction in average tracking error in the case of 3D imagery between the original and our proposed implementation of TRW-S.

@inproceedings{
 title = {InterACTWEL Science Gateway for Adaptation Planning in Food-Energy-Water Sectors of Local Communities: Data, Methods, Lessons Learned and Future Directions},
 type = {inproceedings},
 year = {2019},
 month = {12},
 publisher = {American Geophysical Union},
 day = {11},
 id = {9ce7f592-7a95-34d2-ac00-f54280cf3020},
 created = {2020-04-21T22:47:54.776Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.337Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Rivera2019},
 private_publication = {false},
 abstract = {The management of our limited natural resources has become increasingly critical due to climate change and pressures from growing populations. Unfortunately, these problems continue to be aggravated due to the lack of strategic coordination amongst the food, energy, and water (FEW) sectors. InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land) is envisioned to be transformational decision support cyberinfrastructure (DSC) with state-of-the-art analytics and visualization capabilities that will empower land, water, energy managers and food producers (i.e., FEW actors) to conceptualize and co-plan towards a resilient future for their local communities. Developed as a novel Science Gateway (ScG), InterACTWEL is aimed to help communities effectively coordinate and identify robust natural resources management decisions over time, and for long-term adaptation to acute or chronic perturbations that they do not have control of (e.g., changing state laws or climatic patterns).
To enable large communities to prepare for an evolving future of changes the development of this flexible, secure and human computation-base DSC required the creation of, access to, and use of complex and multi-sectoral datasets of varying sizes, advanced simulation models, large-scale optimization algorithms, visualization techniques for rendering of complex decision and goal spaces, as well as interfaces that facilitate end-user engagement and cognitive learning. Moreover, since the end users of this gateway include not only interdisciplinary technical and social science researchers, but also public and private sectoral stakeholders (e.g., farmers, energy producers, municipalities, governmental agencies, non-governmental organizations, tribes), the DSC had to support the needs and activities of a diverse personas of end-user. Using the Hermiston region, in Oregon, as a case study, this talk will describe the user base, functionalities, and services included in InterACTWEL. Furthermore, we will share current challenges, lessons learned and future directions of DSC that are intended to improve the adaptation planning and resiliency of local communities by sustaining translational research goals and long-lasting collaboration between researchers and citizens.},
 bibtype = {inproceedings},
 author = {Rivera, Samuel J. and Giles, Nicholas Alan and Tilt, Jenna and Reimer, Jeff and Murthy, Ganti and Mukhopadhyay, Snehasis and Durresi, Arjan and Marru, Suresh and Pierce, Marlon Edwin and Babbar-Sebens, Meghna},
 booktitle = {American Geophysical Union Fall Meeting}
}

The management of our limited natural resources has become increasingly critical due to climate change and pressures from growing populations. Unfortunately, these problems continue to be aggravated due to the lack of strategic coordination amongst the food, energy, and water (FEW) sectors. InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land) is envisioned to be transformational decision support cyberinfrastructure (DSC) with state-of-the-art analytics and visualization capabilities that will empower land, water, energy managers and food producers (i.e., FEW actors) to conceptualize and co-plan towards a resilient future for their local communities. Developed as a novel Science Gateway (ScG), InterACTWEL is aimed to help communities effectively coordinate and identify robust natural resources management decisions over time, and for long-term adaptation to acute or chronic perturbations that they do not have control of (e.g., changing state laws or climatic patterns). To enable large communities to prepare for an evolving future of changes the development of this flexible, secure and human computation-base DSC required the creation of, access to, and use of complex and multi-sectoral datasets of varying sizes, advanced simulation models, large-scale optimization algorithms, visualization techniques for rendering of complex decision and goal spaces, as well as interfaces that facilitate end-user engagement and cognitive learning. Moreover, since the end users of this gateway include not only interdisciplinary technical and social science researchers, but also public and private sectoral stakeholders (e.g., farmers, energy producers, municipalities, governmental agencies, non-governmental organizations, tribes), the DSC had to support the needs and activities of a diverse personas of end-user. Using the Hermiston region, in Oregon, as a case study, this talk will describe the user base, functionalities, and services included in InterACTWEL. Furthermore, we will share current challenges, lessons learned and future directions of DSC that are intended to improve the adaptation planning and resiliency of local communities by sustaining translational research goals and long-lasting collaboration between researchers and citizens.

@inproceedings{
 title = {CyberWater—An open and sustainable framework for diverse data and model integration},
 type = {inproceedings},
 year = {2019},
 month = {12},
 publisher = {AGU},
 day = {9},
 id = {d5250e15-a3f3-3065-afe5-083da98bc4dd},
 created = {2020-04-21T22:47:54.980Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.490Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Luna2019},
 private_publication = {false},
 abstract = {To tackle fundamental scientific questions regarding health, resilience and sustainability of water resources which encompass hydrological, biological, environmental, atmospheric, and other geosciences that define the Earth system, researchers need to be able to easily access diverse data sources and to also effectively incorporate these data into heterogeneous models. Furthermore, models need to be easily integrated, and many models require supercomputing power to run. To address these cyberinfrastructure challenges, a new sustainable and easy-to-use Open Data and Open Modeling framework called CyberWater is currently under development, which is a collaborative NSF-funded project among University of Pittsburgh, IUPUI/IU, CUAHSI, University of Iowa, NC State University, and Ball State University. CyberWater addresses the challenges of accessing heterogeneous data sources via the Open Data architecture which adopts a common internal data model and representation to facilitate the integration of various external data sources. Data Agents are used to handle remote data access protocols, metadata standards, and source-specific implementations. The Open Modeling architecture allows different models to be easily integrated into CyberWater via Model Agents. CyberWater adopts a graphical scientific workflow system (VisTrails) and offers generic tools to help users develop model agents without coding.
At present, CyberWater can run a land surface model (LSM) of Variability Infiltration Capacity (VIC) model and a hydrological model of Distributed Hydrology Soil Vegetation Model (DHSVM), and let the user to easily couple an LSM with a routing model. Examples will be presented to illustrate the CyberWater system via the automatic flow from accessing data to model simulation results in a user-friendly workflow-controlled environment. A diverse set of new tools such as spatial and temporal transformations, GIS-related analyses, and high performance computing is under development. All of the current and future functionalities of CyberWater will not only be thoroughly tested and evaluated by the development team but also by the water community. The CyberWater is aimed for the next generation of open data and modeling framework in cyberinfrastructure for broad earth science communities.},
 bibtype = {inproceedings},
 author = {Luna, Daniel and Chen, Ranran and Yuan, Cao and Liang, Yao and Liang, Xu and Bales, Jerad and Castronova, Anthony M and Demir, Ibrahim and Hooper, Richard P and Krajewski, Witold F and Lin, Lan and Mantilla, Ricardo and Pamidighantam, Sudhakar and Song, Fengguang and Zhang, Yang},
 booktitle = {American Geophysical Union Fall Meeting}
}

To tackle fundamental scientific questions regarding health, resilience and sustainability of water resources which encompass hydrological, biological, environmental, atmospheric, and other geosciences that define the Earth system, researchers need to be able to easily access diverse data sources and to also effectively incorporate these data into heterogeneous models. Furthermore, models need to be easily integrated, and many models require supercomputing power to run. To address these cyberinfrastructure challenges, a new sustainable and easy-to-use Open Data and Open Modeling framework called CyberWater is currently under development, which is a collaborative NSF-funded project among University of Pittsburgh, IUPUI/IU, CUAHSI, University of Iowa, NC State University, and Ball State University. CyberWater addresses the challenges of accessing heterogeneous data sources via the Open Data architecture which adopts a common internal data model and representation to facilitate the integration of various external data sources. Data Agents are used to handle remote data access protocols, metadata standards, and source-specific implementations. The Open Modeling architecture allows different models to be easily integrated into CyberWater via Model Agents. CyberWater adopts a graphical scientific workflow system (VisTrails) and offers generic tools to help users develop model agents without coding. At present, CyberWater can run a land surface model (LSM) of Variability Infiltration Capacity (VIC) model and a hydrological model of Distributed Hydrology Soil Vegetation Model (DHSVM), and let the user to easily couple an LSM with a routing model. Examples will be presented to illustrate the CyberWater system via the automatic flow from accessing data to model simulation results in a user-friendly workflow-controlled environment. A diverse set of new tools such as spatial and temporal transformations, GIS-related analyses, and high performance computing is under development. All of the current and future functionalities of CyberWater will not only be thoroughly tested and evaluated by the development team but also by the water community. The CyberWater is aimed for the next generation of open data and modeling framework in cyberinfrastructure for broad earth science communities.

@inproceedings{
 title = {Stakeholder-Driven Adaptation Planning of Food-Energy-Water Nexus in Local Communities},
 type = {inproceedings},
 year = {2019},
 month = {12},
 publisher = {AGU},
 day = {12},
 id = {ece0ee51-362c-312d-927c-5c15ad745a5c},
 created = {2020-04-21T22:47:55.060Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.484Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Babbar-Sebens2019a},
 private_publication = {false},
 abstract = {For long it has been realized that the complexity in the interconnectedness between humans and the natural environment is an important justification for engaging stakeholders in any natural resource management problem. The same argument can be extended to the management of the food-energy-water (FEW) nexus that requires multiple considerations including the relationship of humans with water, energy, and land resources, and the economic sectors that depend on these resources. Especially, managing the nexus for an uncertain future will require better understanding of how humans respond to diverse stresses and threats, such as climate change impacts, changing political and socio-economic considerations, natural hazards, etc. This understanding can then be used to improve how stakeholder knowledge is formulated into “stakeholder-relevant” mitigation and adaptation decision problems. In this presentation, the process of discovering and formulating an adaptation planning problem for a community of large number of FEW stakeholders is examined in the regions of Umatilla and Morrow counties in Oregon. Discovery of the FEW nexus adaptation problem was accomplished via co-production of knowledge, involving researchers and community stakeholders. The iterative process of discovery was then used to formulate and validate a conceptual model of how the community’s FEW nexus would adapt to changes in water rights and availability in the region. The conceptual model was validated using stakeholder engagement, before the model was quantified using hierarchical decision approaches and integrated within a novel cyberinfrastructure, InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land).},
 bibtype = {inproceedings},
 author = {Babbar-Sebens, Meghna and Rivera, Samuel J. and Giles, Nicholas Alan and Tilt, Jenna and Reimer, Jeff and Murthy, Ganti and Mukhopadhyay, Snehasis and Durresi, Arjan and Marru, Suresh and Pierce, Marlon Edwin},
 booktitle = {American Geophysical Union Fall Meeting}
}

For long it has been realized that the complexity in the interconnectedness between humans and the natural environment is an important justification for engaging stakeholders in any natural resource management problem. The same argument can be extended to the management of the food-energy-water (FEW) nexus that requires multiple considerations including the relationship of humans with water, energy, and land resources, and the economic sectors that depend on these resources. Especially, managing the nexus for an uncertain future will require better understanding of how humans respond to diverse stresses and threats, such as climate change impacts, changing political and socio-economic considerations, natural hazards, etc. This understanding can then be used to improve how stakeholder knowledge is formulated into “stakeholder-relevant” mitigation and adaptation decision problems. In this presentation, the process of discovering and formulating an adaptation planning problem for a community of large number of FEW stakeholders is examined in the regions of Umatilla and Morrow counties in Oregon. Discovery of the FEW nexus adaptation problem was accomplished via co-production of knowledge, involving researchers and community stakeholders. The iterative process of discovery was then used to formulate and validate a conceptual model of how the community’s FEW nexus would adapt to changes in water rights and availability in the region. The conceptual model was validated using stakeholder engagement, before the model was quantified using hierarchical decision approaches and integrated within a novel cyberinfrastructure, InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land).

@inproceedings{
 title = {Advancing Access to Global Flood Modeling and Alerting using the PDC DisasterAWARE Platform and Remote Sensing Technologies},
 type = {inproceedings},
 year = {2019},
 month = {12},
 publisher = {AGU},
 day = {9},
 id = {d74b5da7-9269-3187-a2a2-7e71537fc777},
 created = {2020-04-21T22:47:55.062Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.279Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Glasscoe2019},
 private_publication = {false},
 abstract = {Disaster managers face significant challenges in managing essential information for preparedness, response, and recovery efforts. The development of an open access, global flood alerting system for effective classification of potential impacts and the formulation of effective emergency response measures requires incorporation of a wide variety of flood model and remote sensing (RS) data sources from multiple platforms.
We seek to rapidly classify flood severity and alerts based on the potential for impacts in a way similar to the USGS PAGER impact analysis for earthquakes. The proposed approach is to use a model of models that leverages existing capabilities and products, as well as the incorporation of RS products for ground-truthing model results and delineation of flood areas. The ground-truthing would provide continued adjustment of parameters in the model of models approach. Administrative area resilience indicators for flood affected areas will support incorporation of vulnerability and lack of capacity into rapid assessment of potential impacts.

The work will build on the current DisasterAWARE platform, operated by the Pacific Disaster Center (PDC) and currently providing global multi-hazard alerting and Situational Awareness information to the public. DisasterAWARE supports hazard monitoring and early warning needs of Disaster Managers across the globe, as well as providing the Common Operating Picture in support of US DOD’s humanitarian assistance and disaster response mission. However, the current systems lack global flood alerting or incorporation of a RS component that will allow near real-time validation of simulated flood modeling results.

The use of RS images and derivative products will enable users to validate in near real-time the results of high-resolution simulation modeling outputs to be incorporated into DisasterAWARE, and that are often imported into loss estimation software to quantify disaster impacts. The objective is to integrate and leverage publicly-available global flood modeling sources with available RS platforms (satellite and airborne) to produce a robust and comprehensive platform for flood damage assessment and alerting. This will help communities build their capacity and resilience by rapidly responding to flood impacts.},
 bibtype = {inproceedings},
 author = {Glasscoe, Margaret T and Bausch, Douglas B and Chiesa, Chris and Hampe, Greg and Tiampo, Kristy French and Eguchi, Ronald T and Huyck, Charles K and Pierce, Marlon Edwin and Wang, Jun and Chen, ZhiQiang and Kar, Bandana and Schumann, Guy},
 booktitle = {American Geophysical Union Fall Meeting}
}

Disaster managers face significant challenges in managing essential information for preparedness, response, and recovery efforts. The development of an open access, global flood alerting system for effective classification of potential impacts and the formulation of effective emergency response measures requires incorporation of a wide variety of flood model and remote sensing (RS) data sources from multiple platforms. We seek to rapidly classify flood severity and alerts based on the potential for impacts in a way similar to the USGS PAGER impact analysis for earthquakes. The proposed approach is to use a model of models that leverages existing capabilities and products, as well as the incorporation of RS products for ground-truthing model results and delineation of flood areas. The ground-truthing would provide continued adjustment of parameters in the model of models approach. Administrative area resilience indicators for flood affected areas will support incorporation of vulnerability and lack of capacity into rapid assessment of potential impacts. The work will build on the current DisasterAWARE platform, operated by the Pacific Disaster Center (PDC) and currently providing global multi-hazard alerting and Situational Awareness information to the public. DisasterAWARE supports hazard monitoring and early warning needs of Disaster Managers across the globe, as well as providing the Common Operating Picture in support of US DOD’s humanitarian assistance and disaster response mission. However, the current systems lack global flood alerting or incorporation of a RS component that will allow near real-time validation of simulated flood modeling results. The use of RS images and derivative products will enable users to validate in near real-time the results of high-resolution simulation modeling outputs to be incorporated into DisasterAWARE, and that are often imported into loss estimation software to quantify disaster impacts. The objective is to integrate and leverage publicly-available global flood modeling sources with available RS platforms (satellite and airborne) to produce a robust and comprehensive platform for flood damage assessment and alerting. This will help communities build their capacity and resilience by rapidly responding to flood impacts.

@article{
 title = {ManyClasses 1: Assessing the generalizable effect of immediate versus delayed feedback across many college classes},
 type = {article},
 year = {2019},
 keywords = {Cognitive Psychology,Educational Psychology,Social and Behavioral Sciences},
 publisher = {PsyArXiv},
 id = {ab024069-fbff-3859-9487-750285fcd7ae},
 created = {2020-04-22T21:44:56.762Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T20:38:12.643Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fyfe2019},
 private_publication = {false},
 abstract = {Psychology researchers have long attempted to identify educational practices that improve student learning. However, experimental research on these practices is often conducted in laboratory contexts or in a single class, threatening the external validity of the results. In this paper, we establish an experimental paradigm for evaluating the benefits of recommended practices across a variety of authentic educational contexts – a model we call ManyClasses. The core feature is that researchers examine the same research question and measure the same experimental effect across many classes spanning a range of topics, institutions, teacher implementations, and student populations. We report the first ManyClasses study, which examined how the timing of feedback on class assignments, either immediate or delayed by a few days, affected subsequent performance on class assessments. Across XX classes, [summarize effect of feedback timing, including key moderators]. More broadly, these findings provide evidence regarding the feasibility of conducting within-class randomized experiments across a range of naturally occurring learning environments.},
 bibtype = {article},
 author = {Fyfe, Emily and Leeuw, Joshua de and Carvalho, Paulo and Goldstone, Robert and Motz, Benjamin},
 doi = {10.31234/OSF.IO/4MVYH},
 journal = {Advances in Methods and Practices in Psychological Science}
}

Psychology researchers have long attempted to identify educational practices that improve student learning. However, experimental research on these practices is often conducted in laboratory contexts or in a single class, threatening the external validity of the results. In this paper, we establish an experimental paradigm for evaluating the benefits of recommended practices across a variety of authentic educational contexts – a model we call ManyClasses. The core feature is that researchers examine the same research question and measure the same experimental effect across many classes spanning a range of topics, institutions, teacher implementations, and student populations. We report the first ManyClasses study, which examined how the timing of feedback on class assignments, either immediate or delayed by a few days, affected subsequent performance on class assessments. Across XX classes, [summarize effect of feedback timing, including key moderators]. More broadly, these findings provide evidence regarding the feasibility of conducting within-class randomized experiments across a range of naturally occurring learning environments.

@article{
 title = {Rice Galaxy: an open resource for plant science},
 type = {article},
 year = {2019},
 pages = {1-14},
 volume = {8},
 websites = {http://orcid.org/0000-0001-8512-144X},
 id = {1dcdb9df-e76f-384c-9ed4-5838d0a0cad7},
 created = {2020-04-22T21:44:57.004Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.687Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Juanillas2019},
 private_publication = {false},
 abstract = {Background: Rice molecular genetics, breeding, genetic diversity, and allied research (such as rice-pathogen interaction) have adopted sequencing technologies and high-density genotyping platforms for genome variation analysis and gene discovery. Germplasm collections representing rice diversity, improved varieties, and elite breeding materials are accessible through rice gene banks for use in research and breeding, with many having genome sequences and high-density genotype data available. Combining phenotypic and genotypic information on these accessions enables genome-wide association analysis, which is driving quantitative trait loci discovery and molecular marker development. Comparative sequence analyses across quantitative trait loci regions facilitate the discovery of novel alleles. Analyses involving DNA sequences and large genotyping matrices for thousands of samples, however, pose a challenge to non−computer savvy rice researchers. Findings: The Rice Galaxy resource has shared datasets that include high-density genotypes from the 3,000 Rice Genomes project and sequences with corresponding annotations from 9 published rice genomes. The Rice Galaxy web server and deployment installer includes tools for designing single-nucleotide polymorphism assays, analyzing genome-wide association studies, population diversity, rice−bacterial pathogen diagnostics, and a suite of published genomic prediction methods.},
 bibtype = {article},
 author = {Juanillas, Venice and Dereeper, Alexis and Beaume, Nicolas and Droc, Gaetan and Dizon, Joshua and Mendoza, John Robert and Perdon, Jon Peter and Mansueto, Locedie and Triplett, Lindsay and Lang, Jillian and Zhou, Gabriel and Ratharanjan, Kunalan and Plale, Beth and Haga, Jason and Leach, Jan E and Ruiz, Manuel and Thomson, Michael and Alexandrov, Nickolai and Larmande, Pierre and Kretzschmar, Tobias and Mauleon, Ramil P},
 doi = {10.1093/gigascience/giz028}
}

Background: Rice molecular genetics, breeding, genetic diversity, and allied research (such as rice-pathogen interaction) have adopted sequencing technologies and high-density genotyping platforms for genome variation analysis and gene discovery. Germplasm collections representing rice diversity, improved varieties, and elite breeding materials are accessible through rice gene banks for use in research and breeding, with many having genome sequences and high-density genotype data available. Combining phenotypic and genotypic information on these accessions enables genome-wide association analysis, which is driving quantitative trait loci discovery and molecular marker development. Comparative sequence analyses across quantitative trait loci regions facilitate the discovery of novel alleles. Analyses involving DNA sequences and large genotyping matrices for thousands of samples, however, pose a challenge to non−computer savvy rice researchers. Findings: The Rice Galaxy resource has shared datasets that include high-density genotypes from the 3,000 Rice Genomes project and sequences with corresponding annotations from 9 published rice genomes. The Rice Galaxy web server and deployment installer includes tools for designing single-nucleotide polymorphism assays, analyzing genome-wide association studies, population diversity, rice−bacterial pathogen diagnostics, and a suite of published genomic prediction methods.

@inproceedings{
 title = {The validity and utility of activity logs as a measure of student engagement},
 type = {inproceedings},
 year = {2019},
 keywords = {LMS,Student engagement,Trace data,Web logs},
 pages = {300-309},
 websites = {http://dl.acm.org/citation.cfm?doid=3303772.3303789},
 month = {3},
 publisher = {Association for Computing Machinery},
 day = {4},
 city = {New York, New York, USA},
 id = {99e57b6b-2cd4-36ae-89a5-46950b49af4a},
 created = {2020-04-22T21:44:57.118Z},
 accessed = {2020-04-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.708Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Motz2019},
 private_publication = {false},
 abstract = {Learning management system (LMS) web logs provide granular, near-real-time records of student behavior as learners interact with online course materials in digital learning environments. However, it remains unclear whether LMS activity indeed reflects behavioral properties of student engagement, and it also remains unclear how to deal with variability in LMS usage across a diversity of courses. In this study, we evaluate whether instructors' subjective ratings of their students' engagement are related to features of LMS activity for 9,021 students enrolled in 473 for-credit courses. We find that estimators derived from LMS web logs are closely related to instructor ratings of engagement, however, we also observe that there is not a single generic relationship between activity and engagement, and what constitutes the behavioral components of “engagement” will be contingent on course structure. However, for many of these courses, modeled engagement scores are comparable to instructors' ratings in their sensitivity for predicting academic performance. As long as they are tuned to the differences between courses, activity indices from LMS web logs can provide a valid and useful proxy measure of student engagement.},
 bibtype = {inproceedings},
 author = {Motz, Benjamin and Quick, Joshua and Schroeder, Noah and Zook, Jordon and Gunkel, Matthew},
 doi = {10.1145/3303772.3303789},
 booktitle = {Proceedings of the 9th International Conference on Learning Analytics & Knowledge}
}

Learning management system (LMS) web logs provide granular, near-real-time records of student behavior as learners interact with online course materials in digital learning environments. However, it remains unclear whether LMS activity indeed reflects behavioral properties of student engagement, and it also remains unclear how to deal with variability in LMS usage across a diversity of courses. In this study, we evaluate whether instructors' subjective ratings of their students' engagement are related to features of LMS activity for 9,021 students enrolled in 473 for-credit courses. We find that estimators derived from LMS web logs are closely related to instructor ratings of engagement, however, we also observe that there is not a single generic relationship between activity and engagement, and what constitutes the behavioral components of “engagement” will be contingent on course structure. However, for many of these courses, modeled engagement scores are comparable to instructors' ratings in their sensitivity for predicting academic performance. As long as they are tuned to the differences between courses, activity indices from LMS web logs can provide a valid and useful proxy measure of student engagement.

@article{
 title = {Cyberinfrastructure Requirements to Enhance Multi-messenger Astrophysics},
 type = {article},
 year = {2019},
 websites = {http://arxiv.org/abs/1903.04590},
 month = {3},
 day = {11},
 id = {2bcb05ba-9331-34ad-b009-f2f4cba04f58},
 created = {2020-04-22T21:44:57.173Z},
 accessed = {2020-04-22},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.833Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Chang2019},
 private_publication = {false},
 abstract = {The identification of the electromagnetic counterpart of the gravitational wave event, GW170817, and discovery of neutrinos and gamma-rays from TXS 0506+056 heralded the new era of multi-messenger astrophysics. As the number of multi-messenger events rapidly grow over the next decade, the cyberinfrastructure requirements to handle the increase in data rates, data volume, need for event follow up, and analysis across the different messengers will also explosively grow. The cyberinfrastructure requirements to enhance multi-messenger astrophysics will both be a major challenge and opportunity for astronomers, physicists, computer scientists and cyberinfrastructure specialists. Here we outline some of these requirements and argue for a distributed cyberinfrastructure institute for multi-messenger astrophysics to meet these challenges.},
 bibtype = {article},
 author = {Chang, Philip and Allen, Gabrielle and Anderson, Warren and Bianco, Federica B. and Bloom, Joshua S. and Brady, Patrick R. and Brazier, Adam and Cenko, S. Bradley and Couch, Sean M. and DeYoung, Tyce and Deelman, Ewa and Etienne, Zachariah B and Foley, Ryan J. and Fox, Derek B and Golkhou, V. Zach and Grant, Darren R and Hanna, Chad and Holley-Bockelmann, Kelly and Howell, D. Andrew and Huerta, E. A. and Johnson, Margaret W. G. and Juric, Mario and Kaplan, David L. and Katz, Daniel S. and Keivani, Azadeh and Kerzendorf, Wolfgang and Kopper, Claudio and Lam, Michael T. and Lehner, Luis and Marka, Zsuzsa and Marka, Szabolcs and Nabrzyski, Jarek and Narayan, Gautham and O'Shea, Brian W. and Petravick, Donald and Quick, Rob and Street, Rachel A. and Taboada, Ignacio and Timmes, Frank and Turk, Matthew J. and Weltman, Amanda and Zhang, Zhao}
}

The identification of the electromagnetic counterpart of the gravitational wave event, GW170817, and discovery of neutrinos and gamma-rays from TXS 0506+056 heralded the new era of multi-messenger astrophysics. As the number of multi-messenger events rapidly grow over the next decade, the cyberinfrastructure requirements to handle the increase in data rates, data volume, need for event follow up, and analysis across the different messengers will also explosively grow. The cyberinfrastructure requirements to enhance multi-messenger astrophysics will both be a major challenge and opportunity for astronomers, physicists, computer scientists and cyberinfrastructure specialists. Here we outline some of these requirements and argue for a distributed cyberinfrastructure institute for multi-messenger astrophysics to meet these challenges.

@inproceedings{
 title = {Transparency by Design in eScience Research},
 type = {inproceedings},
 year = {2019},
 pages = {428-431},
 month = {3},
 publisher = {Institute of Electrical and Electronics Engineers (IEEE)},
 day = {20},
 id = {1bc9237c-02c5-3984-af73-3ed9e574c76d},
 created = {2020-04-22T21:44:57.327Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.887Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Plale2019a},
 private_publication = {false},
 abstract = {Both the landscape of eScience research and the environment in which the research is conducted are undergoing change. Transparency by design in eScience is proposed as a term to describe transparency in eScience practices, processes, methodologies, and research results. We break down different aspects of transparency and urge the eScience community towards a renewed commitment to scientific rigor because of the important role that we as scientists have to improve society and protect the good will that society has bestowed on science.},
 bibtype = {inproceedings},
 author = {Plale, Beth},
 doi = {10.1109/escience.2019.00055},
 booktitle = {International Conference on eScience (eScience)}
}

Both the landscape of eScience research and the environment in which the research is conducted are undergoing change. Transparency by design in eScience is proposed as a term to describe transparency in eScience practices, processes, methodologies, and research results. We break down different aspects of transparency and urge the eScience community towards a renewed commitment to scientific rigor because of the important role that we as scientists have to improve society and protect the good will that society has bestowed on science.

@article{
 title = {Extending XSEDE Innovations to Campus Cyberinfrastructure - The XSEDE National Integration Toolkit},
 type = {article},
 year = {2019},
 keywords = {Clusters,OpenHPC,Scientific Computing,System Administration,XCRI,XSEDE},
 pages = {16-20},
 volume = {10},
 websites = {https://doi.org/10.22369/issn.2153-4136/10/1/3,http://www.jocse.org/articles/10/1/3/},
 month = {1},
 id = {a60a10c7-8332-30bc-aa0b-dc866d99c2d6},
 created = {2020-04-27T19:54:26.446Z},
 accessed = {2020-04-27},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T23:36:28.819Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Coulter2019d},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {XSEDE Service Providers (SPs) and resources have the benefit of years of testing and implementation, tuning and configuration, and the development of specific tools to help users and systems make the best use of these resources. Cyberinfrastructure professionals at the campus level are often charged with building computer resources which are compared to these national-level resources. While organizations and companies exist that guide cyberinfras-tructure configuration choices down certain paths, there is no easy way to distribute the long-term knowledge of the XSEDE project to campus CI professionals. The XSEDE Cyberinfrastructure Resource Integration team has created a variety of toolkits to enable easy knowledge and best-practice transfer from XESDE SPs to campus CI professionals. The XSEDE National Integration Toolkit (XNIT) provides the software used on most XSEDE systems in an effort to propagate the best practices and knowledge of XSEDE resources. XNIT includes basic tools and configuration that make it simpler for a campus cluster to have the same software set and many of the advantages and XSEDE SP resource affords. In this paper, we will detail the steps taken to build such a library of software and discuss the challenges involved in disseminating awareness of toolkits among cyberin-frastructure professionals. We will also describe our experiences in updating the XNIT to be compatible with the OpenHPC project, which forms the basis of many new HPC systems, and appears situated to become the de-facto choice of management software provider for many HPC centers.},
 bibtype = {article},
 author = {Coulter, Eric and Sprouse, Jodie and Reynolds, Resa and Knepper, Richard},
 doi = {10.22369/issn.2153-4136/10/1/3},
 journal = {The Journal of Computational Science Education},
 number = {1}
}

XSEDE Service Providers (SPs) and resources have the benefit of years of testing and implementation, tuning and configuration, and the development of specific tools to help users and systems make the best use of these resources. Cyberinfrastructure professionals at the campus level are often charged with building computer resources which are compared to these national-level resources. While organizations and companies exist that guide cyberinfras-tructure configuration choices down certain paths, there is no easy way to distribute the long-term knowledge of the XSEDE project to campus CI professionals. The XSEDE Cyberinfrastructure Resource Integration team has created a variety of toolkits to enable easy knowledge and best-practice transfer from XESDE SPs to campus CI professionals. The XSEDE National Integration Toolkit (XNIT) provides the software used on most XSEDE systems in an effort to propagate the best practices and knowledge of XSEDE resources. XNIT includes basic tools and configuration that make it simpler for a campus cluster to have the same software set and many of the advantages and XSEDE SP resource affords. In this paper, we will detail the steps taken to build such a library of software and discuss the challenges involved in disseminating awareness of toolkits among cyberin-frastructure professionals. We will also describe our experiences in updating the XNIT to be compatible with the OpenHPC project, which forms the basis of many new HPC systems, and appears situated to become the de-facto choice of management software provider for many HPC centers.

@techreport{
 title = {Jetstream (NSF Award 1445604) Annual Report: December 1, 2018 – November 28, 2019},
 type = {techreport},
 year = {2019},
 websites = {https://scholarworks.iu.edu/dspace/handle/2022/24806},
 id = {c4707ace-6983-3730-81b8-86c09525f3e3},
 created = {2020-09-11T15:29:42.158Z},
 accessed = {2020-09-11},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:00.749Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {HancockDavidY;MerchantNirav;LoweJohnMichael;FischerJeremy;LimingLee;TaylorJames;AfganEnis;TurnerGeorge;SkidmoreEdwin;BeckBrainW.;Snapp-ChildsWinona;FosterIan;Vaughn2019},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Hancock, David Y; Merchant, Nirav; Lowe, John Michael; Fischer, Jeremy; Liming, Lee; Taylor, James; Afgan, Enis; Turner, George; Skidmore, Edwin; Beck, Brain W.; Snapp-Childs, Winona; Foster, Ian; Vaughn, Matthew}
}

@inproceedings{
 title = {HarpLDA+: Optimizing latent dirichlet allocation for parallel efficiency},
 type = {inproceedings},
 year = {2018},
 keywords = {Algorithm optimization; Collective communications,Big data,Learning systems; Statistics},
 pages = {243-252},
 volume = {2018-Janua},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85047792604&doi=10.1109%2FBigData.2017.8257932&partnerID=40&md5=634b9836d11831c51e661b9e5e90bcd9},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 id = {ce97ebfa-0470-3e41-8c53-c8c9e7f45275},
 created = {2018-06-25T18:22:28.432Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:33:20.152Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Peng2018243},
 source_type = {conference},
 notes = {cited By 0; Conference of 5th IEEE International Conference on Big Data, Big Data 2017 ; Conference Date: 11 December 2017 Through 14 December 2017; Conference Code:134260},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 abstract = {Latent Dirichlet Allocation (LDA) is a widely used machine learning technique in topic modeling and data analysis. Training large LDA models on big datasets involves dynamic and irregular computation patterns and is a major challenge to both algorithm optimization and system design. In this paper, we present a comprehensive benchmarking of our novel synchronized LDA training system HarpLDA+ based on Hadoop and Java. It demonstrates impressive performance when compared to three other MPI/C++ based state-of-the-art systems, which are LightLDA, F+NomadLDA, and WarpLDA. HarpLDA+ uses optimized collective communication with a timer control for load balance, leading to stable scalability in both shared-memory and distributed systems. We demonstrate in the experiments that HarpLDA+ is effective in reducing synchronization and communication overhead and outperforms the other three LDA training systems. © 2017 IEEE.},
 bibtype = {inproceedings},
 author = {Peng, B and Zhang, B and Chen, L and Avram, M and Henschel, R and Stewart, C and Zhu, S and McCallum, E and Smith, L and Zahniser, T and Omer, J and Qiu, J},
 editor = {Obradovic Z. Baeza-Yates R., Kepner J Nambiar R Wang C Toyoda M Suzumura T Hu X Cuzzocrea A Baeza-Yates R Tang J Zang H Nie J.-Y. Ghosh R},
 doi = {10.1109/BigData.2017.8257932},
 booktitle = {Proceedings - 2017 IEEE International Conference on Big Data, Big Data 2017}
}

Latent Dirichlet Allocation (LDA) is a widely used machine learning technique in topic modeling and data analysis. Training large LDA models on big datasets involves dynamic and irregular computation patterns and is a major challenge to both algorithm optimization and system design. In this paper, we present a comprehensive benchmarking of our novel synchronized LDA training system HarpLDA+ based on Hadoop and Java. It demonstrates impressive performance when compared to three other MPI/C++ based state-of-the-art systems, which are LightLDA, F+NomadLDA, and WarpLDA. HarpLDA+ uses optimized collective communication with a timer control for load balance, leading to stable scalability in both shared-memory and distributed systems. We demonstrate in the experiments that HarpLDA+ is effective in reducing synchronization and communication overhead and outperforms the other three LDA training systems. © 2017 IEEE.

@inproceedings{
 title = {The Science Gateways Community Institute at Two Years},
 type = {inproceedings},
 year = {2018},
 keywords = {Science gateways,Software institutes,Software sustainability},
 pages = {1-8},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3219142},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {104eb022-7dd6-3644-a951-7bd81d55ffb6},
 created = {2019-10-01T17:20:13.105Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.952Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wilkins-Diehr2018},
 private_publication = {false},
 abstract = {The Science Gateways Community Institute was one of the first two software institutes funded by the National Science Foundation's Office of Advanced Cyberinfrastructure in August, 2016. The structure of and services offered by the institute were developed as a result of seven years of planning grants that funded focus groups, a 5000-person survey and the development of a strategic plan. Now two years in, we provide an overview of the institute's service offerings and their usage, reflect on the experiences of some early clients, review our approaches to metrics and evaluation, and describe some lessons learned. We also describe the lightweight, adaptive management approach employed by the institute. SGCI is organized into five service areas: Incubator, Extended Developer Support, Scientific Software Collaborative, Community Engagement and Exchange, and Workforce Development. This paper will highlight early successes in all five areas, from client achievements to conference experiences to our impact on students. We highlight areas where the institute has evolved - based on community feedback - from what was originally envisioned. We describe our use of the Entrepreneurial Operating System as a lightweight management approach for a highly adaptive organization. Finally, we include early plans for the execution phase of the institute. © 2018 Copyright is held by the owner/author(s). Publication rights licensed to ACM.},
 bibtype = {inproceedings},
 author = {Wilkins-Diehr, Nancy and Zentner, Michael and Pierce, Marlon and Dahan, Maytal and Lawrence, Katherine and Hayden, Linda and Mullinix, Nayiri},
 doi = {10.1145/3219104.3219142},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

The Science Gateways Community Institute was one of the first two software institutes funded by the National Science Foundation's Office of Advanced Cyberinfrastructure in August, 2016. The structure of and services offered by the institute were developed as a result of seven years of planning grants that funded focus groups, a 5000-person survey and the development of a strategic plan. Now two years in, we provide an overview of the institute's service offerings and their usage, reflect on the experiences of some early clients, review our approaches to metrics and evaluation, and describe some lessons learned. We also describe the lightweight, adaptive management approach employed by the institute. SGCI is organized into five service areas: Incubator, Extended Developer Support, Scientific Software Collaborative, Community Engagement and Exchange, and Workforce Development. This paper will highlight early successes in all five areas, from client achievements to conference experiences to our impact on students. We highlight areas where the institute has evolved - based on community feedback - from what was originally envisioned. We describe our use of the Entrepreneurial Operating System as a lightweight management approach for a highly adaptive organization. Finally, we include early plans for the execution phase of the institute. © 2018 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

@inproceedings{
 title = {The CSBG - LSU Gateway},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,Bioinformatics,Computational System Biology,Science Gateway},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229245},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {ad1aabad-bc79-3f51-aa09-14456eb9b34e},
 created = {2019-10-01T17:20:13.350Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.496Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Abeysinghe2018},
 private_publication = {false},
 abstract = {Science gateways are identified as an effective way to publish and distribute software for research communities without the burden of learning HPC (High Performance Computer) systems. In the past, researchers were expected to have in-depth knowledge about using HPC systems for computations along with their respective science field in order to do effective research. Science gateways eliminate the need to learn HPC systems and allows the research communities to focus more on their science and let the gateway handle communicating with HPCs. In this poster we are presenting the science gateway project of CSBG (Computational System Biology Group - www.brylinski.org) of Department of Biological Sciences with Center for Computation & Technology at LSU (Louisiana State University). The gateway project was initiated in order to provide CSBG software tools as a service through a science gateway.},
 bibtype = {inproceedings},
 author = {Abeysinghe, Eroma and Brylinski, Michal and Christie, Marcus and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/3219104.3229245},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Science gateways are identified as an effective way to publish and distribute software for research communities without the burden of learning HPC (High Performance Computer) systems. In the past, researchers were expected to have in-depth knowledge about using HPC systems for computations along with their respective science field in order to do effective research. Science gateways eliminate the need to learn HPC systems and allows the research communities to focus more on their science and let the gateway handle communicating with HPCs. In this poster we are presenting the science gateway project of CSBG (Computational System Biology Group - www.brylinski.org) of Department of Biological Sciences with Center for Computation & Technology at LSU (Louisiana State University). The gateway project was initiated in order to provide CSBG software tools as a service through a science gateway.

@inproceedings{
 title = {Apache Airavata Resource Allocation System: A Tool for Allocating Resources in Science Gateways},
 type = {inproceedings},
 year = {2018},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229271},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {e3a692a1-682d-34ba-80f7-5d22a35a71ac},
 created = {2019-10-01T17:20:16.005Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.084Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Phulwani2018},
 private_publication = {false},
 abstract = {Science Gateways provide user environments and a set of supporting services that help researchers make effective and enhanced use of a diverse set of computing, storage, and related resources. In a software framework like Airavata, the distributed computing resources such as local clusters, supercomputers, computational grids, and computing clouds are shared among multiple researchers. Hence it requires an allocations process to track the demand for their resources, understand the scientific objectives of their users, and decide among competing needs when faced with user demand that is greater than the available resources can supply. We describe the design and a prototype implementation in this presentation.},
 bibtype = {inproceedings},
 author = {Phulwani, Harsha and Thapa, Madrina and Marru, Suresh and Pierce, Marlon and Pamidighantam, Sudhakar and Christie, Marcus},
 doi = {10.1145/3219104.3229271},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Science Gateways provide user environments and a set of supporting services that help researchers make effective and enhanced use of a diverse set of computing, storage, and related resources. In a software framework like Airavata, the distributed computing resources such as local clusters, supercomputers, computational grids, and computing clouds are shared among multiple researchers. Hence it requires an allocations process to track the demand for their resources, understand the scientific objectives of their users, and decide among competing needs when faced with user demand that is greater than the available resources can supply. We describe the design and a prototype implementation in this presentation.

@inproceedings{
 title = {Supporting Science Gateways Using Apache Airavata and SciGaP Services},
 type = {inproceedings},
 year = {2018},
 keywords = {Cyberinfrastructure,Science gateways,Software as a service},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229240},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {fecae67f-d76e-338d-ae21-1847ddbea86e},
 created = {2019-10-01T17:20:20.323Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.031Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pierce2018},
 private_publication = {false},
 abstract = {The Science Gateways Platform as a service (SciGaP.org) project provides a rapid development and stable hosting platform for a wide range of science gateways that focus on software as a service. Based on the open source Apache Airavata project, SciGaP services include user management, workflow execution management, computational experiment archiving and access, and sharing services that allow users to share results and other digital artifacts. SciGaP services are multi-tenanted, with clients accessing services through a well-defined, programming language-independent API. SciGaP services can be integrated into web, mobile, and desktop clients. To simplify development for new clients, SciGaP includes the PGA, a generic PHP-based gateway client for SciGaP services that also acts as a reference implementation of the API. Several example gateways using these services are summarized. © 2018 Copyright held by the owner/author(s).},
 bibtype = {inproceedings},
 author = {Pierce, Marlon and Marru, Suresh and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Christie, Marcus and Wannipurage, Dimuthu},
 doi = {10.1145/3219104.3229240},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

The Science Gateways Platform as a service (SciGaP.org) project provides a rapid development and stable hosting platform for a wide range of science gateways that focus on software as a service. Based on the open source Apache Airavata project, SciGaP services include user management, workflow execution management, computational experiment archiving and access, and sharing services that allow users to share results and other digital artifacts. SciGaP services are multi-tenanted, with clients accessing services through a well-defined, programming language-independent API. SciGaP services can be integrated into web, mobile, and desktop clients. To simplify development for new clients, SciGaP includes the PGA, a generic PHP-based gateway client for SciGaP services that also acts as a reference implementation of the API. Several example gateways using these services are summarized. © 2018 Copyright held by the owner/author(s).

@article{
 title = {Fracture Advancing Step Tectonics Observed in the Yuha Desert and Ocotillo, CA, Following the 2010 M w 7.2 El Mayor-Cucapah Earthquake},
 type = {article},
 year = {2018},
 keywords = {GPS,UAVSAR,earthquake,fault,geodetic imaging,stepover},
 pages = {456-472},
 volume = {5},
 websites = {http://doi.wiley.com/10.1029/2017EA000351},
 month = {9},
 publisher = {Wiley-Blackwell Publishing Ltd},
 day = {1},
 id = {66121872-12de-3182-a2cd-f1cd035a5237},
 created = {2019-10-01T17:20:21.526Z},
 accessed = {2019-08-19},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.418Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Donnellan2018},
 private_publication = {false},
 abstract = {©2018. The Authors. Uninhabited aerial vehicle synthetic aperture radar (UAVSAR) observations 2009–2017 of the Yuha Desert area and Global Positioning System (GPS) time series encompassing the region reveal a northward migrating pattern of deformation following the 4 April 2010 Mw7.2 El Mayor-Cucapah (EMC) earthquake. The north end of the EMC rupture exhibits an asymmetric pattern of deformation that is substantial and smooth northeast of the rupture and limited but with surface fracturing slip northwest. The earthquake triggered ~1 cm of surface coseismic slip at the Yuha fault, which continued to slip postseismically. 2.5 cm of Yuha fault slip occurred by the time of the 15 June 2010 Mw5.7 Ocotillo aftershock and 5 cm of slip occurred by 2017 following a logarithmic afterslip decay 16-day timescale. The Ocotillo aftershock triggered 1.4 cm of slip on a northwest trend extending to the Elsinore fault and by 7 years after the EMC earthquake 2.4 cm of slip had accumulated with a distribution following an afterslip function with a 16-day timescale consistent with other earthquakes and a rate strengthening upper crustal sedimentary layer. GPS data show broad coseismic uplift of the Salton Trough and delayed postseismic motion that may be indicative of fluid migration there and subsidence west of the rupture extension, which continues following the earthquake. The data indicate that the Elsinore, Laguna Salada, and EMC ruptures are part of the same fault system. The results also suggest that north-south shortening and east-west extension across the region drove fracture advancing step tectonics north of the EMC earthquake rupture.},
 bibtype = {article},
 author = {Donnellan, Andrea and Parker, Jay and Heflin, Michael and Lyzenga, Gregory and Moore, Angelyn and Ludwig, Lisa Grant and Rundle, John and Wang, Jun and Pierce, Marlon},
 doi = {10.1029/2017EA000351},
 journal = {Earth and Space Science},
 number = {9}
}

©2018. The Authors. Uninhabited aerial vehicle synthetic aperture radar (UAVSAR) observations 2009–2017 of the Yuha Desert area and Global Positioning System (GPS) time series encompassing the region reveal a northward migrating pattern of deformation following the 4 April 2010 Mw7.2 El Mayor-Cucapah (EMC) earthquake. The north end of the EMC rupture exhibits an asymmetric pattern of deformation that is substantial and smooth northeast of the rupture and limited but with surface fracturing slip northwest. The earthquake triggered ~1 cm of surface coseismic slip at the Yuha fault, which continued to slip postseismically. 2.5 cm of Yuha fault slip occurred by the time of the 15 June 2010 Mw5.7 Ocotillo aftershock and 5 cm of slip occurred by 2017 following a logarithmic afterslip decay 16-day timescale. The Ocotillo aftershock triggered 1.4 cm of slip on a northwest trend extending to the Elsinore fault and by 7 years after the EMC earthquake 2.4 cm of slip had accumulated with a distribution following an afterslip function with a 16-day timescale consistent with other earthquakes and a rate strengthening upper crustal sedimentary layer. GPS data show broad coseismic uplift of the Salton Trough and delayed postseismic motion that may be indicative of fluid migration there and subsidence west of the rupture extension, which continues following the earthquake. The data indicate that the Elsinore, Laguna Salada, and EMC ruptures are part of the same fault system. The results also suggest that north-south shortening and east-west extension across the region drove fracture advancing step tectonics north of the EMC earthquake rupture.

@inproceedings{
 title = {Evaluating NextCloud as a File Storage for Apache Airavata},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,File Storage,File Transfer,NextCloud,WebDAV},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229270},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {36bce677-451f-3965-a4d4-f542d3b36a5c},
 created = {2019-10-01T17:20:23.138Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.797Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kariyattin2018},
 private_publication = {false},
 abstract = {Science gateways enable researchers from broad communities to access advanced computing and storage resources. The researchers analyze large amounts of data using the compute resources and the generated results, usually files are saved in the storage. Consider a scenario where a researcher has large output data files of historically run experiments on an external server. If the researcher wants to move the data to the gateway storage, then the only way to do it is through data transfer. This task would be cumbersome and time consuming. The paper discusses an approach through which historic or any data existing on a different server or in a cloud storage (Google Drive) or in an object storage (Amazon S3) can be ingested into the existing gateway without actually transferring it to the server. We discuss about a software called NextCloud and how it can be used as a gateway storage by integrating it with Apache Airavata. Airavata currently uses local file storage to store user related data files. On the client side, Airavata clients use different protocols like HTTP and SFTP for file transfer. NextCloud is an open source file share and communication platform that provides a common file access layer through its universal file access to different data sources. Integrating NextCloud with Airavata could solve the problem of providing unified file transfer API across all the Airavata clients. As NextCloud supports various external storages, its integration with Airavata would also enable the data ingestion and importing large data from different storage sources to Airavata.},
 bibtype = {inproceedings},
 author = {Kariyattin, Sachin and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/3219104.3229270},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Science gateways enable researchers from broad communities to access advanced computing and storage resources. The researchers analyze large amounts of data using the compute resources and the generated results, usually files are saved in the storage. Consider a scenario where a researcher has large output data files of historically run experiments on an external server. If the researcher wants to move the data to the gateway storage, then the only way to do it is through data transfer. This task would be cumbersome and time consuming. The paper discusses an approach through which historic or any data existing on a different server or in a cloud storage (Google Drive) or in an object storage (Amazon S3) can be ingested into the existing gateway without actually transferring it to the server. We discuss about a software called NextCloud and how it can be used as a gateway storage by integrating it with Apache Airavata. Airavata currently uses local file storage to store user related data files. On the client side, Airavata clients use different protocols like HTTP and SFTP for file transfer. NextCloud is an open source file share and communication platform that provides a common file access layer through its universal file access to different data sources. Integrating NextCloud with Airavata could solve the problem of providing unified file transfer API across all the Airavata clients. As NextCloud supports various external storages, its integration with Airavata would also enable the data ingestion and importing large data from different storage sources to Airavata.

@inproceedings{
 title = {Scaling JupyterHub Using Kubernetes on Jetstream Cloud},
 type = {inproceedings},
 year = {2018},
 keywords = {Cloud Computing,JupyterHub,Kubernetes,Magnum,OpenStack,Unidata,Workforce Development},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229249},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {fc553946-212b-396c-83bf-e9b38820388d},
 created = {2019-10-01T17:20:23.818Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.786Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Sarajlic2018a},
 private_publication = {false},
 abstract = {Unidata, an NSF funded project that started in 1983, is a diverse community of education and research institutions with the common goal of sharing geoscience data and the tools to access and visualize that data. Unidata provides weather observations and other data, software tools, and support to enhance Earth-system education and research, and continuously examines ways of adapting their workflows for new technologies to maximize the reach of their education and research efforts. In support of Unidata objectives to host workshops for atmospheric data analysis using JupyterHub, we explore a cloud computing approach leveraging Kubernetes coupled with JupyterHub that when combined will provide a solution for researchers and students to pull data from Unidata and burst onto Jetstream cloud by requesting resources dynamically via easy to use JupyterHub. More specifically, on Jetstream, Kubernetes is used for automating deployment and scaling of domain specific containerized applications, and JupyterHub is used for spawning multiple hubs within the same Kubernetes cluster instance that will be used for supporting classroom settings. JupyterHub's modular kernel feature will support dynamic needs of classroom application requirements. The proposed approach will serve as an end-to-end solution for researchers to execute their workflows, with JupyterHub serving as a powerful tool for user training and next-generation workforce development in atmospheric sciences.},
 bibtype = {inproceedings},
 author = {Sarajlic, Semir and Chastang, Julien and Marru, Suresh and Fischer, Jeremy and Lowe, Mike},
 doi = {10.1145/3219104.3229249},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Unidata, an NSF funded project that started in 1983, is a diverse community of education and research institutions with the common goal of sharing geoscience data and the tools to access and visualize that data. Unidata provides weather observations and other data, software tools, and support to enhance Earth-system education and research, and continuously examines ways of adapting their workflows for new technologies to maximize the reach of their education and research efforts. In support of Unidata objectives to host workshops for atmospheric data analysis using JupyterHub, we explore a cloud computing approach leveraging Kubernetes coupled with JupyterHub that when combined will provide a solution for researchers and students to pull data from Unidata and burst onto Jetstream cloud by requesting resources dynamically via easy to use JupyterHub. More specifically, on Jetstream, Kubernetes is used for automating deployment and scaling of domain specific containerized applications, and JupyterHub is used for spawning multiple hubs within the same Kubernetes cluster instance that will be used for supporting classroom settings. JupyterHub's modular kernel feature will support dynamic needs of classroom application requirements. The proposed approach will serve as an end-to-end solution for researchers to execute their workflows, with JupyterHub serving as a powerful tool for user training and next-generation workforce development in atmospheric sciences.

@inproceedings{
 title = {ImageX 3.0: a full stack imaging archive solution},
 type = {inproceedings},
 year = {2018},
 pages = {46},
 month = {7},
 publisher = {SPIE-Intl Soc Optical Eng},
 day = {6},
 id = {fd7aed67-c949-3b19-8de5-42e15d1b0f0b},
 created = {2019-10-01T17:20:27.583Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2021-04-23T19:54:35.027Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Young2018},
 private_publication = {false},
 abstract = {Over the past several years we have faced the need to develop a number of solutions to address the challenge of archiving large-format scientific imaging data and seamlessly visualizing that data-irrespective of the image format-on a web browser. ImageX is a ground-up rewrite and synthesis of our solutions to this issue, with a goal of reducing the workload required to transition from simply storing vast amounts of scientific imaging data on disk to securely archiving and sharing that data with the world. The components that make up the ImageX service stack include a secure and scalable back-end data service optimized for providing imaging data, a pre-processor to harvest metadata and intelligently scale and store the imaging data, and a flexible and embeddable front-end visualization web application. Our latest version of the software suite called ImageX 3.0 has been designed to meet the needs of a single user running locally on their own personal computer or scaled up to provide support for the image storage and visualization needs of a modern observatory with the intention of providing a 'Push button' solution to a fully deployed solution. Each ImageX 3.0 component is provided as a Docker container, and can be rapidly and seamlessly deployed to meet demand. In this paper, we describe the ImageX architecture while demonstrating many of its features, including intelligent image scaling with adaptive histograms, load-balancing, and administrative tools. On the user-facing side we demonstrate how the ImageX 3.0 viewer can be embedded into the content of any web application, and explore the astronomy-specific features and plugins we've written into it. The ImageX service stack is fully open-sourced, and is built upon widely-supported industry standards (Node.js, Angular, etc.). Apart from being deployed as a standalone service stack, ImageX components are currently in use or expected to be deployed on: (1) the ODI-PPA portal serving astronomical images taken at the WIYN Observatory in near real-time; (2) the web portal serving microscopy images taken at the IU Electron Microscopy Center; (3) the RADY-SCA portal supporting radiology and medical imaging as well as neuroscience researchers at IU. © 2018 SPIE.},
 bibtype = {inproceedings},
 author = {Young, Michael D. and Gopu, Arvind and Perigo, Raymond},
 doi = {10.1117/12.2313684},
 booktitle = {SPIE ASTRONOMICAL TELESCOPES + INSTRUMENTATION 10-15 June 2018 Austin, Texas, United States}
}

Over the past several years we have faced the need to develop a number of solutions to address the challenge of archiving large-format scientific imaging data and seamlessly visualizing that data-irrespective of the image format-on a web browser. ImageX is a ground-up rewrite and synthesis of our solutions to this issue, with a goal of reducing the workload required to transition from simply storing vast amounts of scientific imaging data on disk to securely archiving and sharing that data with the world. The components that make up the ImageX service stack include a secure and scalable back-end data service optimized for providing imaging data, a pre-processor to harvest metadata and intelligently scale and store the imaging data, and a flexible and embeddable front-end visualization web application. Our latest version of the software suite called ImageX 3.0 has been designed to meet the needs of a single user running locally on their own personal computer or scaled up to provide support for the image storage and visualization needs of a modern observatory with the intention of providing a 'Push button' solution to a fully deployed solution. Each ImageX 3.0 component is provided as a Docker container, and can be rapidly and seamlessly deployed to meet demand. In this paper, we describe the ImageX architecture while demonstrating many of its features, including intelligent image scaling with adaptive histograms, load-balancing, and administrative tools. On the user-facing side we demonstrate how the ImageX 3.0 viewer can be embedded into the content of any web application, and explore the astronomy-specific features and plugins we've written into it. The ImageX service stack is fully open-sourced, and is built upon widely-supported industry standards (Node.js, Angular, etc.). Apart from being deployed as a standalone service stack, ImageX components are currently in use or expected to be deployed on: (1) the ODI-PPA portal serving astronomical images taken at the WIYN Observatory in near real-time; (2) the web portal serving microscopy images taken at the IU Electron Microscopy Center; (3) the RADY-SCA portal supporting radiology and medical imaging as well as neuroscience researchers at IU. © 2018 SPIE.

@inproceedings{
 title = {Scaling JupyterHub using Kubernetes on Jetstream cloud: Platform as a service for research and educational initiatives in the atmospheric sciences},
 type = {inproceedings},
 year = {2018},
 keywords = {Cloud Computing,JupyterHub,Kubernetes,Magnum,OpenStack,Unidata,Workforce Development},
 month = {7},
 publisher = {Association for Computing Machinery},
 day = {22},
 id = {a7659e9a-5510-30bf-a117-adb14fc82c8f},
 created = {2019-10-01T17:20:27.707Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.008Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Sarajlic2018},
 private_publication = {false},
 abstract = {Unidata, an NSF funded project that started in 1983, is a diverse community of education and research institutions with the common goal of sharing geoscience data and the tools to access and visualize that data. Unidata provides weather observations and other data, software tools, and support to enhance Earth-system education and research, and continuously examines ways of adapting their workflows for new technologies to maximize the reach of their education and research efforts.

In support of Unidata objectives to host workshops for atmospheric data analysis using JupyterHub, we explore a cloud computing approach leveraging Kubernetes coupled with JupyterHub that when combined will provide a solution for researchers and students to pull data from Unidata and burst onto Jetstream cloud by requesting resources dynamically via easy to use JupyterHub. More specifically, on Jetstream, Kubernetes is used for automating deployment and scaling of domain specific containerized applications, and JupyterHub is used for spawning multiple hubs within the same Kubernetes cluster instance that will be used for supporting classroom settings. JupyterHub's modular kernel feature will support dynamic needs of classroom application requirements. The proposed approach will serve as an end-to-end solution for researchers to execute their workflows, with JupyterHub serving as a powerful tool for user training and next-generation workforce development in atmospheric sciences.},
 bibtype = {inproceedings},
 author = {Sarajlic, Semir and Chastang, Julien and Marru, Suresh and Fischer, Jeremy and Lowe, Mike},
 doi = {10.1145/3219104.3229249},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

Unidata, an NSF funded project that started in 1983, is a diverse community of education and research institutions with the common goal of sharing geoscience data and the tools to access and visualize that data. Unidata provides weather observations and other data, software tools, and support to enhance Earth-system education and research, and continuously examines ways of adapting their workflows for new technologies to maximize the reach of their education and research efforts. In support of Unidata objectives to host workshops for atmospheric data analysis using JupyterHub, we explore a cloud computing approach leveraging Kubernetes coupled with JupyterHub that when combined will provide a solution for researchers and students to pull data from Unidata and burst onto Jetstream cloud by requesting resources dynamically via easy to use JupyterHub. More specifically, on Jetstream, Kubernetes is used for automating deployment and scaling of domain specific containerized applications, and JupyterHub is used for spawning multiple hubs within the same Kubernetes cluster instance that will be used for supporting classroom settings. JupyterHub's modular kernel feature will support dynamic needs of classroom application requirements. The proposed approach will serve as an end-to-end solution for researchers to execute their workflows, with JupyterHub serving as a powerful tool for user training and next-generation workforce development in atmospheric sciences.

@article{
 title = {Training children aged 5–10 years in compliance control: tracing smaller figures yields better learning not specific to the scale of drawn figures},
 type = {article},
 year = {2018},
 keywords = {Compliance control,Manual control,Motor development,Prospective control,Specificity},
 pages = {2589-2601},
 volume = {236},
 month = {10},
 publisher = {Springer Verlag},
 day = {1},
 id = {ab147de2-fdc2-377c-a242-5014535bda00},
 created = {2019-10-01T17:20:28.391Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.125Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Snapp-Childs2018},
 private_publication = {false},
 abstract = {© 2018 Springer-Verlag GmbH Germany, part of Springer Nature Previously we developed a method that supports active movement generation to allow practice with improvement of good compliance control in tracing and drawing. We showed that the method allowed children with motor impairments to improve at a 3D tracing task to become as proficient as typically developing children and that the training improved 2D figure copying. In this study, we expanded the training protocol to include a wider variety of ages (5–10-year-olds) and we made the figures traced in training the same as in figure copying, but varied the scale of training and copying figures to assess the generality of learning. Forty-eight children were assigned to groups trained using large or small figures. All were tested before training with a tracing task and a copying task. Then, the children trained over five sessions in the tracing task with either small or large figures. Finally, the tracing and copying tasks were tested again following training. A mean speed measure was used to control for path length variations in the timed task. Performance on both tasks at both baseline and posttest varied as a function of the size of the figure and age. In addition, tracing performance also varied with the level of support. In particular, speeds were higher with more support, larger figures and older children. After training, performance improved. Speeds increased. In tracing, performance improved more for large figures traced by children who trained on large figures. In copying, however, performance only improved significantly for children who had trained on small figures and it improved equally for large and small figures. In conclusion, training by tracing smaller figures yielded better learning that was not, however, specific to the scale of drawn figures. Small figures exhibit greater mean curvature. We infer that it yielded better general improvement.},
 bibtype = {article},
 author = {Snapp-Childs, Winona and Fath, Aaron J. and Bingham, Geoffrey P.},
 doi = {10.1007/s00221-018-5319-y},
 journal = {Experimental Brain Research},
 number = {10}
}

© 2018 Springer-Verlag GmbH Germany, part of Springer Nature Previously we developed a method that supports active movement generation to allow practice with improvement of good compliance control in tracing and drawing. We showed that the method allowed children with motor impairments to improve at a 3D tracing task to become as proficient as typically developing children and that the training improved 2D figure copying. In this study, we expanded the training protocol to include a wider variety of ages (5–10-year-olds) and we made the figures traced in training the same as in figure copying, but varied the scale of training and copying figures to assess the generality of learning. Forty-eight children were assigned to groups trained using large or small figures. All were tested before training with a tracing task and a copying task. Then, the children trained over five sessions in the tracing task with either small or large figures. Finally, the tracing and copying tasks were tested again following training. A mean speed measure was used to control for path length variations in the timed task. Performance on both tasks at both baseline and posttest varied as a function of the size of the figure and age. In addition, tracing performance also varied with the level of support. In particular, speeds were higher with more support, larger figures and older children. After training, performance improved. Speeds increased. In tracing, performance improved more for large figures traced by children who trained on large figures. In copying, however, performance only improved significantly for children who had trained on small figures and it improved equally for large and small figures. In conclusion, training by tracing smaller figures yielded better learning that was not, however, specific to the scale of drawn figures. Small figures exhibit greater mean curvature. We infer that it yielded better general improvement.

@inproceedings{
 title = {IQ-stations: Advances in state-of-the-art low cost immersive displays for research and development},
 type = {inproceedings},
 year = {2018},
 keywords = {Consumer hardware,IQ-Station,Tracking systems,Virtual reality},
 pages = {5},
 month = {7},
 publisher = {Association for Computing Machinery},
 day = {22},
 id = {06398430-4c2b-319c-bc41-b748bfff857b},
 created = {2019-10-01T17:20:30.200Z},
 accessed = {2019-08-27},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.190Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Sherman2018},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Sherman, William R. and Whiting, Eric and Money, James H. and Grover, Shane},
 doi = {10.1145/3219104.3219106},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

@article{
 title = {Limited mutation-rate variation within the Paramecium aurelia species complex},
 type = {article},
 year = {2018},
 keywords = {Ciliated protozoa mutationaccumulation,Neutral evolution},
 pages = {2523-2526},
 volume = {8},
 month = {7},
 publisher = {Genetics Society of America},
 day = {1},
 id = {e4392222-fcdf-3038-87f3-df208718c8ad},
 created = {2019-10-01T17:20:30.695Z},
 accessed = {2019-08-20},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.668Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Long2018},
 private_publication = {false},
 abstract = {© 2018 Long et al. Mutation is one of the most fundamental evolutionary forces. Studying variation in the mutation rate within and among closely-related species can help reveal mechanisms of genome divergence, but such variation is unstudied in the vast majority of organisms. Previous studies on ciliated protozoa have found extremely low mutation rates. In this study, using mutation-accumulation techniques combined with deep whole-genome sequencing, we explore the germline base-substitution mutation-rate variation of three cryptic species in the Paramecium aurelia species complex-P. biaurelia, P. sexaurelia, and P. tetraurelia. We find that there is extremely limited variation of the mutation rate and spectrum in the three species and confirm the extremely low mutation rate of ciliates.},
 bibtype = {article},
 author = {Long, Hongan and Doak, Thomas G. and Lynch, Michael},
 doi = {10.1534/g3.118.200420},
 journal = {G3: Genes, Genomes, Genetics},
 number = {7}
}

© 2018 Long et al. Mutation is one of the most fundamental evolutionary forces. Studying variation in the mutation rate within and among closely-related species can help reveal mechanisms of genome divergence, but such variation is unstudied in the vast majority of organisms. Previous studies on ciliated protozoa have found extremely low mutation rates. In this study, using mutation-accumulation techniques combined with deep whole-genome sequencing, we explore the germline base-substitution mutation-rate variation of three cryptic species in the Paramecium aurelia species complex-P. biaurelia, P. sexaurelia, and P. tetraurelia. We find that there is extremely limited variation of the mutation rate and spectrum in the three species and confirm the extremely low mutation rate of ciliates.

@article{
 title = {Insights into an Extensively Fragmented Eukaryotic Genome: De Novo Genome Sequencing of the Multinuclear Ciliate Uroleptopsis citrina},
 type = {article},
 year = {2018},
 pages = {883-894},
 volume = {10},
 publisher = {Oxford University Press},
 id = {4527e4a4-3cdc-3263-af11-a7f1fbede99b},
 created = {2019-10-01T17:20:31.089Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.686Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zheng2018},
 source_type = {JOUR},
 private_publication = {false},
 abstract = {Ciliated protists are a large group of single-celled eukaryotes with separate germline and somatic nuclei in each cell. The somatic genome is developed from the zygotic nucleus through a series of chromosomal rearrangements, including fragmentation, DNA elimination, de novo telomere addition, and DNA amplification. This unique feature makes them perfect models for research in genome biology and evolution. However, genomic research of ciliates has been limited to a few species, owing to problems with DNA contamination and obstacles in cultivation. Here, we introduce a method combining telomere-primer PCR amplification and high-throughput sequencing, which can reduce DNA contamination and obtain genomic data efficiently. Based on this method, we report a draft somatic genome of a multimacronuclear ciliate, Uroleptopsis citrina. 1) The telomeric sequence in U. citrina is confirmed to be C4A4C4A4C4 by directly blunt-end cloning. 2) Genomic analysis of the resulting chromosomes shows a “one-gene one-chromosome” pattern, with a small number of multiple-gene chromosomes. 3) Amino acid usage is analyzed, and reassignment of stop codons is confirmed. 4) Chromosomal analysis shows an obvious asymmetrical GC skew and high bias between A and T in the subtelomeric regions of the sense-strand, with the detection of an 11-bp high AT motif region in the 3′ subtelomeric region. 5) The subtelomeric sequence also has an obvious 40 nt strand oscillation of nucleotide ratio. 6) In the 5′ subtelomeric region of the coding strand, the distribution of potential TATA-box regions is illustrated, which accumulate between 30 and 50 nt. This work provides a valuable reference for genomic research and furthers our understanding of the dynamic nature of unicellular eukaryotic genomes.},
 bibtype = {article},
 author = {Zheng, Weibo and Wang, Chundi and Yan, Ying and Gao, Feng and Doak, Thomas G and Song, Weibo},
 doi = {https://doi.org/10.1093/gbe/evy055},
 journal = {Genome Biology and Evolution},
 number = {3}
}

Ciliated protists are a large group of single-celled eukaryotes with separate germline and somatic nuclei in each cell. The somatic genome is developed from the zygotic nucleus through a series of chromosomal rearrangements, including fragmentation, DNA elimination, de novo telomere addition, and DNA amplification. This unique feature makes them perfect models for research in genome biology and evolution. However, genomic research of ciliates has been limited to a few species, owing to problems with DNA contamination and obstacles in cultivation. Here, we introduce a method combining telomere-primer PCR amplification and high-throughput sequencing, which can reduce DNA contamination and obtain genomic data efficiently. Based on this method, we report a draft somatic genome of a multimacronuclear ciliate, Uroleptopsis citrina. 1) The telomeric sequence in U. citrina is confirmed to be C4A4C4A4C4 by directly blunt-end cloning. 2) Genomic analysis of the resulting chromosomes shows a “one-gene one-chromosome” pattern, with a small number of multiple-gene chromosomes. 3) Amino acid usage is analyzed, and reassignment of stop codons is confirmed. 4) Chromosomal analysis shows an obvious asymmetrical GC skew and high bias between A and T in the subtelomeric regions of the sense-strand, with the detection of an 11-bp high AT motif region in the 3′ subtelomeric region. 5) The subtelomeric sequence also has an obvious 40 nt strand oscillation of nucleotide ratio. 6) In the 5′ subtelomeric region of the coding strand, the distribution of potential TATA-box regions is illustrated, which accumulate between 30 and 50 nt. This work provides a valuable reference for genomic research and furthers our understanding of the dynamic nature of unicellular eukaryotic genomes.

@article{
 title = {Stilbenoid prenyltransferases define key steps in the diversification of peanut phytoalexins.},
 type = {article},
 year = {2018},
 keywords = {Arachidin,arachidin,hairy root,peanut,plant biochemistry,prenylation,resveratrol,secondary metabolism,small molecule,stilbenoid,transcriptomics},
 pages = {28-46},
 volume = {293},
 websites = {http://www.ncbi.nlm.nih.gov/pubmed/29158266,http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC5766904},
 id = {00ebce8a-ca57-333c-baec-dd957b280018},
 created = {2019-10-01T17:20:31.445Z},
 accessed = {2019-08-27},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.835Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Yang2018},
 private_publication = {false},
 abstract = {Defense responses of peanut (Arachis hypogaea) to biotic and abiotic stresses include the synthesis of prenylated stilbenoids. Members of this compound class show several protective activities in human disease studies, and the list of potential therapeutic targets continues to expand. Despite their medical and biological importance, the biosynthetic pathways of prenylated stilbenoids remain to be elucidated, and the genes encoding stilbenoid-specific prenyltransferases have yet to be identified in any plant species. In this study, we combined targeted transcriptomic and metabolomic analyses to discover prenyltransferase genes in elicitor-treated peanut hairy root cultures. Transcripts encoding five enzymes were identified, and two of these were functionally characterized in a transient expression system consisting of Agrobacterium-infiltrated leaves of Nicotiana benthamiana We observed that one of these prenyltransferases, AhR4DT-1, catalyzes a key reaction in the biosynthesis of prenylated stilbenoids, in which resveratrol is prenylated at its C-4 position to form arachidin-2, whereas another, AhR3'DT-1, added the prenyl group to C-3' of resveratrol. Each of these prenyltransferases was highly specific for stilbenoid substrates, and we confirmed their subcellular location in the plastid by fluorescence microscopy. Structural analysis of the prenylated stilbenoids suggested that these two prenyltransferase activities represent the first committed steps in the biosynthesis of a large number of prenylated stilbenoids and their derivatives in peanut. In summary, we have identified five candidate prenyltransferases in peanut and confirmed that two of them are stilbenoid-specific, advancing our understanding of this specialized enzyme family and shedding critical light onto the biosynthesis of bioactive stilbenoids.},
 bibtype = {article},
 author = {Yang, Tianhong and Fang, Lingling and Sanders, Sheri and Jayanthi, Srinivas and Rajan, Gayathri and Podicheti, Ram and Thallapuranam, Suresh Kumar and Mockaitis, Keithanne and Medina-Bolivar, Fabricio},
 doi = {10.1074/jbc.RA117.000564},
 journal = {The Journal of biological chemistry},
 number = {1}
}

Defense responses of peanut (Arachis hypogaea) to biotic and abiotic stresses include the synthesis of prenylated stilbenoids. Members of this compound class show several protective activities in human disease studies, and the list of potential therapeutic targets continues to expand. Despite their medical and biological importance, the biosynthetic pathways of prenylated stilbenoids remain to be elucidated, and the genes encoding stilbenoid-specific prenyltransferases have yet to be identified in any plant species. In this study, we combined targeted transcriptomic and metabolomic analyses to discover prenyltransferase genes in elicitor-treated peanut hairy root cultures. Transcripts encoding five enzymes were identified, and two of these were functionally characterized in a transient expression system consisting of Agrobacterium-infiltrated leaves of Nicotiana benthamiana We observed that one of these prenyltransferases, AhR4DT-1, catalyzes a key reaction in the biosynthesis of prenylated stilbenoids, in which resveratrol is prenylated at its C-4 position to form arachidin-2, whereas another, AhR3'DT-1, added the prenyl group to C-3' of resveratrol. Each of these prenyltransferases was highly specific for stilbenoid substrates, and we confirmed their subcellular location in the plastid by fluorescence microscopy. Structural analysis of the prenylated stilbenoids suggested that these two prenyltransferase activities represent the first committed steps in the biosynthesis of a large number of prenylated stilbenoids and their derivatives in peanut. In summary, we have identified five candidate prenyltransferases in peanut and confirmed that two of them are stilbenoid-specific, advancing our understanding of this specialized enzyme family and shedding critical light onto the biosynthesis of bioactive stilbenoids.

@article{
 title = {Escherichia coli cultures maintain stable subpopulation structure during long-term evolution},
 type = {article},
 year = {2018},
 keywords = {Article; bacterial colonization; bacterial gene; b},
 pages = {E4642-E4650},
 volume = {115},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85046974773&doi=10.1073%2Fpnas.1708371115&partnerID=40&md5=f03cd74612020bba927dfd16284a2ae6},
 publisher = {National Academy of Sciences},
 id = {5cfc23ba-c9dd-30e6-9bc3-522d0ffde292},
 created = {2019-10-01T17:20:32.089Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:32.089Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Behringer2018E4642},
 source_type = {article},
 notes = {cited By 0},
 private_publication = {false},
 abstract = {How genetic variation is generated and maintained remains a central question in evolutionary biology. When presented with a complex environment, microbes can take advantage of genetic variation to exploit new niches. Here we present a massively parallel experiment where WT and repair-deficient (ΔmutL) Escherichia coli populations have evolved over 3 y in a spatially heterogeneous and nutritionally complex environment. Metage-nomic sequencing revealed that these initially isogenic populations evolved and maintained stable subpopulation structure in just 10 mL of medium for up to 10,000 generations, consisting of up to five major haplotypes with many minor haplotypes. We characterized the genomic, transcriptomic, exometabolomic, and phenotypic differences between clonal isolates, revealing subpopulation structure driven primarily by spatial segregation followed by differential utilization of nutrients. In addition to genes regulating the import and catabolism of nutrients, major polymorphisms of note included insertion elements transposing into fimE (regulator of the type I fimbriae) and upstream of hns (global regulator of environmental-change and stress-response genes), both known to regulate biofilm formation. Interestingly, these genes have also been identified as critical to colonization in uro-pathogenic E. coli infections. Our findings illustrate the complexity that can arise and persist even in small cultures, raising the possibility that infections may often be promoted by an evolving and complex pathogen population. © 2018 National Academy of Sciences. All rights reserved.},
 bibtype = {article},
 author = {Behringer, M G and Choi, B I and Miller, S F and Doak, T G and Karty, J A and Guo, W and Lynch, M},
 doi = {10.1073/pnas.1708371115},
 journal = {Proceedings of the National Academy of Sciences of the United States of America},
 number = {20}
}

How genetic variation is generated and maintained remains a central question in evolutionary biology. When presented with a complex environment, microbes can take advantage of genetic variation to exploit new niches. Here we present a massively parallel experiment where WT and repair-deficient (ΔmutL) Escherichia coli populations have evolved over 3 y in a spatially heterogeneous and nutritionally complex environment. Metage-nomic sequencing revealed that these initially isogenic populations evolved and maintained stable subpopulation structure in just 10 mL of medium for up to 10,000 generations, consisting of up to five major haplotypes with many minor haplotypes. We characterized the genomic, transcriptomic, exometabolomic, and phenotypic differences between clonal isolates, revealing subpopulation structure driven primarily by spatial segregation followed by differential utilization of nutrients. In addition to genes regulating the import and catabolism of nutrients, major polymorphisms of note included insertion elements transposing into fimE (regulator of the type I fimbriae) and upstream of hns (global regulator of environmental-change and stress-response genes), both known to regulate biofilm formation. Interestingly, these genes have also been identified as critical to colonization in uro-pathogenic E. coli infections. Our findings illustrate the complexity that can arise and persist even in small cultures, raising the possibility that infections may often be promoted by an evolving and complex pathogen population. © 2018 National Academy of Sciences. All rights reserved.

@inproceedings{
 title = {Return on Investment for Three Cyberinfrastructure Facilities: A Local Campus Supercomputer, the NSF-Funded Jetstream Cloud System, and XSEDE (the eXtreme Science and Engineering Discovery Environment)},
 type = {inproceedings},
 year = {2018},
 keywords = {Cost benefit analysis,High performance computing,Scientific computing,Supercomputing},
 pages = {223-236},
 websites = {https://ieeexplore.ieee.org/document/8603169/},
 month = {12},
 publisher = {IEEE},
 day = {4},
 id = {1697f3d0-2337-3ee4-8856-d7a4d04fa5e3},
 created = {2019-10-01T17:20:35.064Z},
 accessed = {2019-08-14},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T18:58:47.660Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2018},
 private_publication = {false},
 abstract = {The economics of high performance computing are rapidly changing. Commercial cloud offerings, private research clouds, and pressure on the budgets of institutions of higher education and federally-funded research organizations are all contributing factors. As such, it has become a necessity that all expenses and investments be analyzed and considered carefully. In this paper we will analyze the return on investment (ROI) for three different kinds of cyberinfrastructure resources: the eXtreme Science and Engineering Discovery Environment (XSEDE); the NSF-funded Jetstream cloud system; and the Indiana University (IU) Big Red II supercomputer, funded exclusively by IU for use of the IU community and collaborators. We determined the ROI for these three resources by assigning financial values to services by either comparison with commercially available services, or by surveys of value of these resources to their users. In all three cases, the ROI for these very different types of cyberinfrastructure resources was well greater than 1-meaning that investors are getting more than $1 in returned value for every $1 invested. While there are many ways to measure the value and impact of investment in cyberinfrastructure resources, we are able to quantify the short-term ROI and show that it is a net positive for campuses and the federal government respectively.},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Hancock, David Y. and Wernert, Julie and Link, Matthew R. and Wilkins-Diehr, Nancy and Miller, Therese and Gaither, Kelly and Snapp-Childs, Winona},
 doi = {10.1109/UCC.2018.00031},
 booktitle = {2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)}
}

The economics of high performance computing are rapidly changing. Commercial cloud offerings, private research clouds, and pressure on the budgets of institutions of higher education and federally-funded research organizations are all contributing factors. As such, it has become a necessity that all expenses and investments be analyzed and considered carefully. In this paper we will analyze the return on investment (ROI) for three different kinds of cyberinfrastructure resources: the eXtreme Science and Engineering Discovery Environment (XSEDE); the NSF-funded Jetstream cloud system; and the Indiana University (IU) Big Red II supercomputer, funded exclusively by IU for use of the IU community and collaborators. We determined the ROI for these three resources by assigning financial values to services by either comparison with commercially available services, or by surveys of value of these resources to their users. In all three cases, the ROI for these very different types of cyberinfrastructure resources was well greater than 1-meaning that investors are getting more than $1 in returned value for every $1 invested. While there are many ways to measure the value and impact of investment in cyberinfrastructure resources, we are able to quantify the short-term ROI and show that it is a net positive for campuses and the federal government respectively.

@inproceedings{
 title = {High Availability on Jetstream: Practices and Lessons Learned},
 type = {inproceedings},
 year = {2018},
 keywords = {Atmosphere,XSEDE,acm reference format,atmosphere,availability,cloud,george turner,hpc,jeremy fischer,john michael lowe,research,sanjana sudarshan,xsede},
 pages = {4:1--4:7},
 websites = {http://doi.acm.org/10.1145/3217880.3217884},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {ScienceCloud'18},
 id = {5d4797f1-4581-3845-971d-edfa984ebcc2},
 created = {2019-10-01T17:20:35.320Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:35.320Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lowe:2018:HAJ:3217880.3217884},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {Research computing has traditionally used high performance computing (HPC) clusters and has been a service not given to high availability without a doubling of computational and storage capacity. System maintenance such as security patching, firmware updates, and other system upgrades generally meant that the system would be unavailable for the duration of the work unless one has redundant HPC systems and storage. While efforts were often made to limit downtimes, when it became necessary, maintenance windows might be one to two hours or as much as an entire day. As the National Science Foundation (NSF) began funding non-traditional research systems, looking at ways to provide higher availability for researchers became one focus for service providers. One of the design elements of Jetstream was to have geographic dispersion to maximize availability. This was the first step in a number of design elements intended to make Jetstream exceed the NSF's availability requirements. We will examine the design steps employed, the components of the system and how the availability for each was considered in deployment, how maintenance is handled, and the lessons learned from the design and implementation of the Jetstream cloud.},
 bibtype = {inproceedings},
 author = {Lowe, John Michael and Fischer, Jeremy and Sudarshan, Sanjana and Turner, George and Stewart, Craig A and Hancock, David Y},
 doi = {10.1145/3217880.3217884},
 booktitle = {Proceedings of the 9th Workshop on Scientific Cloud Computing (ScienceCloud'18)}
}

Research computing has traditionally used high performance computing (HPC) clusters and has been a service not given to high availability without a doubling of computational and storage capacity. System maintenance such as security patching, firmware updates, and other system upgrades generally meant that the system would be unavailable for the duration of the work unless one has redundant HPC systems and storage. While efforts were often made to limit downtimes, when it became necessary, maintenance windows might be one to two hours or as much as an entire day. As the National Science Foundation (NSF) began funding non-traditional research systems, looking at ways to provide higher availability for researchers became one focus for service providers. One of the design elements of Jetstream was to have geographic dispersion to maximize availability. This was the first step in a number of design elements intended to make Jetstream exceed the NSF's availability requirements. We will examine the design steps employed, the components of the system and how the availability for each was considered in deployment, how maintenance is handled, and the lessons learned from the design and implementation of the Jetstream cloud.

@inproceedings{
 title = {Methodologies and Practices for Adoption of a Novel National Research Environment},
 type = {inproceedings},
 year = {2018},
 keywords = {XSEDE,cloud,hpc,research},
 pages = {21:1--21:7},
 websites = {http://doi.acm.org/10.1145/3219104.3219115},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {e1170dc2-8029-3f4d-a796-d6e2b44f8a45},
 created = {2019-10-01T17:20:38.579Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:38.579Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fischer:2018:MPA:3219104.3219115},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {There are numerous domains of science that have been using high performance computing (HPC) systems for decades. Historically, when new HPC resources are introduced, specific variations may require researchers to make minor adjustments to their workflows but the general usage and expectations remain much the same. This consistency means that domain scientists can generally move from system to system as necessary and as new resources come online, they can be fairly easily adopted by these researchers. However, as novel resources, such as cloud computing systems, become available, additional work may be required in order to help researchers find and use the resource. When the goal of a system's funding and deployment is to find non-traditional research groups that have been under-served by the national cyberinfrastructure, a different approach to system adoption and training is required. When Jetstream was funded by the NSF as the first production research cloud, it became clear that to attract non-traditional or under-served researchers, a very proactive approach would be required. Here we show how the Jetstream team 1) developed methods and practices for increasing awareness of the system to both traditional HPC users as well as under-served and non-traditional users of HPC systems, 2) developed training approaches which highlight the capabilities that a cloud system may offer that are different from traditional HPC systems. We also discuss areas of success and failure, and plans for future efforts.},
 bibtype = {inproceedings},
 author = {Fischer, Jeremy and Beck, Brian W and Sudarshan, Sanjana and Turner, George and Snapp-Childs, Winona and Stewart, Craig A and Hancock, David Y},
 doi = {10.1145/3219104.3219115},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

There are numerous domains of science that have been using high performance computing (HPC) systems for decades. Historically, when new HPC resources are introduced, specific variations may require researchers to make minor adjustments to their workflows but the general usage and expectations remain much the same. This consistency means that domain scientists can generally move from system to system as necessary and as new resources come online, they can be fairly easily adopted by these researchers. However, as novel resources, such as cloud computing systems, become available, additional work may be required in order to help researchers find and use the resource. When the goal of a system's funding and deployment is to find non-traditional research groups that have been under-served by the national cyberinfrastructure, a different approach to system adoption and training is required. When Jetstream was funded by the NSF as the first production research cloud, it became clear that to attract non-traditional or under-served researchers, a very proactive approach would be required. Here we show how the Jetstream team 1) developed methods and practices for increasing awareness of the system to both traditional HPC users as well as under-served and non-traditional users of HPC systems, 2) developed training approaches which highlight the capabilities that a cloud system may offer that are different from traditional HPC systems. We also discuss areas of success and failure, and plans for future efforts.

@article{
 title = {XD Metrics on Demand Value Analytics: Visualizing the Impact of Internal Information Technology Investments on External Funding, Publications, and Collaboration Networks},
 type = {article},
 year = {2018},
 volume = {2},
 websites = {http://journal.frontiersin.org/article/10.3389/frma.2017.00010/full},
 month = {1},
 publisher = {Frontiers Media SA},
 day = {29},
 id = {251d5ed4-c9ec-381d-b9b4-8db9f0f35c1c},
 created = {2019-10-01T17:20:40.312Z},
 accessed = {2019-08-26},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.198Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Scrivner2018},
 private_publication = {false},
 abstract = {Many universities invest substantial resources in the design, deployment, and maintenance of campus-based cyberinfrastructure. To justify the expense, it is important that university administrators and others understand and communicate the value of these internal investments in terms of scholarly impact as measured by external funding, publications, and research collaborations. This paper introduces two visualizations and their usage in the Value Analytics (VA) module for Open XD Metrics on Demand (XDMoD). The VA module was developed by Indiana University’s (IU) Research Technologies division in conjunction with IU’s Cyberinfrastructure for Network Science Center (CNS) and the University at Buffalo’s Center for Computational Research (CCR). It interrelates quantitative measures of information technology (IT) usage, external funding, and publications in support of IT strategic decision making. This paper details the data, analysis workflows, and visual mappings used in the two VA visualizations that aim to communicate the value of different IT usage in terms of NSF and NIH funding, resulting publications, and associated research collaborations. To illustrate the feasibility of measuring IT values on research, we measured its financial and academic impact from the period between 2012 and 2017. The financial return on investment (ROI) is measured in terms of the funding, totaling $ 21,016,055 for NIH and NSF projects, and the academic ROI constitutes 1,531 NIH and NSF awards and 968 publications associated with 83 NSF and NIH awards. In addition, the results show that Medical Specialties, Brain Research, and Infectious Diseases are the top three scientific disciplines ranked by their publication records during the given time period.},
 bibtype = {article},
 author = {Scrivner, Olga and Singh, Gagandeep and Bouchard, Sara E. and Hutcheson, Scott C. and Fulton, Ben and Link, Matthew R. and Börner, Katy},
 doi = {10.3389/frma.2017.00010},
 journal = {Frontiers in Research Metrics and Analytics}
}

Many universities invest substantial resources in the design, deployment, and maintenance of campus-based cyberinfrastructure. To justify the expense, it is important that university administrators and others understand and communicate the value of these internal investments in terms of scholarly impact as measured by external funding, publications, and research collaborations. This paper introduces two visualizations and their usage in the Value Analytics (VA) module for Open XD Metrics on Demand (XDMoD). The VA module was developed by Indiana University’s (IU) Research Technologies division in conjunction with IU’s Cyberinfrastructure for Network Science Center (CNS) and the University at Buffalo’s Center for Computational Research (CCR). It interrelates quantitative measures of information technology (IT) usage, external funding, and publications in support of IT strategic decision making. This paper details the data, analysis workflows, and visual mappings used in the two VA visualizations that aim to communicate the value of different IT usage in terms of NSF and NIH funding, resulting publications, and associated research collaborations. To illustrate the feasibility of measuring IT values on research, we measured its financial and academic impact from the period between 2012 and 2017. The financial return on investment (ROI) is measured in terms of the funding, totaling $ 21,016,055 for NIH and NSF projects, and the academic ROI constitutes 1,531 NIH and NSF awards and 968 publications associated with 83 NSF and NIH awards. In addition, the results show that Medical Specialties, Brain Research, and Infectious Diseases are the top three scientific disciplines ranked by their publication records during the given time period.

@inproceedings{
 title = {High Performance Photogrammetry for Academic Research},
 type = {inproceedings},
 year = {2018},
 keywords = {HPC,acm reference format,benchmarking,distributed processing,evaluation,hpc,performance,performance evaluation,photogrammetry,scalability},
 pages = {45:1--45:8},
 websites = {http://doi.acm.org/10.1145/3219104.3219148},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {df2a4d8e-14f7-349e-bed8-b41cda3e7d85},
 created = {2019-10-01T17:20:40.900Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:40.900Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ruan:2018:HPP:3219104.3219148},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {Photogrammetry is the process of computationally extracting a three-dimensional surface model from a set of two-dimensional photographs of an object or environment. It is used to build models of everything from terrains to statues to ancient artifacts. In the past, the computational process was done on powerful PCs and could take weeks for large datasets. Even relatively small objects often required many hours of compute time to stitch together. With the availability of parallel processing options in the latest release of state-of-the-art photogrammetry software, it is possible to leverage the power of high performance computing systems on large datasets. In this paper we present a particular implementation of a high performance photogrammetry service. Though the service is currently based on a specific software package (Agisoft's PhotoScan), our system architecture is designed around a general photogrammetry process that can be easily adapted to leverage other photogrammetry tools. In addition, we report on an extensive performance study that measured the relative impacts of dataset size, software quality settings, and processing cluster size. Furthermore, we share lessons learned that are useful to system administrators looking to establish a similar service, and we describe the user-facing support components that are crucial for the success of the service.},
 bibtype = {inproceedings},
 author = {Ruan, Guangchen and Wernert, Eric and Sherman, William and Gniady, Tassie and Tuna, Esen and Sherman, William},
 doi = {10.1145/3219104.3219148},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

Photogrammetry is the process of computationally extracting a three-dimensional surface model from a set of two-dimensional photographs of an object or environment. It is used to build models of everything from terrains to statues to ancient artifacts. In the past, the computational process was done on powerful PCs and could take weeks for large datasets. Even relatively small objects often required many hours of compute time to stitch together. With the availability of parallel processing options in the latest release of state-of-the-art photogrammetry software, it is possible to leverage the power of high performance computing systems on large datasets. In this paper we present a particular implementation of a high performance photogrammetry service. Though the service is currently based on a specific software package (Agisoft's PhotoScan), our system architecture is designed around a general photogrammetry process that can be easily adapted to leverage other photogrammetry tools. In addition, we report on an extensive performance study that measured the relative impacts of dataset size, software quality settings, and processing cluster size. Furthermore, we share lessons learned that are useful to system administrators looking to establish a similar service, and we describe the user-facing support components that are crucial for the success of the service.

@inproceedings{
 title = {A Computational Notebook Approach to Large-scale Text Analysis},
 type = {inproceedings},
 year = {2018},
 keywords = {HPC,Spark,computational notebook,interactive analysis,scalability,text analysis},
 pages = {1-8},
 websites = {http://doi.acm.org/10.1145/3219104.3219153,http://dl.acm.org/citation.cfm?doid=3219104.3219153},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 series = {PEARC '18},
 id = {eaa29482-e732-3483-96e7-45c85711cd2e},
 created = {2019-10-01T17:20:43.081Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:43.081Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ruan:2018:CNA:3219104.3219153},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {Large-scale text analysis algorithms are important to many fields as they interrogate reams of textual data to extract evidence, correlations, and trends not readily discoverable by a human reader. Unfortunately, there is often an expertise mismatch between computational researchers who have the technical and programming skills necessary to develop workflows at scale and domain scholars who have knowledge of the literary, historical, scientific, or social factors that can affect data as it is manipulated. Our work focuses on the use of scalable computational notebooks as a model to bridge the accessibility gap for domain scholars, putting the power of HPC resources directly in the hands of the researchers who have scholarly questions. The computational notebook approach offers many benefits, including: fine-grained control through modularized functions, interactive analysis that puts the "human in the loop", scalable analysis that leverages Spark-as-a-Service, and complexity hiding interfaces that minimize the need for HPC expertise. In addition, the notebook approach makes it easy to share, reproduce, and sustain research workflows. We illustrate the applicability of our approach with usage scenarios on HPC systems as well as within a restricted computing environment to access sensitive, in-copyright data, and demonstrate the usefulness of the notebook approach with three examples from three different domains and data sources. These sources include historical topic trends in ten thousand scientific articles, sentiment analysis of tweets, and literary analysis of the copyrighted works of Kurt Vonnegut using non-consumptive techniques.},
 bibtype = {inproceedings},
 author = {Ruan, Guangchen and Gniady, Tassie and Kloster, David and Wernert, Eric and Tuna, Esen},
 doi = {10.1145/3219104.3219153},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Large-scale text analysis algorithms are important to many fields as they interrogate reams of textual data to extract evidence, correlations, and trends not readily discoverable by a human reader. Unfortunately, there is often an expertise mismatch between computational researchers who have the technical and programming skills necessary to develop workflows at scale and domain scholars who have knowledge of the literary, historical, scientific, or social factors that can affect data as it is manipulated. Our work focuses on the use of scalable computational notebooks as a model to bridge the accessibility gap for domain scholars, putting the power of HPC resources directly in the hands of the researchers who have scholarly questions. The computational notebook approach offers many benefits, including: fine-grained control through modularized functions, interactive analysis that puts the "human in the loop", scalable analysis that leverages Spark-as-a-Service, and complexity hiding interfaces that minimize the need for HPC expertise. In addition, the notebook approach makes it easy to share, reproduce, and sustain research workflows. We illustrate the applicability of our approach with usage scenarios on HPC systems as well as within a restricted computing environment to access sensitive, in-copyright data, and demonstrate the usefulness of the notebook approach with three examples from three different domains and data sources. These sources include historical topic trends in ten thousand scientific articles, sentiment analysis of tweets, and literary analysis of the copyrighted works of Kurt Vonnegut using non-consumptive techniques.

@inproceedings{
 title = {Toward sustainable deployment of distributed services on the cloud: dockerized ODI-PPA on Jetstream},
 type = {inproceedings},
 year = {2018},
 pages = {108},
 month = {7},
 publisher = {SPIE-Intl Soc Optical Eng},
 day = {6},
 id = {7ac34819-1928-3a63-90f7-91b70849f28e},
 created = {2019-10-01T17:20:43.947Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.053Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Bao2018},
 private_publication = {false},
 abstract = {The One Degree Imager - Portal, Pipeline and Archive (ODI-PPA) - a mature and fully developed product - has been a workhorse for astronomers observing on the WIYN ODI. It not only provides access to data stored in a secure archive, it also has a rich search and visualization interface, as well as integrated pipeline capabilities connected with supercomputers at Indiana University in a manner transparent to the user. As part of our ongoing sustainability review process, and given the increasing age of the ODI-PPA codebase, we have considered various approaches to modernization. While industry currently trends toward Node.js based architectures, we concluded that porting an entire legacy PHP and Python-based system like ODI-PPA with its complex and distributed service stack would require too significant an amount of human development/testing/deployment hours. Aging deployment hardware with tight budgets is another issue we identified, a common one especially when deploying complex distributed service stacks. In this paper, we present DockStream (https://jsportal.odi.iu.edu), an elegant solution that addresses both of the aforementioned issues. Using ODI-PPA as a case study, we present a proof of concept solution combining a suite of Docker containers built for each PPA service and a mechanism to acquire cost-free computational and storage resources. The dockerized ODI-PPA services can be deployed on one Dockerenabled host or several depending on the availability of hardware resources and the expected levels of use. In this paper, we describe the process of designing, creating, and deploying such custom containers. The NSF-funded Jetstream led by the Indiana University Pervasive Technology Institute (PTI), provides cloud-based, on-demand computing and data analysis resources, and a pathway to tackle the issue of insufficient hardware refreshment funds. We briefly describe the process to acquiring computational and storage resources on Jetstream, and the use of the Atmosphere web interface to create and maintain virtual machines on Jetstream. Finally, we present a summary of security refinements to a dockerized service stack on the cloud using nginx, custom docker networks, and Linux firewalls that significant decrease the risk of security vulnerabilities and incidents while improving scalability.},
 bibtype = {inproceedings},
 author = {Bao, Yuanzhi and Gopu, Arvind and Perigo, Raymond and Young, Michael D.},
 doi = {10.1117/12.2313647},
 booktitle = {SPIE ASTRONOMICAL TELESCOPES + INSTRUMENTATION 10-15 June 2018 Austin, Texas, United States}
}

The One Degree Imager - Portal, Pipeline and Archive (ODI-PPA) - a mature and fully developed product - has been a workhorse for astronomers observing on the WIYN ODI. It not only provides access to data stored in a secure archive, it also has a rich search and visualization interface, as well as integrated pipeline capabilities connected with supercomputers at Indiana University in a manner transparent to the user. As part of our ongoing sustainability review process, and given the increasing age of the ODI-PPA codebase, we have considered various approaches to modernization. While industry currently trends toward Node.js based architectures, we concluded that porting an entire legacy PHP and Python-based system like ODI-PPA with its complex and distributed service stack would require too significant an amount of human development/testing/deployment hours. Aging deployment hardware with tight budgets is another issue we identified, a common one especially when deploying complex distributed service stacks. In this paper, we present DockStream (https://jsportal.odi.iu.edu), an elegant solution that addresses both of the aforementioned issues. Using ODI-PPA as a case study, we present a proof of concept solution combining a suite of Docker containers built for each PPA service and a mechanism to acquire cost-free computational and storage resources. The dockerized ODI-PPA services can be deployed on one Dockerenabled host or several depending on the availability of hardware resources and the expected levels of use. In this paper, we describe the process of designing, creating, and deploying such custom containers. The NSF-funded Jetstream led by the Indiana University Pervasive Technology Institute (PTI), provides cloud-based, on-demand computing and data analysis resources, and a pathway to tackle the issue of insufficient hardware refreshment funds. We briefly describe the process to acquiring computational and storage resources on Jetstream, and the use of the Atmosphere web interface to create and maintain virtual machines on Jetstream. Finally, we present a summary of security refinements to a dockerized service stack on the cloud using nginx, custom docker networks, and Linux firewalls that significant decrease the risk of security vulnerabilities and incidents while improving scalability.

@inproceedings{
 title = {Data Capsule Appliance for Restricted Data in Libraries},
 type = {inproceedings},
 year = {2018},
 websites = {https://www.tacc.utexas.edu/documents/1084364/1627230/06_Plale-DC-IMLS-CMD18.pdf/792ac021-b8b8-432d-aceb-4a8d8c9a6dac},
 city = {Fort Worth, TX},
 id = {5110758a-7762-3ad0-b481-57371ba7784a},
 created = {2019-10-01T17:20:46.224Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:46.224Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Withana2018},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Withana, Sachith and Kouper, Inna and Plale, Beth A.},
 booktitle = {Workshop on Cyberinfrastructure and Machine Learning for Digital Libraries and Archives, in conjunction with Joint Conference on Digital Libraries 2018}
}

@inproceedings{
 title = {Subject headings and beyond: Mapping the HathiTrust Digital Library content for wider use},
 type = {inproceedings},
 year = {2018},
 websites = {https://osf.io/ak9u8/},
 city = {Berkeley, CA},
 id = {46cd618a-8ea7-34fe-a19d-961d9255b34b},
 created = {2019-10-01T17:20:51.873Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:51.873Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Edelblute2018},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Edelblute, Trevor and Zoss, Angela and Kouper, Inna},
 booktitle = {HathiTrust Research Center UnCamp 2018}
}

@article{
 title = {Rice Galaxy: an open resource for plant science},
 type = {article},
 year = {2018},
 pages = {358754},
 publisher = {Cold Spring Harbor Laboratory},
 id = {256d5614-4bfa-37b2-9be3-569521c3e09f},
 created = {2019-10-01T17:20:52.123Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.942Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Juanillas2018},
 source_type = {JOUR},
 private_publication = {false},
 bibtype = {article},
 author = {Juanillas, Venice Margarette J and Dereeper, Alexis and Beaume, Nicolas and Droc, Gaetan and Dizon, Joshua and Mendoza, John Robert and Perdon, Jon Peter and Mansueto, Locedie and Triplett, Lindsay and Lang, Jillian},
 journal = {bioRxiv}
}

@inproceedings{
 title = {Narrative visualization},
 type = {inproceedings},
 year = {2018},
 websites = {https://www.researchgate.net/publication/325271444_Narrative_Visualization},
 city = {Ames, Iowa},
 id = {0e727d8b-c226-36eb-9e8a-4b3b120c6755},
 created = {2019-10-01T17:20:52.574Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:52.574Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kouper2018},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Kouper, Inna},
 booktitle = {2018 Midwest Big Data Summer School}
}

@inproceedings{
 title = {Restricted data types used in secure computing environments},
 type = {inproceedings},
 year = {2018},
 id = {253bf04b-6c9c-34d4-8349-b362f2b472cf},
 created = {2019-10-01T17:20:52.807Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:52.807Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kouper2018a},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Kouper, Inna and Mitchell, Erik},
 booktitle = {HathiTrust Research Center UnCamp 2018}
}

@article{
 title = {Harp-DAAL for High Performance Big Data Computing},
 type = {article},
 year = {2018},
 websites = {https://software.intel.com/sites/default/files/parallel-universe-issue-32.pdf},
 month = {3},
 id = {362b4d16-bf65-3c60-a6d4-aacdd96ef897},
 created = {2019-10-01T17:20:53.992Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.906Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Qiu2018},
 private_publication = {false},
 abstract = {Large-scale data analytics is revolutionizing many business and scientific domains. Easy-to-use scalable parallel techniques are necessary to process big data and gain meaningful insights. We introduce a novel HPC-Cloud convergence framework named Harp-DAAL and demonstrate that the combination of Big Data (Hadoop) and HPC techniques can simultaneously achieve productivity and performance. Harp is a distributed Hadoop-based framework that orchestrates efficient node synchronization [1]. Harp uses Intel ® Data Analytics Accelerator Library (DAAL) [2], for its highly optimized kernels on Intel ® Xeon and Xeon Phi architectures. This way the high-level API of Big Data tools can be combined with intra-node fine-grained parallelism that is optimized for HPC platforms. We illustrate this framework in detail with K-means clustering, a computation-bounded algorithm used in image clustering. We also show the broad applicability of Harp-DAAL by discussing the performance of three other big data algorithms: Subgraph Counting by color coding, Matrix Factorization and Latent Dirichlet Allocation. They share issues such as load imbalance, irregular structure, and communication issues that create difficult challenges. Figure 1 Cloud-HPC interoperable software for High Performance Big Data Analytics at Scale The categories in Figure 1 illustrate a classification of data intensive computation into five computation models that map into five distinct system architectures. It starts with Sequential, followed by centralized batch architectures corresponding exactly to the three forms of MapReduce: Map-Only, MapReduce and Iterative MapReduce. Category five is the classic MPI model. Harp brings Hadoop users the benefits of supporting all 5 classes of data-intensive computation, from pleasingly parallel to machine learning and simulations. We have expanded the applicability of Hadoop (with Harp plugin) for more classes of Big Data applications, especially complex data analytics such as machine learning and graph. We redesign a modular software stack with native kernels (with DAAL) to effectively utilize scale-up servers for machine learning and data analytics applications. Harp-DAAL shows how simulations and Big Data can use common programming environments with a runtime based on a rich set of collectives and libraries.},
 bibtype = {article},
 author = {Qiu, Judy},
 journal = {Parallel Universe}
}

Large-scale data analytics is revolutionizing many business and scientific domains. Easy-to-use scalable parallel techniques are necessary to process big data and gain meaningful insights. We introduce a novel HPC-Cloud convergence framework named Harp-DAAL and demonstrate that the combination of Big Data (Hadoop) and HPC techniques can simultaneously achieve productivity and performance. Harp is a distributed Hadoop-based framework that orchestrates efficient node synchronization [1]. Harp uses Intel ® Data Analytics Accelerator Library (DAAL) [2], for its highly optimized kernels on Intel ® Xeon and Xeon Phi architectures. This way the high-level API of Big Data tools can be combined with intra-node fine-grained parallelism that is optimized for HPC platforms. We illustrate this framework in detail with K-means clustering, a computation-bounded algorithm used in image clustering. We also show the broad applicability of Harp-DAAL by discussing the performance of three other big data algorithms: Subgraph Counting by color coding, Matrix Factorization and Latent Dirichlet Allocation. They share issues such as load imbalance, irregular structure, and communication issues that create difficult challenges. Figure 1 Cloud-HPC interoperable software for High Performance Big Data Analytics at Scale The categories in Figure 1 illustrate a classification of data intensive computation into five computation models that map into five distinct system architectures. It starts with Sequential, followed by centralized batch architectures corresponding exactly to the three forms of MapReduce: Map-Only, MapReduce and Iterative MapReduce. Category five is the classic MPI model. Harp brings Hadoop users the benefits of supporting all 5 classes of data-intensive computation, from pleasingly parallel to machine learning and simulations. We have expanded the applicability of Hadoop (with Harp plugin) for more classes of Big Data applications, especially complex data analytics such as machine learning and graph. We redesign a modular software stack with native kernels (with DAAL) to effectively utilize scale-up servers for machine learning and data analytics applications. Harp-DAAL shows how simulations and Big Data can use common programming environments with a runtime based on a rich set of collectives and libraries.

@article{
 title = {Structure of pion photoproduction amplitudes},
 type = {article},
 year = {2018},
 keywords = {doi:10.1103/PhysRevD.98.014041 url:https://doi.org},
 pages = {014041},
 volume = {98},
 websites = {https://link.aps.org/doi/10.1103/PhysRevD.98.014041},
 publisher = {American Physical Society},
 id = {b8e6deae-84b9-342a-8fd9-f25d1b4f9a7e},
 created = {2019-10-01T17:20:53.992Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.833Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Mathieu2018a},
 private_publication = {false},
 bibtype = {article},
 author = {Mathieu, V. and Nys, J. and Fernández-Ramírez, C. and Blin, A. N. Hiller and Jackura, A. and Pilloni, A. and Szczepaniak, A. P. and Fox, G.},
 doi = {10.1103/PhysRevD.98.014041},
 journal = {Physical Review D},
 number = {1}
}

@techreport{
 title = {Machine Learning for Parameter Auto-tuning in Molecular Dynamics Simulations: Efficient Dynamics of Ions near Polarizable Nanoparticles},
 type = {techreport},
 year = {2018},
 keywords = {Auto-tuning,Energy Minimization,Hybrid MPI/OpenMP,Machine Learning,Nanoscale Simulations,Parallel Computing},
 pages = {15},
 websites = {www.sagepub.com/,http://dsc.soic.indiana.edu/publications/Manuscript.IJHPCA.Nov2018.pdf},
 id = {93e36f16-de3c-38ac-8569-20194aa0e1dd},
 created = {2019-10-01T17:20:54.076Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.762Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kadupitiya2018},
 private_publication = {false},
 abstract = {Simulating the dynamics of ions near polarizable nanoparticles (NPs) is extremely challenging due to the need to solve the Poisson equation at every simulation timestep. Recently, a molecular dynamics (MD) method based on a dynamical optimization framework bypassed this obstacle by representing the polarization charge density as virtual dynamic variables, and evolving them in parallel with the physical dynamics of ions. We highlight the computational gains accessible with the integration of machine learning (ML) methods for parameter prediction in MD simulations by demonstrating how they were realized in MD simulations of ions near polarizable NPs. An artificial neural network based regression model was integrated with MD and predicted the optimal simulation timestep and critical parameters characterizing the virtual system on-the-fly with 94.3% success. The integration of ML method with hybrid OpenMP/MPI parallelized MD simulations generated accurate dynamics of thousands of ions in the presence of polarizable NPs for over 10 million steps (with a maximum simulated physical time over 30 ns) while reducing the computational time from thousands of hours to tens of hours yielding a maximum speedup of ≈ 3 from ML-only acceleration and a maximum overall speedup of ≈ 600 from ML-hybrid Open/MPI combined method. Extraction of ionic structure in concentrated electrolytes near oil-water emulsions demonstrates the success of the method. The approach can be generalized to select optimal parameters in other molecular dynamics applications and energy minimization problems.},
 bibtype = {techreport},
 author = {Kadupitiya, Jcs and Fox, Geoffrey C and Jadhao, Vikram},
 doi = {10.1177/ToBeAssigned}
}

Simulating the dynamics of ions near polarizable nanoparticles (NPs) is extremely challenging due to the need to solve the Poisson equation at every simulation timestep. Recently, a molecular dynamics (MD) method based on a dynamical optimization framework bypassed this obstacle by representing the polarization charge density as virtual dynamic variables, and evolving them in parallel with the physical dynamics of ions. We highlight the computational gains accessible with the integration of machine learning (ML) methods for parameter prediction in MD simulations by demonstrating how they were realized in MD simulations of ions near polarizable NPs. An artificial neural network based regression model was integrated with MD and predicted the optimal simulation timestep and critical parameters characterizing the virtual system on-the-fly with 94.3% success. The integration of ML method with hybrid OpenMP/MPI parallelized MD simulations generated accurate dynamics of thousands of ions in the presence of polarizable NPs for over 10 million steps (with a maximum simulated physical time over 30 ns) while reducing the computational time from thousands of hours to tens of hours yielding a maximum speedup of ≈ 3 from ML-only acceleration and a maximum overall speedup of ≈ 600 from ML-hybrid Open/MPI combined method. Extraction of ionic structure in concentrated electrolytes near oil-water emulsions demonstrates the success of the method. The approach can be generalized to select optimal parameters in other molecular dynamics applications and energy minimization problems.

@article{
 title = {Vector meson photoproduction with a linearly polarized beam},
 type = {article},
 year = {2018},
 volume = {97},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85049067436&doi=10.1103%2FPhysRevD.97.094003&partnerID=40&md5=c3e8c7a8d17dcaf2f3e6e356802ef69c},
 publisher = {American Physical Society},
 id = {fa9c47f0-8641-384e-bdcd-40a289fea8f3},
 created = {2019-10-01T17:20:54.176Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:54.176Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Mathieu2018},
 source_type = {article},
 notes = {cited By 0},
 private_publication = {false},
 abstract = {We propose a model based on Regge theory to describe photoproduction of light vector mesons. We fit the SLAC data and make predictions for the energy and momentum-transfer dependence of the spin-density matrix elements in photoproduction of ω, ρ0 and φ mesons at Eγ∼8.5 GeV, which are soon to be measured at Jefferson Lab. © 2018 authors. Published by the American Physical Society. Published by the American Physical Society under the terms of the »https://creativecommons.org/licenses/by/4.0/» Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI. Funded by SCOAP3.},
 bibtype = {article},
 author = {Mathieu, V and Nys, J and Fernández-Ramírez, C and Jackura, A and Pilloni, A and Sherrill, N and Szczepaniak, A P and Fox, G},
 doi = {10.1103/PhysRevD.97.094003},
 journal = {Physical Review D},
 number = {9}
}

We propose a model based on Regge theory to describe photoproduction of light vector mesons. We fit the SLAC data and make predictions for the energy and momentum-transfer dependence of the spin-density matrix elements in photoproduction of ω, ρ0 and φ mesons at Eγ∼8.5 GeV, which are soon to be measured at Jefferson Lab. © 2018 authors. Published by the American Physical Society. Published by the American Physical Society under the terms of the »https://creativecommons.org/licenses/by/4.0/» Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI. Funded by SCOAP3.

@article{
 title = {Crossover analysis and automated layer-tracking assessment of the extracted DEM of the basal topography of the canadian arctic archipelago ice-cap},
 type = {article},
 year = {2018},
 keywords = {DEM,SAR,Synthetic aperture radar imaging,ice,ice-bottom tracking,tomography},
 pages = {862-867},
 id = {16f65691-dd6e-33eb-ab4a-7fc604924405},
 created = {2019-10-01T17:20:54.553Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.251Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Al-Ibadi2018},
 private_publication = {false},
 abstract = {© 2018 IEEE. In 2014, as part of the NASA Operation IceBridge project, the Center for Remote Sensing of Ice Sheets operated a multi-beam synthetic aperture radar depth sounder/imager over the Canadian Arctic Archipelago (CAA) to generate digital elevation models (DEMs) of the glacial basal topography. In this work, we briefly describe the processing steps that led to the generation of these DEMs, algorithm improvements over previously published results, and assess the results from two different perspectives. First, we evaluate the self-consistency of the DEMs where flight paths cross over each other and two measurements are made at the same location. Secondly, we compare the quality of the outputs of the ice-bottom tracker before and after applying manual corrections to the tracker results; the tracker is an algorithm that we implemented to automatically track the ice-bottom. Even though the CAA ice-caps are mountainous areas, where the scenes often have ice and no ice regions, which makes the imaging complicated, the statistical results show good tracking performance and a good match between the overlapped DEMs, where the mean error of the crossover DEMs is 37±9 m.},
 bibtype = {article},
 author = {Al-Ibadi, Mohanad and Sprick, Jordan and Athinarapu, Sravya and Berger, Victor and Stumpf, Theresa and Paden, John and Leuschen, Carl and Rodriguez, Fernando and Xu, Mingze and Crandall, David and Fox, Geoffrey and Burgess, David and Sharp, Martin and Copland, Luke and Van Wychen, Wesley},
 doi = {10.1109/RADAR.2018.8378673},
 journal = {2018 IEEE Radar Conference, RadarConf 2018}
}

© 2018 IEEE. In 2014, as part of the NASA Operation IceBridge project, the Center for Remote Sensing of Ice Sheets operated a multi-beam synthetic aperture radar depth sounder/imager over the Canadian Arctic Archipelago (CAA) to generate digital elevation models (DEMs) of the glacial basal topography. In this work, we briefly describe the processing steps that led to the generation of these DEMs, algorithm improvements over previously published results, and assess the results from two different perspectives. First, we evaluate the self-consistency of the DEMs where flight paths cross over each other and two measurements are made at the same location. Secondly, we compare the quality of the outputs of the ice-bottom tracker before and after applying manual corrections to the tracker results; the tracker is an algorithm that we implemented to automatically track the ice-bottom. Even though the CAA ice-caps are mountainous areas, where the scenes often have ice and no ice regions, which makes the imaging complicated, the statistical results show good tracking performance and a good match between the overlapped DEMs, where the mean error of the crossover DEMs is 37±9 m.

@inproceedings{
 title = {Performance Characterization of Multi-threaded Graph Processing Applications on Intel Many-Integrated-Core Architecture},
 type = {inproceedings},
 year = {2018},
 pages = {199–208},
 websites = {http://arxiv.org/abs/1708.04701},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 id = {b3933d22-d9f1-354a-b19d-b88c96edc1e5},
 created = {2019-10-01T17:20:54.687Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.021Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Liu2018},
 private_publication = {false},
 abstract = {Intel Xeon Phi many-integrated-core (MIC) architectures usher in a new era of terascale integration. Among emerging killer applications, parallel graph processing has been a critical technique to analyze connected data. In this paper, we empirically evaluate various computing platforms including an Intel Xeon E5 CPU, a Nvidia Geforce GTX1070 GPU and an Xeon Phi 7210 processor codenamed Knights Landing (KNL) in the domain of parallel graph processing. We show that the KNL gains encouraging performance when processing graphs, so that it can become a promising solution to accelerating multi-threaded graph applications. We further characterize the impact of KNL architectural enhancements on the performance of a state-of-the art graph framework.We have four key observations: 1 Different graph applications require distinctive numbers of threads to reach the peak performance. For the same application, various datasets need even different numbers of threads to achieve the best performance. 2 Only a few graph applications benefit from the high bandwidth MCDRAM, while others favor the low latency DDR4 DRAM. 3 Vector processing units executing AVX512 SIMD instructions on KNLs are underutilized when running the state-of-the-art graph framework. 4 The sub-NUMA cache clustering mode offering the lowest local memory access latency hurts the performance of graph benchmarks that are lack of NUMA awareness. At last, We suggest future works including system auto-tuning tools and graph framework optimizations to fully exploit the potential of KNL for parallel graph processing.},
 bibtype = {inproceedings},
 author = {Liu, Xu and Chen, Langshi and Firoz, Jesun S. and Qiu, Judy and Jiang, Lei},
 doi = {10.1109/ISPASS.2018.00033},
 booktitle = {2018 IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2018}
}

Intel Xeon Phi many-integrated-core (MIC) architectures usher in a new era of terascale integration. Among emerging killer applications, parallel graph processing has been a critical technique to analyze connected data. In this paper, we empirically evaluate various computing platforms including an Intel Xeon E5 CPU, a Nvidia Geforce GTX1070 GPU and an Xeon Phi 7210 processor codenamed Knights Landing (KNL) in the domain of parallel graph processing. We show that the KNL gains encouraging performance when processing graphs, so that it can become a promising solution to accelerating multi-threaded graph applications. We further characterize the impact of KNL architectural enhancements on the performance of a state-of-the art graph framework.We have four key observations: 1 Different graph applications require distinctive numbers of threads to reach the peak performance. For the same application, various datasets need even different numbers of threads to achieve the best performance. 2 Only a few graph applications benefit from the high bandwidth MCDRAM, while others favor the low latency DDR4 DRAM. 3 Vector processing units executing AVX512 SIMD instructions on KNLs are underutilized when running the state-of-the-art graph framework. 4 The sub-NUMA cache clustering mode offering the lowest local memory access latency hurts the performance of graph benchmarks that are lack of NUMA awareness. At last, We suggest future works including system auto-tuning tools and graph framework optimizations to fully exploit the potential of KNL for parallel graph processing.

@inproceedings{
 title = {Object Detection by a Super-Resolution Method and a Convolutional Neural Networks},
 type = {inproceedings},
 year = {2018},
 keywords = {CNN,convolution neural networks,deep learning,machine learning,object detection,super-resolution},
 pages = {2263-2269},
 month = {1},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {22},
 id = {452c5a40-913a-3376-9eea-672542e6d6d6},
 created = {2019-10-01T17:20:55.050Z},
 accessed = {2019-08-21},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.323Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Na2018},
 private_publication = {false},
 abstract = {Recently with many blurless or slightly blurred images, convolutional neural networks classify objects with around 90 percent classification rates, even if there are variable sized images. However, small object regions or cropping of images make object detection or classification difficult and decreases the detection rates. In many methods related to convolutional neural network (CNN), Bilinear or Bicubic algorithms are popularly used to interpolate region of interests. To overcome the limitations of these algorithms, we introduce a super-resolution method applied to the cropped regions or candidates, and this leads to improve recognition rates for object detection and classification. Large object candidates comparable in size of the full image have good results for object detections using many popular conventional methods. However, for smaller region candidates, using our super-resolution preprocessing and region candidates, allows a CNN to outperform conventional methods in the number of detected objects when tested on the VOC2007 and MSO datasets.},
 bibtype = {inproceedings},
 author = {Na, Bokyoon and Fox, Geoffrey C.},
 doi = {10.1109/BigData.2018.8622135},
 booktitle = {2018 IEEE International Conference on Big Data, Big Data}
}

Recently with many blurless or slightly blurred images, convolutional neural networks classify objects with around 90 percent classification rates, even if there are variable sized images. However, small object regions or cropping of images make object detection or classification difficult and decreases the detection rates. In many methods related to convolutional neural network (CNN), Bilinear or Bicubic algorithms are popularly used to interpolate region of interests. To overcome the limitations of these algorithms, we introduce a super-resolution method applied to the cropped regions or candidates, and this leads to improve recognition rates for object detection and classification. Large object candidates comparable in size of the full image have good results for object detections using many popular conventional methods. However, for smaller region candidates, using our super-resolution preprocessing and region candidates, allows a CNN to outperform conventional methods in the number of detected objects when tested on the VOC2007 and MSO datasets.

@inproceedings{
 title = {Evaluating the scientific impact of XSEDE},
 type = {inproceedings},
 year = {2018},
 keywords = {Bibliometrics,H-index,Scientific impact,Technology Audit Service,XDMoD,XSEDE},
 month = {7},
 publisher = {Association for Computing Machinery},
 day = {22},
 id = {20b72ad4-d5bb-362b-93c0-b0800fbf4f77},
 created = {2019-10-01T17:20:55.350Z},
 accessed = {2019-09-03},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.270Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Wang2018},
 private_publication = {false},
 abstract = {We use the bibliometrics approach to evaluate the scientific impact of XSEDE. By utilizing publication data from various sources, e.g., ISI Web of Science and Microsoft Academic Graph, we calculate the impact metrics of XSEDE publications and show how they compare with non-XSEDE publication from the same field of study, or non-XSEDE peers from the same journal issue. We explain the dataset and data soruces involved and how we retrieved, cleaned, and curated millions of related publication entries. We then introduce the metrics we used for evaluation and comparison, and the methods used to calculate them. Detailed analysis results of Field Weighted Citation Impact (FWCI) and the peers comparison will be presented and discussed. We also explain how the same approaches could be used to evaluate publications from a similar organization or institute, to demonstrate the general applicability of the present evaluation approach providing impact even beyond XSEDE.},
 bibtype = {inproceedings},
 author = {Wang, Fugang and Fox, Geoffrey C. and Von Laszewski, Gregor and Furlani, Thomas R. and Gallo, Steven M. and Whitson, Timothy and DeLeon, Robert L.},
 doi = {10.1145/3219104.3219124},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

We use the bibliometrics approach to evaluate the scientific impact of XSEDE. By utilizing publication data from various sources, e.g., ISI Web of Science and Microsoft Academic Graph, we calculate the impact metrics of XSEDE publications and show how they compare with non-XSEDE publication from the same field of study, or non-XSEDE peers from the same journal issue. We explain the dataset and data soruces involved and how we retrieved, cleaned, and curated millions of related publication entries. We then introduce the metrics we used for evaluation and comparison, and the methods used to calculate them. Detailed analysis results of Field Weighted Citation Impact (FWCI) and the peers comparison will be presented and discussed. We also explain how the same approaches could be used to evaluate publications from a similar organization or institute, to demonstrate the general applicability of the present evaluation approach providing impact even beyond XSEDE.

@inproceedings{
 title = {Evaluation of Production Serverless Computing Environments},
 type = {inproceedings},
 year = {2018},
 keywords = {Amazon Lambda,Event-driven Computing,FaaS,Google Functions,IBM OpenWhisk,Microsoft Azure Functions,Serverless},
 pages = {442-450},
 volume = {2018-July},
 month = {9},
 publisher = {IEEE Computer Society},
 day = {7},
 id = {24abffda-00d9-3409-b1ce-c6f2eb1f4a07},
 created = {2019-10-01T17:20:55.549Z},
 accessed = {2019-09-04},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.536Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lee2018a},
 private_publication = {false},
 abstract = {Serverless computing provides a small runtime container to execute lines of codes without infrastructure management which is similar to Platform as a Service (PaaS) but a functional level. Amazon started the event-driven compute named Lambda functions in 2014 with a 25 concurrent limitation, but it now supports at least a thousand of concurrent invocation to process event messages generated by resources like databases, storage and system logs. Other providers, i.e., Google, Microsoft, and IBM offer a dynamic scaling manager to handle parallel requests of stateless functions in which additional containers are provisioning on new compute nodes for distribution. However, while functions are often developed for microservices and lightweight workload, they are associated with distributed data processing using the concurrent invocations. We claim that the current serverless computing environments can support dynamic applications in parallel when a partitioned task is executable on a small function instance. We present results of throughput, network bandwidth, a file I/O and compute performance regarding the concurrent invocations. We deployed a series of functions for distributed data processing to address the elasticity and then demonstrated the differences between serverless computing and virtual machines for cost efficiency and resource utilization.},
 bibtype = {inproceedings},
 author = {Lee, Hyungro and Satyam, Kumar and Fox, Geoffrey},
 doi = {10.1109/CLOUD.2018.00062},
 booktitle = {IEEE International Conference on Cloud Computing, CLOUD}
}

Serverless computing provides a small runtime container to execute lines of codes without infrastructure management which is similar to Platform as a Service (PaaS) but a functional level. Amazon started the event-driven compute named Lambda functions in 2014 with a 25 concurrent limitation, but it now supports at least a thousand of concurrent invocation to process event messages generated by resources like databases, storage and system logs. Other providers, i.e., Google, Microsoft, and IBM offer a dynamic scaling manager to handle parallel requests of stateless functions in which additional containers are provisioning on new compute nodes for distribution. However, while functions are often developed for microservices and lightweight workload, they are associated with distributed data processing using the concurrent invocations. We claim that the current serverless computing environments can support dynamic applications in parallel when a partitioned task is executable on a small function instance. We present results of throughput, network bandwidth, a file I/O and compute performance regarding the concurrent invocations. We deployed a series of functions for distributed data processing to address the elasticity and then demonstrated the differences between serverless computing and virtual machines for cost efficiency and resource utilization.

@inproceedings{
 title = {Automated tracking of 2D and 3D ice radar imagery using Viterbi and TRW-S},
 type = {inproceedings},
 year = {2018},
 keywords = {Glaciology,Ice thickness,Ice-bottom tracking,Image classification,Radar tomography},
 pages = {4162-4165},
 volume = {2018-July},
 month = {10},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {31},
 id = {5011258d-5477-30ff-9625-616395a58f6e},
 created = {2019-10-01T17:20:55.671Z},
 accessed = {2019-09-04},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.744Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Berger2018},
 private_publication = {false},
 abstract = {We present improvements to existing implementations of the Viterbi and TRW-S algorithms applied to ice-bottom layer tracking on 2D and 3D radar imagery, respectively. Along with an explanation of our modifications and the reasoning behind them, we present a comparison between our results, the results obtained with the original implementations, and those obtained with other proposed methods of performing ice-bottom layer tracking.},
 bibtype = {inproceedings},
 author = {Berger, Victor and Xu, Mingze and Chu, Shane and Crandall, David and Paden, John and Fox, Geoffrey C.},
 doi = {10.1109/IGARSS.2018.8519411},
 booktitle = {International Geoscience and Remote Sensing Symposium (IGARSS)}
}

We present improvements to existing implementations of the Viterbi and TRW-S algorithms applied to ice-bottom layer tracking on 2D and 3D radar imagery, respectively. Along with an explanation of our modifications and the reasoning behind them, we present a comparison between our results, the results obtained with the original implementations, and those obtained with other proposed methods of performing ice-bottom layer tracking.

@article{
 title = {Finding and counting tree-like subgraphs using MapReduce},
 type = {article},
 year = {2018},
 pages = {217-230},
 volume = {4},
 websites = {https://ieeexplore.ieee.org/document/8090537/},
 month = {7},
 day = {1},
 id = {1abad963-1212-3047-a2d7-667a53ae393e},
 created = {2019-10-01T17:20:57.186Z},
 accessed = {2019-08-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.716Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhao2018},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {IEEE Several variants of the subgraph isomorphism problem, e.g., finding, counting and estimating frequencies of subgraphs in networks arise in a number of real world applications, such as web analysis, disease diffusion prediction and social network analysis. These problems are computationally challenging in having to scale to very large networks with millions of vertices. In this paper, we present SAHAD, a MapReduce algorithm for detecting and counting trees of bounded size using the elegant color coding technique developed by N. Alon et al. SAHAD is a randomized algorithm, and we show rigorous bounds on the approximation quality and the performance of it. SAHAD scales to very large networks comprising of < formula > < tex > $10^7$ < /tex > < /formula > - < formula > < tex > $10^8$ < /tex > < /formula > edges and tree-like (acyclic) templates with up to 12 vertices. Further, we extend our results by implementing SAHAD in the Harp framework, which is more of a high performance computing environment. The new implementation gives 100x improvement in performance over the standard Hadoop implementation and achieves better performance than state-of-the-art MPI solutions on larger graphs.},
 bibtype = {article},
 author = {Zhao, Zhao and Chen, Langshi and Avram, Mihai and Li, Meng and Wang, Guanying and Butt, Ali and Khan, Maleq and Marathe, Madhav and Qiu, Judy and Vullikanti, Anil},
 doi = {10.1109/TMSCS.2017.2768426},
 journal = {IEEE Transactions on Multi-Scale Computing Systems},
 number = {3}
}

IEEE Several variants of the subgraph isomorphism problem, e.g., finding, counting and estimating frequencies of subgraphs in networks arise in a number of real world applications, such as web analysis, disease diffusion prediction and social network analysis. These problems are computationally challenging in having to scale to very large networks with millions of vertices. In this paper, we present SAHAD, a MapReduce algorithm for detecting and counting trees of bounded size using the elegant color coding technique developed by N. Alon et al. SAHAD is a randomized algorithm, and we show rigorous bounds on the approximation quality and the performance of it. SAHAD scales to very large networks comprising of < formula > < tex > $10^7$ < /tex > < /formula > - < formula > < tex > $10^8$ < /tex > < /formula > edges and tree-like (acyclic) templates with up to 12 vertices. Further, we extend our results by implementing SAHAD in the Harp framework, which is more of a high performance computing environment. The new implementation gives 100x improvement in performance over the standard Hadoop implementation and achieves better performance than state-of-the-art MPI solutions on larger graphs.

@inproceedings{
 title = {Deep hybrid wavelet network for ice boundary detection in radra imagery},
 type = {inproceedings},
 year = {2018},
 keywords = {Deep learning,Holistically nested edge detection,Ice Boundary detection,Radar,Wavelet transform},
 pages = {3449-3452},
 volume = {2018-July},
 month = {10},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {31},
 id = {93918f87-4ab8-37b7-8607-f8e885d01bc9},
 created = {2019-10-01T17:20:57.298Z},
 accessed = {2019-09-03},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.568Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kamangir2018},
 private_publication = {false},
 abstract = {This paper proposes a deep convolutional neural network approach to detect ice surface and bottom layers from radar imagery. Radar images are capable to penetrate the ice surface and provide us with valuable information from the underlying layers of the ice surface. In recent years, deep hierarchical learning techniques for object detection and segmentation greatly improved the performance of traditional techniques based on hand-crafted feature engineering. We designed a deep convo-lutional network to produce the images of the surface and bottom ice boundary. Our network takes advantage of undecimated wavelet transform to provide the highest level of information from radar images, as well as multilayer and multi-scale optimized architecture. In this work, radar images from 2009-2016 NASA Operation IceBridge Mission are used to train and test the network. Our network outperformed the state-of-the art accuracy.},
 bibtype = {inproceedings},
 author = {Kamangir, Hamid and Rahnemoonfar, Maryam and Dobbs, Dugan and Paden, John and Fox, Geoffrey},
 doi = {10.1109/IGARSS.2018.8518617},
 booktitle = {International Geoscience and Remote Sensing Symposium (IGARSS)}
}

This paper proposes a deep convolutional neural network approach to detect ice surface and bottom layers from radar imagery. Radar images are capable to penetrate the ice surface and provide us with valuable information from the underlying layers of the ice surface. In recent years, deep hierarchical learning techniques for object detection and segmentation greatly improved the performance of traditional techniques based on hand-crafted feature engineering. We designed a deep convo-lutional network to produce the images of the surface and bottom ice boundary. Our network takes advantage of undecimated wavelet transform to provide the highest level of information from radar images, as well as multilayer and multi-scale optimized architecture. In this work, radar images from 2009-2016 NASA Operation IceBridge Mission are used to train and test the network. Our network outperformed the state-of-the art accuracy.

@techreport{
 title = {Big Data Benchmarks on Bare Metal Cloud},
 type = {techreport},
 year = {2018},
 pages = {1-5},
 id = {f1245ec0-41d2-354a-a60b-557a174f28ee},
 created = {2019-10-01T17:21:00.989Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.664Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Lee2018},
 private_publication = {false},
 abstract = {High performance computing requires to deal with a large number of applications running on different environments, and bare metal cloud is promising to enable new hardware features but in a easier way than traditional HPC systems. Data that we need to deal with is growing exponentially although many big data software support processing them at scale. We perform big data benchmark on public bare metal clouds to demonstrate computing performance with direct hardware access and block storage using up to 25000 and 32000 IOPS respectively for Oracle and Amazon. The preliminary results indicate that Amazon and Oracle are competitive supporting high throughput and low latency with operations in parallel. We investigate further on storage options available on Oracle bare metal with different data sets and anticipate to evaluate petabyte-scale workloads on cluster configurations in the future.},
 bibtype = {techreport},
 author = {Lee, Hyungro and Fox, Geoffrey C},
 doi = {10.13140/RG.2.2.12204.36486}
}

High performance computing requires to deal with a large number of applications running on different environments, and bare metal cloud is promising to enable new hardware features but in a easier way than traditional HPC systems. Data that we need to deal with is growing exponentially although many big data software support processing them at scale. We perform big data benchmark on public bare metal clouds to demonstrate computing performance with direct hardware access and block storage using up to 25000 and 32000 IOPS respectively for Oracle and Amazon. The preliminary results indicate that Amazon and Oracle are competitive supporting high throughput and low latency with operations in parallel. We investigate further on storage options available on Oracle bare metal with different data sets and anticipate to evaluate petabyte-scale workloads on cluster configurations in the future.

@book{
 title = {Big data and extreme-scale computing: Pathways to Convergence-Toward a shaping strategy for a future software and data ecosystem for scientific inquiry},
 type = {book},
 year = {2018},
 source = {International Journal of High Performance Computing Applications},
 keywords = {Big data,extreme-scale computing,future software,high-end data analysis,traditional HPC},
 pages = {435-479},
 volume = {32},
 issue = {4},
 id = {5dffcc0c-bc3c-3cd7-b2bf-2cfcc04edd31},
 created = {2019-10-01T17:21:01.192Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.903Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Asch2018},
 private_publication = {false},
 abstract = {© The Author(s) 2018. Over the past four years, the Big Data and Exascale Computing (BDEC) project organized a series of five international workshops that aimed to explore the ways in which the new forms of data-centric discovery introduced by the ongoing revolution in high-end data analysis (HDA) might be integrated with the established, simulation-centric paradigm of the high-performance computing (HPC) community. Based on those meetings, we argue that the rapid proliferation of digital data generators, the unprecedented growth in the volume and diversity of the data they generate, and the intense evolution of the methods for analyzing and using that data are radically reshaping the landscape of scientific computing. The most critical problems involve the logistics of wide-area, multistage workflows that will move back and forth across the computing continuum, between the multitude of distributed sensors, instruments and other devices at the networks edge, and the centralized resources of commercial clouds and HPC centers. We suggest that the prospects for the future integration of technological infrastructures and research ecosystems need to be considered at three different levels. First, we discuss the convergence of research applications and workflows that establish a research paradigm that combines both HPC and HDA, where ongoing progress is already motivating efforts at the other two levels. Second, we offer an account of some of the problems involved with creating a converged infrastructure for peripheral environments, that is, a shared infrastructure that can be deployed throughout the network in a scalable manner to meet the highly diverse requirements for processing, communication, and buffering/storage of massive data workflows of many different scientific domains. Third, we focus on some opportunities for software ecosystem convergence in big, logically centralized facilities that execute large-scale simulations and models and/or perform large-scale data analytics. We close by offering some conclusions and recommendations for future investment and policy review.},
 bibtype = {book},
 author = {Asch, M. and Moore, T. and Badia, R. and Beck, M. and Beckman, P. and Bidot, T. and Bodin, F. and Cappello, F. and Choudhary, A. and de Supinski, B. and Deelman, E. and Dongarra, J. and Dubey, A. and Fox, G. and Fu, H. and Girona, S. and Gropp, W. and Heroux, M. and Ishikawa, Y. and Keahey, K. and Keyes, D. and Kramer, W. and Lavignon, J. F. and Lu, Y. and Matsuoka, S. and Mohr, B. and Reed, D. and Requena, S. and Saltz, J. and Schulthess, T. and Stevens, R. and Swany, M. and Szalay, A. and Tang, W. and Varoquaux, G. and Vilotte, J. P. and Wisniewski, R. and Xu, Z. and Zacharov, I.},
 doi = {10.1177/1094342018778123}
}

© The Author(s) 2018. Over the past four years, the Big Data and Exascale Computing (BDEC) project organized a series of five international workshops that aimed to explore the ways in which the new forms of data-centric discovery introduced by the ongoing revolution in high-end data analysis (HDA) might be integrated with the established, simulation-centric paradigm of the high-performance computing (HPC) community. Based on those meetings, we argue that the rapid proliferation of digital data generators, the unprecedented growth in the volume and diversity of the data they generate, and the intense evolution of the methods for analyzing and using that data are radically reshaping the landscape of scientific computing. The most critical problems involve the logistics of wide-area, multistage workflows that will move back and forth across the computing continuum, between the multitude of distributed sensors, instruments and other devices at the networks edge, and the centralized resources of commercial clouds and HPC centers. We suggest that the prospects for the future integration of technological infrastructures and research ecosystems need to be considered at three different levels. First, we discuss the convergence of research applications and workflows that establish a research paradigm that combines both HPC and HDA, where ongoing progress is already motivating efforts at the other two levels. Second, we offer an account of some of the problems involved with creating a converged infrastructure for peripheral environments, that is, a shared infrastructure that can be deployed throughout the network in a scalable manner to meet the highly diverse requirements for processing, communication, and buffering/storage of massive data workflows of many different scientific domains. Third, we focus on some opportunities for software ecosystem convergence in big, logically centralized facilities that execute large-scale simulations and models and/or perform large-scale data analytics. We close by offering some conclusions and recommendations for future investment and policy review.

@inproceedings{
 title = {Task Scheduling in Big Data - Review, Research Challenges, and Prospects},
 type = {inproceedings},
 year = {2018},
 keywords = {Big Data,Dataflow,MapReduce,Static and Dynamic Task Scheduling,Task Scheduling Model,Twister2},
 pages = {165-173},
 month = {8},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {20},
 id = {1caf2685-0aa3-3c64-a879-97271e93d71d},
 created = {2019-10-01T17:21:01.239Z},
 accessed = {2019-09-04},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.443Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Govindarajan2018},
 private_publication = {false},
 abstract = {—In a Big data computing, the processing of data requires a large amount of CPU cycles and network bandwidth and disk I/O. Dataflow is a programming model for processing Big data which consists of tasks organized in a graph structure. Scheduling these tasks is one of the key active research areas which mainly aims to place the tasks on available resources. It is essential to effectively schedule the tasks, in a manner that minimizes task completion time and increases utilization of resources. In recent years, various researchers have discussed and presented different task scheduling algorithms. In this research study, we have investigated the state-of-art of various types of task scheduling algorithms, scheduling considerations for batch and streaming processing, and task scheduling algorithms in the well-known open-source big data platforms. Furthermore, this study proposes a new task scheduling system to alleviate the problems persists in the existing task scheduling for big data.},
 bibtype = {inproceedings},
 author = {Govindarajan, Kannan and Kamburugamuve, Supun and Wickramasinghe, Pulasthi and Abeykoon, Vibhatha and Fox, Geoffrey},
 doi = {10.1109/ICoAC.2017.8441494},
 booktitle = {2017 9th International Conference on Advanced Computing, ICoAC 2017}
}

—In a Big data computing, the processing of data requires a large amount of CPU cycles and network bandwidth and disk I/O. Dataflow is a programming model for processing Big data which consists of tasks organized in a graph structure. Scheduling these tasks is one of the key active research areas which mainly aims to place the tasks on available resources. It is essential to effectively schedule the tasks, in a manner that minimizes task completion time and increases utilization of resources. In recent years, various researchers have discussed and presented different task scheduling algorithms. In this research study, we have investigated the state-of-art of various types of task scheduling algorithms, scheduling considerations for batch and streaming processing, and task scheduling algorithms in the well-known open-source big data platforms. Furthermore, this study proposes a new task scheduling system to alleviate the problems persists in the existing task scheduling for big data.

@article{
 title = {Features of πΔ photoproduction at high energies},
 type = {article},
 year = {2018},
 pages = {77-81},
 volume = {779},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85044396630&doi=10.1016%2Fj.physletb.2018.01.075&partnerID=40&md5=12e6f3f9ea386dbf28749cd0713aa855},
 publisher = {Elsevier B.V.},
 id = {d8476aea-069c-3fcb-ae84-b03c75d5bbca},
 created = {2019-10-01T17:21:01.444Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:01.444Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Nys201877},
 source_type = {article},
 notes = {cited By 0},
 private_publication = {false},
 abstract = {Hybrid/exotic meson spectroscopy searches at Jefferson Lab require the accurate theoretical description of the production mechanism in peripheral photoproduction. We develop a model for πΔ photoproduction at high energies (5≤Elab≤16 GeV) that incorporates both the absorbed pion and natural-parity cut contributions. We fit the available observables, providing a good description of the energy and angular dependencies of the experimental data. We also provide predictions for the photon beam asymmetry of charged pions at Elab=9 GeV which is expected to be measured by GlueX and CLAS12 experiments in the near future. © 2018 The Author},
 bibtype = {article},
 author = {Nys, J and Mathieu, V and Fernández-Ramírez, C and Jackura, A and Mikhasenko, M and Pilloni, A and Sherrill, N and Ryckebusch, J and Szczepaniak, A P and Fox, Geoffrey Charles and Center, Joint Physics Analysis},
 doi = {10.1016/j.physletb.2018.01.075},
 journal = {Physics Letters, Section B: Nuclear, Elementary Particle and High-Energy Physics}
}

Hybrid/exotic meson spectroscopy searches at Jefferson Lab require the accurate theoretical description of the production mechanism in peripheral photoproduction. We develop a model for πΔ photoproduction at high energies (5≤Elab≤16 GeV) that incorporates both the absorbed pion and natural-parity cut contributions. We fit the available observables, providing a good description of the energy and angular dependencies of the experimental data. We also provide predictions for the photon beam asymmetry of charged pions at Elab=9 GeV which is expected to be measured by GlueX and CLAS12 experiments in the near future. © 2018 The Author

@article{
 title = {Analyticity Constraints for Hadron Amplitudes: Going High to Heal Low Energy Issues},
 type = {article},
 year = {2018},
 pages = {41001-p1-p5},
 volume = {122},
 websites = {http://arxiv.org/abs/1708.07779},
 id = {ab2c1e5a-8274-3bb1-adbc-3c5e90009426},
 created = {2019-10-01T17:21:01.590Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.147Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Mathieu2018b},
 private_publication = {false},
 abstract = {Analyticity constitutes a rigid constraint on hadron scattering amplitudes. This property is used to relate models in different energy regimes. Using meson photoproduction as a benchmark, we show how to test contemporary low energy models directly against high energy data. This method pinpoints deficiencies of the models and treads a path to further improvement. The implementation of this technique enables one to produce more stable and reliable partial waves for future use in hadron spectroscopy and new physics searches.},
 bibtype = {article},
 author = {Mathieu, V. and Nys, J. and Pilloni, A. and Fernández-Ramírez, C. and Jackura, A. and Mikhasenko, M. and Pauk, V. and Szczepaniak, A. P. and Fox, G.},
 journal = {Europhysics Letters},
 number = {4}
}

Analyticity constitutes a rigid constraint on hadron scattering amplitudes. This property is used to relate models in different energy regimes. Using meson photoproduction as a benchmark, we show how to test contemporary low energy models directly against high energy data. This method pinpoints deficiencies of the models and treads a path to further improvement. The implementation of this technique enables one to produce more stable and reliable partial waves for future use in hadron spectroscopy and new physics searches.

@techreport{
 title = {Contributions to High-Performance Big Data Computing},
 type = {techreport},
 year = {2018},
 keywords = {Big Data,Biomolecular simulations,Clouds,Graph Analytics,HPC,MIDAS,Network Science,Pathology,Polar Science,SPIDAL},
 websites = {http://dsc.soic.indiana.edu/publications/FormattedSPIDALPaperJune2019.pdf,https://www.researchgate.net/publication/328090399_Contributions_to_High-Performance_Big_Data_Computing},
 id = {18ff2b58-d289-3380-8a8c-61e4d8ece87b},
 created = {2019-10-01T17:21:01.803Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.348Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Fox2018},
 private_publication = {false},
 abstract = {Our project is at the interface of Big Data and HPC-High-Performance Big Data computing and this paper describes a collaboration between 7 collaborating Universities at Arizona State, Indiana (lead), Kansas, Rutgers, Stony Brook, Virginia Tech, and Utah. It addresses the intersection of High-performance and Big Data computing with several different application areas or communities driving the requirements for software systems and algorithms. We describe the base architecture, including the HPC-ABDS, High-Performance Computing enhanced Apache Big Data Stack, and an application use case study identifying key features that determine software and algorithm requirements. We summarize middleware including Harp-DAAL collective communication layer, Twister2 Big Data toolkit, and pilot jobs. Then we present the SPIDAL Scalable Parallel Interoperable Data Analytics Library and our work for it in core machine-learning, image processing and the application communities, Network science, Polar Science, Biomolecular Simulations, Pathology, and Spatial systems. We describe basic algorithms and their integration in end-to-end use cases.},
 bibtype = {techreport},
 author = {Fox, Geoffrey and Qiu, Judy and Crandall, David and Laszewski, Gregor Von and Beckstein, Oliver and Paden, John and Paraskevakos, Ioannis and Jha, Shantenu and Wang, Fusheng and Marathe, Madhav and Vullikanti, Anil and Cheatham, Thomas}
}

Our project is at the interface of Big Data and HPC-High-Performance Big Data computing and this paper describes a collaboration between 7 collaborating Universities at Arizona State, Indiana (lead), Kansas, Rutgers, Stony Brook, Virginia Tech, and Utah. It addresses the intersection of High-performance and Big Data computing with several different application areas or communities driving the requirements for software systems and algorithms. We describe the base architecture, including the HPC-ABDS, High-Performance Computing enhanced Apache Big Data Stack, and an application use case study identifying key features that determine software and algorithm requirements. We summarize middleware including Harp-DAAL collective communication layer, Twister2 Big Data toolkit, and pilot jobs. Then we present the SPIDAL Scalable Parallel Interoperable Data Analytics Library and our work for it in core machine-learning, image processing and the application communities, Network science, Polar Science, Biomolecular Simulations, Pathology, and Spatial systems. We describe basic algorithms and their integration in end-to-end use cases.

@techreport{
 title = {Detecting ice layers in radar images with deep learning},
 type = {techreport},
 year = {2018},
 pages = {2-5},
 issue = {April},
 websites = {https://pdfs.semanticscholar.org/e24d/e190e0e01e0e53003fa83daeb4557859f9f6.pdf},
 id = {fb72b8b1-5f54-3eb5-96b9-2e0835f2b601},
 created = {2019-10-01T17:21:01.973Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.071Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {HamidKamangirMaryamRahnemoonfarDuganDobbsJPaden2018},
 private_publication = {false},
 abstract = {This paper proposes a Deep Convolutional Neural Network approach to detect Ice surface and bottom layers from radar imagery. Radar images are capable to penetrate the earth surface and provide us with valuable information from the underlying layers of ice surface. In recent years, deep hierarchical learning techniques for object detection and segmentation greatly improved the performance of traditional techniques based on hand-crafted feature engineering. We designed a hybrid Deep Convolutional Network to produce the images of surface and bottom ice boundary as outputs. Our network takes advantage of undecimated wavelet transform to provide the highest level of information from radar images, as well as multi-layer and multi-scale optimized architecture. In this work, radar images from 2009-2016 NASA Operation IceBridge Mission are used to train and test the network. Our network outperformed the state-of-the art accuracy.},
 bibtype = {techreport},
 author = {Hamid Kamangir, Maryam Rahnemoonfar, Dugan Dobbs, J Paden, Geoffrey Fox}
}

This paper proposes a Deep Convolutional Neural Network approach to detect Ice surface and bottom layers from radar imagery. Radar images are capable to penetrate the earth surface and provide us with valuable information from the underlying layers of ice surface. In recent years, deep hierarchical learning techniques for object detection and segmentation greatly improved the performance of traditional techniques based on hand-crafted feature engineering. We designed a hybrid Deep Convolutional Network to produce the images of surface and bottom ice boundary as outputs. Our network takes advantage of undecimated wavelet transform to provide the highest level of information from radar images, as well as multi-layer and multi-scale optimized architecture. In this work, radar images from 2009-2016 NASA Operation IceBridge Mission are used to train and test the network. Our network outperformed the state-of-the art accuracy.

@inproceedings{
 title = {Multi-task Spatiotemporal Neural Networks for Structured Surface Reconstruction},
 type = {inproceedings},
 year = {2018},
 pages = {1273-1282},
 volume = {2018-Janua},
 month = {5},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {3},
 id = {8d825f5a-82a4-3345-885e-4258ee92181b},
 created = {2019-10-01T17:21:02.013Z},
 accessed = {2019-09-04},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.100Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Xu2018},
 private_publication = {false},
 abstract = {Deep learning methods have surpassed the performance of traditional techniques on a wide range of problems in computer vision, but nearly all of this work has studied consumer photos, where precisely correct output is often not critical. It is less clear how well these techniques may apply on structured prediction problems where fine-grained output with high precision is required, such as in scientific imaging domains. Here we consider the problem of segmenting echogram radar data collected from the polar ice sheets, which is challenging because segmentation boundaries are often very weak and there is a high degree of noise. We propose a multi-task spatiotemporal neural network that combines 3D ConvNets and Recurrent Neural Networks (RNNs) to estimate ice surface boundaries from sequences of tomographic radar images. We show that our model outperforms the state-of-the-art on this problem by (1) avoiding the need for hand-tuned parameters, (2) extracting multiple surfaces (ice-air and ice-bed) simultaneously, (3) requiring less non-visual metadata, and (4) being about 6 times faster.},
 bibtype = {inproceedings},
 author = {Xu, Mingze and Fan, Chenyou and Paden, John D. and Fox, Geoffrey C. and Crandall, David J.},
 doi = {10.1109/WACV.2018.00144},
 booktitle = {Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018}
}

Deep learning methods have surpassed the performance of traditional techniques on a wide range of problems in computer vision, but nearly all of this work has studied consumer photos, where precisely correct output is often not critical. It is less clear how well these techniques may apply on structured prediction problems where fine-grained output with high precision is required, such as in scientific imaging domains. Here we consider the problem of segmenting echogram radar data collected from the polar ice sheets, which is challenging because segmentation boundaries are often very weak and there is a high degree of noise. We propose a multi-task spatiotemporal neural network that combines 3D ConvNets and Recurrent Neural Networks (RNNs) to estimate ice surface boundaries from sequences of tomographic radar images. We show that our model outperforms the state-of-the-art on this problem by (1) avoiding the need for hand-tuned parameters, (2) extracting multiple surfaces (ice-air and ice-bed) simultaneously, (3) requiring less non-visual metadata, and (4) being about 6 times faster.

@inproceedings{
 title = {Twister: Net - Communication Library for Big Data Processing in HPC and Cloud Environments},
 type = {inproceedings},
 year = {2018},
 keywords = {Big-data,Collectives,HPC,MPI,Streaming},
 pages = {383-391},
 volume = {2018-July},
 month = {9},
 publisher = {IEEE Computer Society},
 day = {7},
 id = {5805ab79-5f24-3379-ac55-473415607664},
 created = {2019-10-01T17:21:02.072Z},
 accessed = {2019-09-04},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.126Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kamburugamuve2018},
 private_publication = {false},
 abstract = {Streaming processing and batch data processing are the dominant forms of big data analytics today, with numerous systems such as Hadoop, Spark, and Heron designed to process the ever-increasing explosion of data. Generally, these systems are developed as single projects with aspects such as communication, task management, and data management integrated together. By contrast, we take a component-based approach to big data by developing the essential features of a big data system as independent components with polymorphic implementations to support different requirements. Consequently, we recognize the requirements of both dataflow used in popular Apache Systems and the Bulk Synchronous Processing communication style common in High-Performance Computing (HPC) for different applications. Message Passing Interface (MPI) implementations are dominant in HPC but there are no such standard libraries available for big data. Twister:Net is a stand-alone, highly optimized dataflow style parallel communication library which can be used by big data systems or advanced users. Twister:Net can work both in cloud environments using TCP or HPC environments using MPI implementations. This paper introduces Twister:Net and compares it with existing systems to highlight its design and performance. © 2018 IEEE.},
 bibtype = {inproceedings},
 author = {Kamburugamuve, Supun and Wickramasinghe, Pulasthi and Govindarajan, Kannan and Uyar, Ahmet and Gunduz, Gurhan and Abeykoon, Vibhatha and Fox, Geoffrey},
 doi = {10.1109/CLOUD.2018.00055},
 booktitle = {IEEE International Conference on Cloud Computing, CLOUD}
}

Streaming processing and batch data processing are the dominant forms of big data analytics today, with numerous systems such as Hadoop, Spark, and Heron designed to process the ever-increasing explosion of data. Generally, these systems are developed as single projects with aspects such as communication, task management, and data management integrated together. By contrast, we take a component-based approach to big data by developing the essential features of a big data system as independent components with polymorphic implementations to support different requirements. Consequently, we recognize the requirements of both dataflow used in popular Apache Systems and the Bulk Synchronous Processing communication style common in High-Performance Computing (HPC) for different applications. Message Passing Interface (MPI) implementations are dominant in HPC but there are no such standard libraries available for big data. Twister:Net is a stand-alone, highly optimized dataflow style parallel communication library which can be used by big data systems or advanced users. Twister:Net can work both in cloud environments using TCP or HPC environments using MPI implementations. This paper introduces Twister:Net and compares it with existing systems to highlight its design and performance. © 2018 IEEE.

@article{
 title = {Anatomy of machine learning algorithm implementations in MPI, Spark, and Flink},
 type = {article},
 year = {2018},
 keywords = {Artificial intelligence; Big data; Data flow analy,Data flow modeling; Flink; High performance compu,Learning algorithms},
 pages = {61-73},
 volume = {32},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85039854240&doi=10.1177%2F1094342017712976&partnerID=40&md5=0a1048e69609d95f438e0b2f01466624},
 publisher = {SAGE Publications Inc.},
 id = {fee0ce1f-2b00-3b30-aa0a-c17641db8593},
 created = {2019-10-01T17:21:02.665Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:02.665Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kamburugamuve201861},
 source_type = {article},
 notes = {cited By 1},
 private_publication = {false},
 abstract = {With the ever-increasing need to analyze large amounts of data to get useful insights, it is essential to develop complex parallel machine learning algorithms that can scale with data and number of parallel processes. These algorithms need to run on large data sets as well as they need to be executed with minimal time in order to extract useful information in a time-constrained environment. Message passing interface (MPI) is a widely used model for developing such algorithms in high-performance computing paradigm, while Apache Spark and Apache Flink are emerging as big data platforms for large-scale parallel machine learning. Even though these big data frameworks are designed differently, they follow the data flow model for execution and user APIs. Data flow model offers fundamentally different capabilities than the MPI execution model, but the same type of parallelism can be used in applications developed in both models. This article presents three distinct machine learning algorithms implemented in MPI, Spark, and Flink and compares their performance and identifies strengths and weaknesses in each platform. © 2017, © The Author(s) 2017.},
 bibtype = {article},
 author = {Kamburugamuve, S and Wickramasinghe, P and Ekanayake, S and Fox, Geoffrey Charles},
 doi = {10.1177/1094342017712976},
 journal = {International Journal of High Performance Computing Applications},
 number = {1}
}

With the ever-increasing need to analyze large amounts of data to get useful insights, it is essential to develop complex parallel machine learning algorithms that can scale with data and number of parallel processes. These algorithms need to run on large data sets as well as they need to be executed with minimal time in order to extract useful information in a time-constrained environment. Message passing interface (MPI) is a widely used model for developing such algorithms in high-performance computing paradigm, while Apache Spark and Apache Flink are emerging as big data platforms for large-scale parallel machine learning. Even though these big data frameworks are designed differently, they follow the data flow model for execution and user APIs. Data flow model offers fundamentally different capabilities than the MPI execution model, but the same type of parallelism can be used in applications developed in both models. This article presents three distinct machine learning algorithms implemented in MPI, Spark, and Flink and compares their performance and identifies strengths and weaknesses in each platform. © 2017, © The Author(s) 2017.

@article{
 title = {Global analysis of charge exchange meson production at high energies},
 type = {article},
 year = {2018},
 keywords = {doi:10.1103/PhysRevD.98.034020 url:https://doi.org},
 id = {2f92bdd6-79a9-3829-84ca-a27996b929c2},
 created = {2019-10-01T17:21:02.857Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.387Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Nys2018},
 private_publication = {false},
 abstract = {Many experiments that are conducted to study the hadron spectrum rely on peripheral resonance production. Hereby, the rapidity gap allows the process to be viewed as an independent fragmentation of the beam and the target, with the beam fragmentation dominated by production and decays of meson resonances. We test this separation by determining the kinematic regimes that are dominated by factorizable contributions, indicating the most favorable regions to perform this kind of experiments. In doing so, we use a Regge model to analyze the available world data of charge exchange meson production with beam momentum above 5 GeV in the laboratory frame that are not dominated by either pion or Pomeron exchanges. We determine the Regge residues and point out the kinematic regimes which are dominated by factorizable contributions.},
 bibtype = {article},
 author = {Nys, J and Hiller Blin, A N and Mathieu, V and Fernández-Ramírez, C and Jackura, A and Pilloni, A and Ryckebusch, J and Szczepaniak, A P and Fox, G},
 doi = {10.1103/PhysRevD.98.034020}
}

Many experiments that are conducted to study the hadron spectrum rely on peripheral resonance production. Hereby, the rapidity gap allows the process to be viewed as an independent fragmentation of the beam and the target, with the beam fragmentation dominated by production and decays of meson resonances. We test this separation by determining the kinematic regimes that are dominated by factorizable contributions, indicating the most favorable regions to perform this kind of experiments. In doing so, we use a Regge model to analyze the available world data of charge exchange meson production with beam momentum above 5 GeV in the laboratory frame that are not dominated by either pion or Pomeron exchanges. We determine the Regge residues and point out the kinematic regimes which are dominated by factorizable contributions.

@inproceedings{
 title = {Task-parallel analysis of molecular dynamics trajectories},
 type = {inproceedings},
 year = {2018},
 keywords = {Data analytics,MD analysis,MD simulations analysis,Task-parallel},
 month = {8},
 publisher = {Association for Computing Machinery},
 day = {13},
 id = {ec60cc2e-ca18-3e16-8fac-313503fbbe74},
 created = {2019-10-01T17:21:02.886Z},
 accessed = {2019-09-03},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.324Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Paraskevakos2018},
 private_publication = {false},
 abstract = {Different parallel frameworks for implementing data analysis applications have been proposed by the HPC and Big Data communities. In this paper, we investigate three task-parallel frameworks: Spark, Dask and RADICAL-Pilot with respect to their ability to support data analytics on HPC resources and compare them with MPI. We investigate the data analysis requirements of Molecular Dynamics (MD) simulations which are significant consumers of supercomputing cycles, producing immense amounts of data. A typical large-scale MD simulation of a physical system of O(100k) atoms over \musecs can produce from O(10) GB to O(1000) GBs of data. We propose and evaluate different approaches for parallelization of a representative set of MD trajectory analysis algorithms, in particular the computation of path similarity and leaflet identification. We evaluate Spark, Dask and RADICAL-Pilot with respect to their abstractions and runtime engine capabilities to support these algorithms. We provide a conceptual basis for comparing and understanding different frameworks that enable users to select the optimal system for each application. We also provide a quantitative performance analysis of the different algorithms across the three frameworks.},
 bibtype = {inproceedings},
 author = {Paraskevakos, Ioannis and Chantzialexiou, George and Luckow, Andre and Cheatham, Thomas E. and Khoshlessan, Mahzad and Beckstein, Oliver and Fox, Geoffrey C. and Jha, Shantenu},
 doi = {10.1145/3225058.3225128},
 booktitle = {Proceedings of the 47th International Conference on Parallel Processing (ICPP 2018)}
}

Different parallel frameworks for implementing data analysis applications have been proposed by the HPC and Big Data communities. In this paper, we investigate three task-parallel frameworks: Spark, Dask and RADICAL-Pilot with respect to their ability to support data analytics on HPC resources and compare them with MPI. We investigate the data analysis requirements of Molecular Dynamics (MD) simulations which are significant consumers of supercomputing cycles, producing immense amounts of data. A typical large-scale MD simulation of a physical system of O(100k) atoms over \musecs can produce from O(10) GB to O(1000) GBs of data. We propose and evaluate different approaches for parallelization of a representative set of MD trajectory analysis algorithms, in particular the computation of path similarity and leaflet identification. We evaluate Spark, Dask and RADICAL-Pilot with respect to their abstractions and runtime engine capabilities to support these algorithms. We provide a conceptual basis for comparing and understanding different frameworks that enable users to select the optimal system for each application. We also provide a quantitative performance analysis of the different algorithms across the three frameworks.

@inproceedings{
 title = {Searching the Sequence Read Archive Using Jetstream and Wrangler},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,Bacteriophage,Credential Store,Jetstream,Metagenomics,Metagenomics Discovery Challenge,SRA,SRA Gateway,SciGaP,Search SRA,Sequence Read Archive,Wrangler},
 pages = {50:1--50:7},
 websites = {http://doi.acm.org/10.1145/3219104.3229278},
 month = {7},
 publisher = {ACM},
 day = {22},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {0aa8664d-88c2-338b-bb19-e0153c638275},
 created = {2019-10-01T17:21:06.731Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:30:53.978Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Levi:2018:SSR:3219104.3229278},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {The Sequence Read Archive (SRA), the world’s largest database of sequences, hosts approximately 10 petabases (1016 bp) of sequence data and is growing at the alarming rate of 10 TB per day. Yet this rich trove of data is inaccessible to most researchers: searching through the SRA requires large storage and computing facilities that are beyond the capacity of most laboratories. Enabling scientists to analyze existing sequence data will provide insight into ecology, medicine, and industrial applications. In this project we specifi- cally focus on metagenomic sequences (whole community data sets from different environments). We are developing a set of tools to enable biologists to mine the metagenomes in the SRA using the NSF-funded cloud computing resources, Jetstream and Wrangler. We have developed a proof-of-principle pipeline to demonstrate the feasibility of the approach. We are leveraging our existing in- frastructure to enable all scientists to access the SRA metagenomes regardless of their computational ability and are working to create a stable pipeline with a science gateway portal that is accessible to all researchers.تنيىسبت},
 bibtype = {inproceedings},
 author = {Levi, Kyle and Rynge, Mats and Abeysinghe, Eroma and Edwards, Robert A},
 doi = {10.1145/3219104.3229278},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

The Sequence Read Archive (SRA), the world’s largest database of sequences, hosts approximately 10 petabases (1016 bp) of sequence data and is growing at the alarming rate of 10 TB per day. Yet this rich trove of data is inaccessible to most researchers: searching through the SRA requires large storage and computing facilities that are beyond the capacity of most laboratories. Enabling scientists to analyze existing sequence data will provide insight into ecology, medicine, and industrial applications. In this project we specifi- cally focus on metagenomic sequences (whole community data sets from different environments). We are developing a set of tools to enable biologists to mine the metagenomes in the SRA using the NSF-funded cloud computing resources, Jetstream and Wrangler. We have developed a proof-of-principle pipeline to demonstrate the feasibility of the approach. We are leveraging our existing in- frastructure to enable all scientists to access the SRA metagenomes regardless of their computational ability and are working to create a stable pipeline with a science gateway portal that is accessible to all researchers.تنيىسبت

@inproceedings{
 title = {Grid Technology for Supporting Health Education and Measuring the Health Outcome},
 type = {inproceedings},
 year = {2018},
 keywords = {Community health,Cyberinfrastructure,Data grid,Data integration,Grid computing,Health education,IRODS,Mobile technology,Portal,Virtual,XSEDE,community health,cyberinfrastructure,data grid,data integration,grid computing,health education,iRODS,mobile technology,portal,virtual},
 pages = {89:1--89:4},
 websites = {http://doi.acm.org/10.1145/3219104.3229247},
 month = {7},
 publisher = {ACM},
 day = {22},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {b451500f-8b78-390a-adc2-ac93018d72dc},
 created = {2019-10-01T17:21:06.910Z},
 accessed = {2019-08-26},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:30:52.752Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Sukhija:2018:GTS:3219104.3229247},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {In this paper, we present our developed health-IT solution that address these challenges: developing the strategies not only to store such a vast amount of data but also making those available to the researchers for further analysis that can measure the outcome on the participant's health. The developed community health grid (C-Grid) solution to store, manage and share large amounts of these instruction materials and participant's health related data, where the remote management and analysis of this data grid is performed using iRODS, the Integrated Rule-Oriented Data System is discussed and presented in this paper.},
 bibtype = {inproceedings},
 author = {Sukhija, Nitin and Datta, Arun K. and Sevin, Sonny and Coulter, Eric and Datta, Arun K. and Coulter, Eric},
 doi = {10.1145/3219104.3229247},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

In this paper, we present our developed health-IT solution that address these challenges: developing the strategies not only to store such a vast amount of data but also making those available to the researchers for further analysis that can measure the outcome on the participant's health. The developed community health grid (C-Grid) solution to store, manage and share large amounts of these instruction materials and participant's health related data, where the remote management and analysis of this data grid is performed using iRODS, the Integrated Rule-Oriented Data System is discussed and presented in this paper.

@article{
 title = {Jetstream—Early operations performance, adoption, and impacts},
 type = {article},
 year = {2018},
 keywords = {OpenStack,cloud computing,long tail of science,science impacts,user adoption,virtual machines},
 websites = {https://onlinelibrary.wiley.com/doi/abs/10.1002/cpe.4683},
 month = {9},
 publisher = {John Wiley and Sons Ltd},
 day = {2},
 id = {920fbf64-e3b7-31fd-adea-0341c9525761},
 created = {2019-10-01T17:21:07.185Z},
 accessed = {2019-08-14},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:59:22.166Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hancock2018},
 private_publication = {false},
 abstract = {Jetstream is a first of its kind system for the NSF — a distributed production cloud resource. We review the purpose for creating Jetstream, discuss Jetstream's key characteristics, describe our experiences from the first year of maintaining an OpenStack‐based cloud environment, and share some of the early scientific impacts achieved by Jetstream users. Jetstream offers a unique capability within the XSEDE‐supported US national cyberinfrastructure, delivering interactive virtual machines (VMs) via the Atmosphere interface. As a multi‐region deployment that operates as an integrated system, Jetstream is proving effective in supporting modes and disciplines of research traditionally underrepresented on larger XSEDE‐supported clusters and supercomputers. Already, Jetstream has been used to perform research and education in biology, biochemistry, atmospheric science, earth science, and computer science.},
 bibtype = {article},
 author = {Hancock, David Y. and Stewart, Craig A. and Vaughn, Matthew and Fischer, Jeremy and Lowe, John Michael and Turner, George and Swetnam, Tyson L. and Chafin, Tyler K. and Afgan, Enis and Pierce, Marlon E. and Snapp‐Childs, Winona and Snapp-Childs, Winona},
 doi = {10.1002/cpe.4683},
 journal = {Concurrency and Computation: Practice and Experience}
}

Jetstream is a first of its kind system for the NSF — a distributed production cloud resource. We review the purpose for creating Jetstream, discuss Jetstream's key characteristics, describe our experiences from the first year of maintaining an OpenStack‐based cloud environment, and share some of the early scientific impacts achieved by Jetstream users. Jetstream offers a unique capability within the XSEDE‐supported US national cyberinfrastructure, delivering interactive virtual machines (VMs) via the Atmosphere interface. As a multi‐region deployment that operates as an integrated system, Jetstream is proving effective in supporting modes and disciplines of research traditionally underrepresented on larger XSEDE‐supported clusters and supercomputers. Already, Jetstream has been used to perform research and education in biology, biochemistry, atmospheric science, earth science, and computer science.

@techreport{
 title = {The Report of the 2018 NSF Cybersecurity Summit for Large Facilities and Cyberinfrastructure},
 type = {techreport},
 year = {2018},
 websites = {http://hdl.handle.net/2022/22588.},
 id = {d046dd10-b0f1-3ffb-a184-366947174419},
 created = {2019-10-01T17:21:18.861Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.471Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {AndrewAdamsJeannetteDopheideMarkKrenzJamesMarsteller2018},
 private_publication = {false},
 bibtype = {techreport},
 author = {Andrew Adams, Jeannette Dopheide, Mark Krenz, James Marsteller, Von Welch and and John Zage, undefined}
}

@techreport{
 title = {Trusted CI Annual Report for 2018},
 type = {techreport},
 year = {2018},
 websites = {hp://hdl.handle.net/2022/22597},
 id = {8521c330-0ce3-30de-87d8-ceb3ed5be4d5},
 created = {2019-10-01T17:21:21.337Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.283Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Adams2018},
 private_publication = {false},
 bibtype = {techreport},
 author = {Adams, Andrew and Avila, Kay and Atkins, Joel and Basney, Jim and Bohland, Leslee and Borecky, Diana and Cowles, Robert and Dopheide, Jeannee and Fleury, Terry and Harbour, Grayson and Heymann, Elisa and Hudson, Florence and Jackson, Craig and Kiser, Ryan and Krenz, Mark and Marsteller, Jim and Miller, Barton and Raquel, Warren and Ruff, Preston and Russell, Scoo and Shah, Zalak and Shankar, Anurag and Sons, Susan and Welch, Von and Zage, John}
}

@inproceedings{
 title = {A New Science Gateway to Provide Decision Support on Carbon Capture and Storage Technologies},
 type = {inproceedings},
 year = {2018},
 keywords = {Carbon capture,Science gateways},
 pages = {1-3},
 websites = {http://doi.acm.org/10.1145/3219104.3229244,http://dl.acm.org/citation.cfm?doid=3219104.3229244},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 series = {PEARC '18},
 id = {9a7802b5-afcb-3485-bcc1-6a50f5bb5320},
 created = {2019-10-01T17:21:22.663Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:22.663Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wang:2018:NSG:3219104.3229244},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {Carbon dioxide capture and storage (CCS) is a promising technology for mitigating climate change, and its implementation is considered critical to meeting threshold targets for global warming in the 21st century. We have developed a new science gateway application for the successful modeling software known as SimCCS that is used for evaluating complex, integrated CCS infrastructure. Using the Apache Airavata middleware and high-performance computing resources made available by the Extreme Science and Engineering Discovery Environment, we built the SimCCS Gateway to expand the tool's scalability for decision support and risk assessment. Case studies developed for evaluating a proposed CCS technology at Duke Energy's Gibson Station coal-fired power plant in southwest Indiana demonstrate its improved ability in data analysis as well as risk assessment at various uncertainty levels. Further work is continuing to expand the functionality of both web and desktop clients, and to develop an active user group community in research and industry via the SimCCS Gateway interface.},
 bibtype = {inproceedings},
 author = {Wang, Yinzhi and Pamidighantam, Sudhakar and Yaw, Sean and Abeysinghe, Eroma and Marru, Suresh and Christie, Marcus and Ellett, Kevin and Pierce, Marlon and Middleton, Richard},
 doi = {10.1145/3219104.3229244},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Carbon dioxide capture and storage (CCS) is a promising technology for mitigating climate change, and its implementation is considered critical to meeting threshold targets for global warming in the 21st century. We have developed a new science gateway application for the successful modeling software known as SimCCS that is used for evaluating complex, integrated CCS infrastructure. Using the Apache Airavata middleware and high-performance computing resources made available by the Extreme Science and Engineering Discovery Environment, we built the SimCCS Gateway to expand the tool's scalability for decision support and risk assessment. Case studies developed for evaluating a proposed CCS technology at Duke Energy's Gibson Station coal-fired power plant in southwest Indiana demonstrate its improved ability in data analysis as well as risk assessment at various uncertainty levels. Further work is continuing to expand the functionality of both web and desktop clients, and to develop an active user group community in research and industry via the SimCCS Gateway interface.

@inproceedings{
 title = {Science Gateway Implementation at the University of South Dakota},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,Gateway,Keycloak,SciGaP},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229265},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {c9d11ec4-7bfb-3ded-8c6c-ab4bc1411d9d},
 created = {2019-10-01T17:21:23.986Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.546Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Madison2018},
 private_publication = {false},
 abstract = {Science Gateways are virtual environments that accelerate scientific discovery by enabling scientific communities to more easily and effectively utilize distributed computing and data resources. Successful Science Gateways provide access to sophisticated and powerful resources, while shielding their users from the underlying complexities. Here we present work completed by the University of South Dakota (USD) Research Computing Group in conjunction with the Science Gateways Community Institute (SGCI) [1] and Indiana University on setting up a Science Gateway to access USD's high-performance computing resources. These resources are now available to both faculty and students and allow ease of access and use of USD's distributed computing and data resources. The implementation of this gateway project has been multifaceted and has included placement of federated user login, user facilitation and outreach, and integration of USD's cyberinfrastructure resources. We present this project as an example for other research computing groups so that they may learn from our successes and the challenges that we have overcome in providing this user resource. Additionally, this project serves to exemplify the importance of creating a broad user base of research computing infrastructure through the development of alternative user interfaces such as Science Gateways.},
 bibtype = {inproceedings},
 author = {Madison, Joseph D. and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Marru, Suresh and Christie, Marcus and Jennewein, Douglas M. and Pierce, Marlon},
 doi = {10.1145/3219104.3229265},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Science Gateways are virtual environments that accelerate scientific discovery by enabling scientific communities to more easily and effectively utilize distributed computing and data resources. Successful Science Gateways provide access to sophisticated and powerful resources, while shielding their users from the underlying complexities. Here we present work completed by the University of South Dakota (USD) Research Computing Group in conjunction with the Science Gateways Community Institute (SGCI) [1] and Indiana University on setting up a Science Gateway to access USD's high-performance computing resources. These resources are now available to both faculty and students and allow ease of access and use of USD's distributed computing and data resources. The implementation of this gateway project has been multifaceted and has included placement of federated user login, user facilitation and outreach, and integration of USD's cyberinfrastructure resources. We present this project as an example for other research computing groups so that they may learn from our successes and the challenges that we have overcome in providing this user resource. Additionally, this project serves to exemplify the importance of creating a broad user base of research computing infrastructure through the development of alternative user interfaces such as Science Gateways.

@inproceedings{
 title = {Django Content Management System Evaluation and Integration with Apache Airavata},
 type = {inproceedings},
 year = {2018},
 keywords = {ACM proceedings,Apache Airavata,Django Framework,Science Gateway,Text tagging,Wagtail CMS},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229272},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {c6abbe69-ea24-3a68-bae1-288f60799003},
 created = {2019-10-01T17:21:24.843Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.127Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Adithela2018},
 private_publication = {false},
 abstract = {Apache Airavata is an open-source software framework that enables scientific researchers to compose, manage, execute and monitor large-scale applications and workflows on distributed computing resources. Airavata is currently leveraged by many science gateways to perform computations on shared clusters. Currently, Gateway Administrators managing content on their websites will require the assistance of the Airavata Developer Team to make the slightest of change to their website. This paper will overcome this challenge by presenting the benefits of integrating a content management system. It will also briefly evaluate various options available for choosing a Content Management Platform which complies with the Airavata Architecture Standards. This feature will enable researchers with minimal web design knowledge to easily manage content across their gateway. It is also poised to drastically increase the productivity of the Airavata developer team and the gateway administrators. © 2018 Association for Computing Machinery.},
 bibtype = {inproceedings},
 author = {Adithela, Stephen Paul and Christie, Marcus and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/3219104.3229272},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Apache Airavata is an open-source software framework that enables scientific researchers to compose, manage, execute and monitor large-scale applications and workflows on distributed computing resources. Airavata is currently leveraged by many science gateways to perform computations on shared clusters. Currently, Gateway Administrators managing content on their websites will require the assistance of the Airavata Developer Team to make the slightest of change to their website. This paper will overcome this challenge by presenting the benefits of integrating a content management system. It will also briefly evaluate various options available for choosing a Content Management Platform which complies with the Airavata Architecture Standards. This feature will enable researchers with minimal web design knowledge to easily manage content across their gateway. It is also poised to drastically increase the productivity of the Airavata developer team and the gateway administrators. © 2018 Association for Computing Machinery.

@inbook{
 type = {inbook},
 year = {2018},
 pages = {101-116},
 publisher = {Springer},
 id = {e0e289b6-bcaa-38e7-9be3-96b2820b9089},
 created = {2019-10-01T17:21:26.366Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.268Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Parker2018},
 source_type = {CHAP},
 private_publication = {false},
 bibtype = {inbook},
 author = {Parker, Jay and Glasscoe, Margaret and Donnellan, Andrea and Stough, Timothy and Pierce, Marlon and Wang, Jun},
 chapter = {Radar Determination of Fault Slip and Location in Partially Decorrelated Images},
 title = {Earthquakes and Multi-hazards Around the Pacific Rim, Vol. I}
}

@inproceedings{
 title = {Your Good Health is a Workforce Issue},
 type = {inproceedings},
 year = {2018},
 keywords = {SAD,Standard American Diet,acm reference format,cancer,health,life balance,preventive testing,sad,standard american diet,stress,work,work/life balance,workforce development},
 pages = {75:1--75:8},
 websites = {http://doi.acm.org/10.1145/3219104.3219107},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {5ce71fed-56ff-3ad9-b68f-36fcd7e5e191},
 created = {2019-10-01T17:21:27.662Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:27.662Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart:2018:YGH:3219104.3219107},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {The high performance computing (HPC), cyberinfrastructure, and research and academic information technology communities are small - too small to fulfill current needs for such professionals in the US. Members of this community are also often under a lot of stress, and with that can come health problems. The senior author was diagnosed with Stage IV cancer in early 2017. In this paper, we share what we have learned about health management in general and dealing with cancer in particular, focusing on lessons that are portable to other members of the HPC, cyberinfrastructure, and research and academic information technology communities. We also make recommendations to the National Science Foundation regarding changes the NSF could make to reduce some of the stress this community feels on a day-in, day-out basis. The key point of this report is to provide information to members of the cyberinfrastructure community that they might not already have - and might not receive from their primary care physicians - that will help them live longer and healthier lives. While our own experiences are based on one of the author's diagnosis of cancer, the information presented here should be of general value to all in terms of strategies for reducing and detecting long-term health risks. Our hope is that this information will help you be as healthy as possible until you reach retirement age and then healthy during a well-deserved and long period of retirement!},
 bibtype = {inproceedings},
 author = {Stewart, Craig A and Krefeldt, Marion},
 doi = {10.1145/3219104.3219107},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

The high performance computing (HPC), cyberinfrastructure, and research and academic information technology communities are small - too small to fulfill current needs for such professionals in the US. Members of this community are also often under a lot of stress, and with that can come health problems. The senior author was diagnosed with Stage IV cancer in early 2017. In this paper, we share what we have learned about health management in general and dealing with cancer in particular, focusing on lessons that are portable to other members of the HPC, cyberinfrastructure, and research and academic information technology communities. We also make recommendations to the National Science Foundation regarding changes the NSF could make to reduce some of the stress this community feels on a day-in, day-out basis. The key point of this report is to provide information to members of the cyberinfrastructure community that they might not already have - and might not receive from their primary care physicians - that will help them live longer and healthier lives. While our own experiences are based on one of the author's diagnosis of cancer, the information presented here should be of general value to all in terms of strategies for reducing and detecting long-term health risks. Our hope is that this information will help you be as healthy as possible until you reach retirement age and then healthy during a well-deserved and long period of retirement!

@inproceedings{
 title = {Using a Science Gateway to Deliver SimVascular Software As a Service for Classroom Instruction},
 type = {inproceedings},
 year = {2018},
 keywords = {ACM proceedings,Apache Airavata,Science Gateway},
 pages = {102:1--102:4},
 websites = {http://doi.acm.org/10.1145/3219104.3229242},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {da3e301a-0208-3822-ba51-ece0d36db76f},
 created = {2019-10-01T17:21:29.609Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:29.609Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wilson:2018:USG:3219104.3229242},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {SimVascular (http://www.simvascular.org) is open source software enabling users to construct image-based, patient-specific anatomic models and perform realistic blood flow simulation useful in disease research, medical device design, and surgical planning. The software consists of two core executables: a front-end application and a flow solver. The front-end application enables users to create patient-specific anatomic models from imaging data, generate finite-element meshes, prescribe boundary conditions, and set up an analysis. The finite-element based blood flow solver utilizes MPI and is massively scalable. SimVascular has been successfully integrated into graduate level courses on cardiovascular modeling at multiple institutions including Stanford, UC Berkeley, Purdue, and Marquette to introduce state-of-the-art modeling to the students and provide a basis for hands-on projects. While the front-end application can be installed and run on a laptop, the flow solver requires high performance computing (HPC) for realistic problem sizes. This provides a significant challenge for instructors as many students are unfamiliar with HPC, and local resources might be limited or difficult to administer. There is also a need to provide user and group management capabilities for courses: students should authenticate using campus credentials, instructors should be able to access students' work, and students' access to computing allocations should be limited. Our poster will detail an Apache Airavata-based science gateway to address these needs. XSEDE's Comet provides the backend computing power. This approach allows the SimVascular team to provision HPC resources and install and maintain the software providing students access at institutions across the country. The science gateway interface provides access to SimVascular's flow solver, while allowing students to use SimVascular's desktop interfaces.},
 bibtype = {inproceedings},
 author = {Wilson, Nathan M and Marru, Suresh and Abeysinghe, Eroma and Christie, Marcus A and Maher, Gabriel D and Updegrove, Adam R and Pierce, Marlon and Marsden, Alison L},
 doi = {10.1145/3219104.3229242},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing}
}

SimVascular (http://www.simvascular.org) is open source software enabling users to construct image-based, patient-specific anatomic models and perform realistic blood flow simulation useful in disease research, medical device design, and surgical planning. The software consists of two core executables: a front-end application and a flow solver. The front-end application enables users to create patient-specific anatomic models from imaging data, generate finite-element meshes, prescribe boundary conditions, and set up an analysis. The finite-element based blood flow solver utilizes MPI and is massively scalable. SimVascular has been successfully integrated into graduate level courses on cardiovascular modeling at multiple institutions including Stanford, UC Berkeley, Purdue, and Marquette to introduce state-of-the-art modeling to the students and provide a basis for hands-on projects. While the front-end application can be installed and run on a laptop, the flow solver requires high performance computing (HPC) for realistic problem sizes. This provides a significant challenge for instructors as many students are unfamiliar with HPC, and local resources might be limited or difficult to administer. There is also a need to provide user and group management capabilities for courses: students should authenticate using campus credentials, instructors should be able to access students' work, and students' access to computing allocations should be limited. Our poster will detail an Apache Airavata-based science gateway to address these needs. XSEDE's Comet provides the backend computing power. This approach allows the SimVascular team to provision HPC resources and install and maintain the software providing students access at institutions across the country. The science gateway interface provides access to SimVascular's flow solver, while allowing students to use SimVascular's desktop interfaces.

@inproceedings{
 title = {PHASTA Science Gateway for High Performance Computational Fluid Dynamics},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,CFD,Paralllel Unstructured Mesh,Pervasive Technology Institute,Science Gateway,Science Gateways Research Center},
 pages = {1-4},
 websites = {http://dl.acm.org/citation.cfm?doid=3219104.3229243},
 month = {7},
 publisher = {ACM Press},
 day = {22},
 city = {New York, New York, USA},
 id = {467c8437-2ddb-3d8e-96f0-d91b2dcf4881},
 created = {2019-10-01T17:21:30.353Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.407Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Smith2018},
 private_publication = {false},
 abstract = {The Parallel Hierarchic Adaptive Stabilized Transient Analysis (PHASTA) software supports modeling compressible or incompressible, laminar or turbulent, steady or unsteady flows in 3D using unstructured grids. PHASTA has been applied to industrial and academic flows on complex, as-designed geometric models with over one billion mesh elements using upwards of one million compute cores. The PHASTA Science Gateway (phasta.scigap.org) brings these increasingly critical technologies to a larger user base by providing a central hub for simulation execution, simulation data management, and documentation. Researchers and engineers using the gateway can easily define and execute simulations on the TACC Stampede2 Skylake and Knights Landing nodes without being burdened by the details of remote access, the job scheduler, and filesystem configuration. In addition to simplifying the simulation execution process, the gateway creates a searchable archive of past jobs that can be shared with other users to support reproducibility and increase productivity. Our poster presents the construction of the gateway with Apache Airavata, the simulation definition process, applications it currently supports, and our ongoing efforts to expand functionality, the user base, and the community.},
 bibtype = {inproceedings},
 author = {Smith, Cameron W. and Abeysinghe, Eroma and Marru, Suresh and Jansen, Kenneth E.},
 doi = {10.1145/3219104.3229243},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

The Parallel Hierarchic Adaptive Stabilized Transient Analysis (PHASTA) software supports modeling compressible or incompressible, laminar or turbulent, steady or unsteady flows in 3D using unstructured grids. PHASTA has been applied to industrial and academic flows on complex, as-designed geometric models with over one billion mesh elements using upwards of one million compute cores. The PHASTA Science Gateway (phasta.scigap.org) brings these increasingly critical technologies to a larger user base by providing a central hub for simulation execution, simulation data management, and documentation. Researchers and engineers using the gateway can easily define and execute simulations on the TACC Stampede2 Skylake and Knights Landing nodes without being burdened by the details of remote access, the job scheduler, and filesystem configuration. In addition to simplifying the simulation execution process, the gateway creates a searchable archive of past jobs that can be shared with other users to support reproducibility and increase productivity. Our poster presents the construction of the gateway with Apache Airavata, the simulation definition process, applications it currently supports, and our ongoing efforts to expand functionality, the user base, and the community.

@inproceedings{
 title = {Simplifying Access to Campus Resources at Southern Illinois University with a Science Gateway},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,MaSuRCA,Science Gateway},
 pages = {1-4},
 websites = {http://doi.acm.org/10.1145/3219104.3229252,http://dl.acm.org/citation.cfm?doid=3219104.3229252},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 series = {PEARC '18},
 id = {520a034a-d0f2-3f63-9666-5aae497e287b},
 created = {2019-10-01T17:21:30.464Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:30.464Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Sunkara:2018:SAC:3219104.3229252},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {Not all the researchers are comfortable in using High Performance Computing (HPC) Systems. Southern Illinois University's Office of Information Technology (OIT) Research Computing assists our researchers to get on these systems and use them for their research activities. One such potential use case at SIU involves the group of researchers from Life Sciences who were trying to use MaSuRCA [3], a genome-sequencing tool for their research. Although the leaders of this research are well versed in using BigDog [4](SIU's HPC Cluster), other fellow researchers did have issues in using the cluster for their work. It was time to look for efficient ways of enabling them to use the cluster. We examined using Science Gateways which can help our researchers to use the computational cluster without logging on to a Linux-based HPC system. OIT is currently collaborating with the Science Gateways Research Center at Indiana University (IU) on the use of Apache Airavata [1] as a Science Gateway framework for the MaSuRCA user community at SIU. The IU team members provide hosting and operations for Apache Airavata middleware as part of the SciGaP.org project. IU collaborators also provide a basic science gateway user interface, the PGA. The SIU gateway, although hosted off campus, is integrated with SIU's BigDog cluster.},
 bibtype = {inproceedings},
 author = {Sunkara, Sai Susheel and Langin, Chet and Pierce, Marlon and Abeysinghe, Eroma and Pamidighantam, Sudhakar and Marru, Suresh},
 doi = {10.1145/3219104.3229252},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing - PEARC '18}
}

Not all the researchers are comfortable in using High Performance Computing (HPC) Systems. Southern Illinois University's Office of Information Technology (OIT) Research Computing assists our researchers to get on these systems and use them for their research activities. One such potential use case at SIU involves the group of researchers from Life Sciences who were trying to use MaSuRCA [3], a genome-sequencing tool for their research. Although the leaders of this research are well versed in using BigDog [4](SIU's HPC Cluster), other fellow researchers did have issues in using the cluster for their work. It was time to look for efficient ways of enabling them to use the cluster. We examined using Science Gateways which can help our researchers to use the computational cluster without logging on to a Linux-based HPC system. OIT is currently collaborating with the Science Gateways Research Center at Indiana University (IU) on the use of Apache Airavata [1] as a Science Gateway framework for the MaSuRCA user community at SIU. The IU team members provide hosting and operations for Apache Airavata middleware as part of the SciGaP.org project. IU collaborators also provide a basic science gateway user interface, the PGA. The SIU gateway, although hosted off campus, is integrated with SIU's BigDog cluster.

@inproceedings{
 title = {Building a Science Gateway For Processing and Modeling Sequencing Data Via Apache Airavata},
 type = {inproceedings},
 year = {2018},
 keywords = {Apache Airavata,Next Generation Sequencing,Science gateway,cloud computing,sequencing data,software-as-a-service},
 pages = {39:1--39:7},
 websites = {http://doi.acm.org/10.1145/3219104.3219141},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {PEARC '18},
 id = {dc6648d0-be5e-3642-8535-a6360c187f25},
 created = {2019-10-01T17:21:30.492Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:30.492Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wang:2018:BSG:3219104.3219141},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {The amount of DNA sequencing data has been exponentially growing during the past decade due to advances in sequencing technology. Processing and modeling large amounts of sequencing data can be computationally intractable for desktop computing platforms. High performance computing (HPC) resources offer advantages in terms of computing power, and can be a general solution to these problems. Using HPCs directly for computational needs requires skilled users who know their way around HPCs and acquiring such skills take time. Science gateways acts as the middle layer between users and HPCs, providing users with the resources to accomplish compute-intensive tasks without requiring specialized expertise. We developed a web-based computing platform for genome biologists by customizing the PHP Gateway for Airavata (PGA) framework that accesses publicly accessible HPC resources via Apache Airavata. This web computing platform takes advantage of the Extreme Science and Engineering Discovery Environment (XSEDE) which provides the resources for gateway development, including access to CPU, GPU, and storage resources. We used this platform to develop a gateway for the dREG algorithm, an online computing tool for finding functional regions in mammalian genomes using nascent RNA sequencing data. The dREG gateway provides its users a free, powerful and user-friendly GPU computing resource based on XSEDE, circumventing the need of specialized knowledge about installation, configuration, and execution on an HPC for biologists. The dREG gateway is available at: https://dREG.dnasequence.org/.},
 bibtype = {inproceedings},
 author = {Wang, Zhong and Christie, Marcus A and Abeysinghe, Eroma and Chu, Tinyi and Marru, Suresh and Pierce, Marlon and Danko, Charles G},
 doi = {10.1145/3219104.3219141},
 booktitle = {Proceedings of the Practice and Experience on Advanced Research Computing (PEARC '18)}
}

The amount of DNA sequencing data has been exponentially growing during the past decade due to advances in sequencing technology. Processing and modeling large amounts of sequencing data can be computationally intractable for desktop computing platforms. High performance computing (HPC) resources offer advantages in terms of computing power, and can be a general solution to these problems. Using HPCs directly for computational needs requires skilled users who know their way around HPCs and acquiring such skills take time. Science gateways acts as the middle layer between users and HPCs, providing users with the resources to accomplish compute-intensive tasks without requiring specialized expertise. We developed a web-based computing platform for genome biologists by customizing the PHP Gateway for Airavata (PGA) framework that accesses publicly accessible HPC resources via Apache Airavata. This web computing platform takes advantage of the Extreme Science and Engineering Discovery Environment (XSEDE) which provides the resources for gateway development, including access to CPU, GPU, and storage resources. We used this platform to develop a gateway for the dREG algorithm, an online computing tool for finding functional regions in mammalian genomes using nascent RNA sequencing data. The dREG gateway provides its users a free, powerful and user-friendly GPU computing resource based on XSEDE, circumventing the need of specialized knowledge about installation, configuration, and execution on an HPC for biologists. The dREG gateway is available at: https://dREG.dnasequence.org/.

@inproceedings{
 title = {Big provenance stream processing for data intensive computations},
 type = {inproceedings},
 year = {2018},
 keywords = {Big Data,Big Provenance,Stream Processing},
 pages = {245-255},
 month = {12},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {24},
 id = {02faffe9-74a8-370f-9c62-67aef6795cd5},
 created = {2020-04-22T21:44:56.973Z},
 accessed = {2020-04-21},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.831Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Suriarachchi2018},
 private_publication = {false},
 abstract = {In the business and research landscape of today, data analysis consumes public and proprietary data from numerous sources, and utilizes any one or more of popular data-parallel frameworks such as Hadoop, Spark and Flink. In the Data Lake setting these frameworks co-exist. Our earlier work has shown that data provenance in Data Lakes can aid with both traceability and management. The sheer volume of fine-grained provenance generated in a multi-framework application motivates the need for on-the-fly provenance processing. We introduce a new parallel stream processing algorithm that reduces fine-grained provenance while preserving backward and forward provenance. The algorithm is resilient to provenance events arriving out-of-order. It is evaluated using several strategies for partitioning a provenance stream. The evaluation shows that the parallel algorithm performs well in processing out-of-order provenance streams, with good scalability and accuracy.},
 bibtype = {inproceedings},
 author = {Suriarachchi, Isuru and Withana, Sachith and Plale, Beth},
 doi = {10.1109/eScience.2018.00039},
 booktitle = {Proceedings - IEEE 14th International Conference on eScience, e-Science 2018}
}

In the business and research landscape of today, data analysis consumes public and proprietary data from numerous sources, and utilizes any one or more of popular data-parallel frameworks such as Hadoop, Spark and Flink. In the Data Lake setting these frameworks co-exist. Our earlier work has shown that data provenance in Data Lakes can aid with both traceability and management. The sheer volume of fine-grained provenance generated in a multi-framework application motivates the need for on-the-fly provenance processing. We introduce a new parallel stream processing algorithm that reduces fine-grained provenance while preserving backward and forward provenance. The algorithm is resilient to provenance events arriving out-of-order. It is evaluated using several strategies for partitioning a provenance stream. The evaluation shows that the parallel algorithm performs well in processing out-of-order provenance streams, with good scalability and accuracy.

@techreport{
 title = {ABI Sustaining: The National Center for Genome Analysis Support 2018 Annual Report},
 type = {techreport},
 year = {2018},
 keywords = {NCGAS,National Science Foundation},
 websites = {http://creativecommons.org/licenses/by/4.0/.},
 month = {9},
 day = {10},
 id = {91895bbb-5bce-3750-8a88-83f0d62b156c},
 created = {2020-09-09T20:50:48.696Z},
 accessed = {2020-09-09},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.372Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Doak2018},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {National Science Foundation ABI-1458641},
 bibtype = {techreport},
 author = {Doak, Thomas G. and Stewart, Craig A. and Michaels, Scott D}
}

National Science Foundation ABI-1458641

@techreport{
 title = {Summary of the National Center for Genome Analysis Support (NCGAS) 2018 de Novo Transcriptome Workflow and Workshop},
 type = {techreport},
 year = {2018},
 keywords = {NCGAS,NSF Report,Transcriptome Assembly,Workshop},
 websites = {https://scholarworks.iu.edu/dspace/handle/2022/22254},
 month = {6},
 day = {7},
 id = {d81eb260-ae70-3b34-929a-07aca2b774bc},
 created = {2020-09-09T21:03:42.928Z},
 accessed = {2020-09-09},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.385Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Sanders2018},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {The National Center for Genome Analysis Support (NCGAS) held a workshop entitled "de novo Assembly of Transcriptomes using HPC Resources" on April 30th, 2018 through May 1, 2018. This workshop was in serving NCGAS's mission of enabling the biological research community to analyze, understand, and make use of the genomic information now available by packaging our now seven years of experience assisting with de novo transcriptome assemblies and running High Performance Computing (HPC) resources into a documented, easily approachable workflow for our users. The workshop covered common questions and problems that our users have had in HC (such as job handling, resource availability, data management, and troubleshooting) and in the construction of transcriptomes (such as software choices, combination of assemblies, and downstream analyses). The two-day workshop also highlighted the available resources for US scientists, concentrating heavily on available XSEDE resources for analyses, visualization, and archiving of data.},
 bibtype = {techreport},
 author = {Sanders, Sheri A and Ganote, Carrie L and Papudeshi, Bhavya and Stewart, Craig A and Doak, Thomas G}
}

The National Center for Genome Analysis Support (NCGAS) held a workshop entitled "de novo Assembly of Transcriptomes using HPC Resources" on April 30th, 2018 through May 1, 2018. This workshop was in serving NCGAS's mission of enabling the biological research community to analyze, understand, and make use of the genomic information now available by packaging our now seven years of experience assisting with de novo transcriptome assemblies and running High Performance Computing (HPC) resources into a documented, easily approachable workflow for our users. The workshop covered common questions and problems that our users have had in HC (such as job handling, resource availability, data management, and troubleshooting) and in the construction of transcriptomes (such as software choices, combination of assemblies, and downstream analyses). The two-day workshop also highlighted the available resources for US scientists, concentrating heavily on available XSEDE resources for analyses, visualization, and archiving of data.

@article{
 title = {Hybrid Supervised-unsupervised Image Topic Visualization with Convolutional Neural Network and LDA},
 type = {article},
 year = {2017},
 id = {5f2ce4ba-5118-393a-82e2-7c871f086bb9},
 created = {2017-11-28T17:33:40.887Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.912Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhen2017},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Zhen, Kai and Birla, Mridul and Crandall, David and Zhang, Bingjing and Qiu, Judy},
 journal = {arXiv preprint arXiv:1703.05243}
}

@inproceedings{
 title = {Teaching, Learning and Collaborating through Cloud Computing Online Classes},
 type = {inproceedings},
 year = {2017},
 id = {25b62c01-4711-3dbc-86e6-32fe308c42d5},
 created = {2017-11-28T17:33:41.112Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.698Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Qiu2017},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Qiu, Judy and Kamburugamuve, Supun and Lee, Hyungro and Mitchell, Jerome and Caldwell, Rebecca and Bullock, Gina and Hayden, Linda},
 booktitle = {EduHPC'17 workshop at SC17}
}

@techreport{
 title = {Harp-DAAL: A High Performance Data-Intensive Machine Learning Framework},
 type = {techreport},
 year = {2017},
 id = {08a8ae71-7156-39f8-b2b4-6b8ce5855c75},
 created = {2017-11-28T17:33:41.161Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.764Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2017a},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Chen, Langshi and Qiu, Judy}
}

@article{
 title = {Parallelizing Big Data Machine Learning Applications with Model Rotation},
 type = {article},
 year = {2017},
 id = {656d218a-533a-35d4-800d-3b1cec7bb43a},
 created = {2017-11-28T17:33:41.377Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.709Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2017a},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Zhang, Bingjing and Peng, Bo and Qiu, Judy},
 journal = {ser. Advances in Parallel Computing}
}

@article{
 title = {A Survey: Runtime Software Systems for High Performance Computing},
 type = {article},
 year = {2017},
 pages = {48-68},
 volume = {4},
 id = {28e95f87-2747-3f11-ae8a-60325de7080e},
 created = {2017-12-18T17:36:21.043Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-07-12T19:57:41.029Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Sterling2017},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Sterling, Thomas and Anderson, Matthew and Brodowicz, Maciej},
 journal = {Supercomputing Frontiers and Innovations},
 number = {1}
}

@article{
 title = {Simultac Fonton: A Fine-Grain Architecture for Extreme Performance beyond Moore's Law},
 type = {article},
 year = {2017},
 pages = {27-37},
 volume = {4},
 id = {e05c2ac9-1a09-3e21-8f7a-cdb324051694},
 created = {2017-12-18T17:36:21.249Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-07-12T19:57:39.337Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Brodowicz2017},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Brodowicz, Maciej and Sterling, Thomas},
 journal = {Supercomputing Frontiers and Innovations},
 number = {2}
}

@article{
 title = {Asymptotic Computing–Undoing the Damage},
 type = {article},
 year = {2017},
 pages = {55},
 volume = {30},
 publisher = {IOS Press},
 id = {80f5bea3-4cb2-3695-b850-a8a86ea9a2b3},
 created = {2017-12-18T17:36:21.346Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-07-12T19:57:38.402Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {STERLING2017},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {STERLING, Thomas and BRODOWICZ, Maciej and KOGLER, Danny},
 journal = {New Frontiers in High Performance Computing and Big Data}
}

@article{
 title = {Declarative Guide Creation},
 type = {article},
 year = {2017},
 pages = {22-33},
 volume = {2017},
 publisher = {Society for Imaging Science and Technology},
 id = {fd5496c1-a73f-3d5f-b7a6-f2b80a22fea5},
 created = {2017-12-18T18:01:20.636Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-07-12T19:57:46.397Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Cottam2017},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Cottam, Joseph A and Lumsdaine, Andrew},
 journal = {Electronic Imaging},
 number = {1}
}

@article{
 title = {High-throughput cancer hypothesis testing with an integrated PhysiCell-EMEWS workflow},
 type = {article},
 year = {2017},
 pages = {196709},
 publisher = {Cold Spring Harbor Laboratory},
 id = {9271829b-299f-3969-ac33-7186870e10fb},
 created = {2017-12-18T21:43:37.137Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.292Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ozik2017},
 source_type = {JOUR},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Ozik, Jonathan and Collier, Nicholson and Wozniak, Justin and Macal, Charles and Cockrell, Chase and Friedman, Samuel and Ghaffarizadeh, Ahmadreza and Heiland, Randy and An, Gary and Macklin, Paul},
 journal = {bioRxiv}
}

@techreport{
 title = {QuakeSim lessons for NASA Earth Science sensor webs},
 type = {techreport},
 year = {2017},
 id = {e966334e-82a3-3dca-9d9e-dff6494c2086},
 created = {2019-10-01T17:20:14.333Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:52.289Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {parkerquakesim},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Parker, Jay and Donnellan, Andrea and Pierce, Marlon}
}

@inproceedings{
 title = {Better Data Discoverability in Science Gateways},
 type = {inproceedings},
 year = {2017},
 id = {c372fe4d-fd65-35c1-8753-5ba1b8b00307},
 created = {2019-10-01T17:20:18.278Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:57.123Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {nakandalabetter},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Nakandala, Supun and Pamidighantam, Sudhakar and Marru, Suresh and Pierce, Marlon},
 booktitle = {Science Gateways}
}

@inproceedings{
 title = {Science gateways incubator: Software sustainability meets community needs},
 type = {inproceedings},
 year = {2017},
 keywords = {community,incubator,science gateways bootcamp,science gateways community institute,sustainability},
 pages = {477-485},
 month = {11},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 day = {14},
 id = {0488edce-8667-3714-b349-dfe66492170c},
 created = {2019-10-01T17:20:18.954Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.009Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gesing2017},
 private_publication = {false},
 abstract = {© 2017 IEEE. The main goal of the US Science Gateways Community Institute (SGCI) is to serve science gateways to achieve sustainability and growth. Science gateways allow science and engineering communities to access shared data, software, computing services, instruments, educational materials, and other resources specific to their disciplines. Thus, science gateways are a subgroup of scientific software and the means for addressing software sustainability are also suitable for science gateways and vice versa, e.g., best practices for software engineering. Since science gateways are tailored to specific communities, understanding users' requirements is critical for sustainability. SGCI consists of five service areas that closely interact with each other. The Incubator acknowledges the value of business strategy to inform well-designed science gateways and offers two main types of services: Individualized consultancy, tailored to specific challenges a gateway faces, and the Science Gateways Bootcamp. The cornerstone of the Bootcamp is a one-week onsite intensive workshop where participants create their own roadmap for a sustainable science gateway via sessions with experts, hands-on exercises, and group work. This paper offers an overview of the work of the Incubator and shares lessons learned from the inaugural session of the Bootcamp in April 2017.},
 bibtype = {inproceedings},
 author = {Gesing, Sandra and Zentner, Michael and Casavan, Juliana and Hillery, Betsy and Vorvoreanu, Mihaela and Heiland, Randy and Marru, Suresh and Pierce, Marlon and Mullinix, Nayiri and Maron, Nancy},
 doi = {10.1109/eScience.2017.77},
 booktitle = {Proceedings - 13th IEEE International Conference on eScience, eScience 2017}
}

© 2017 IEEE. The main goal of the US Science Gateways Community Institute (SGCI) is to serve science gateways to achieve sustainability and growth. Science gateways allow science and engineering communities to access shared data, software, computing services, instruments, educational materials, and other resources specific to their disciplines. Thus, science gateways are a subgroup of scientific software and the means for addressing software sustainability are also suitable for science gateways and vice versa, e.g., best practices for software engineering. Since science gateways are tailored to specific communities, understanding users' requirements is critical for sustainability. SGCI consists of five service areas that closely interact with each other. The Incubator acknowledges the value of business strategy to inform well-designed science gateways and offers two main types of services: Individualized consultancy, tailored to specific challenges a gateway faces, and the Science Gateways Bootcamp. The cornerstone of the Bootcamp is a one-week onsite intensive workshop where participants create their own roadmap for a sustainable science gateway via sessions with experts, hands-on exercises, and group work. This paper offers an overview of the work of the Incubator and shares lessons learned from the inaugural session of the Bootcamp in April 2017.

@inproceedings{
 title = {The PHASTA Science Gateway},
 type = {inproceedings},
 year = {2017},
 keywords = {ACM proceedings,Apache Airavata,LATEX,Science Gateway,Text tagging},
 pages = {1-4},
 volume = {Part F1287},
 websites = {http://dl.acm.org/citation.cfm?doid=3093338.3104151},
 month = {7},
 publisher = {ACM Press},
 day = {9},
 city = {New York, New York, USA},
 id = {694f8bb7-9ee8-30ae-9c01-cc454dd2479e},
 created = {2019-10-01T17:20:22.510Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.797Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Smith2017},
 private_publication = {false},
 abstract = {The Parallel Hierarchic Adaptive Stabilized Transient Analysis (PHASTA) software supports modeling compressible or incompressible, laminar or turbulent, steady or unsteady flows in 3D using unstructured grids. PHASTA, coupled with the Parallel Unstructured Mesh Infrastructure (PUMI), supports parallel, automated, adaptive simulation workflows. Researchers can easily execute these workflows on the TACC Stampede Xeon and Knights Landing nodes without being burdened by the details of each system using the PHASTA science gateway (created with Apache Airavata). In addition to abstracting away job execution and filesystem details, the gateway creates a searchable archive of past jobs to support reproducibility. Our poster presents the construction of the PHASTA gateway, the workflows it currently supports, and our ongoing efforts to expand functionality and the user base.},
 bibtype = {inproceedings},
 author = {Smith, Cameron W. and Abeysinghe, Eroma},
 doi = {10.1145/3093338.3104151},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact - PEARC17}
}

The Parallel Hierarchic Adaptive Stabilized Transient Analysis (PHASTA) software supports modeling compressible or incompressible, laminar or turbulent, steady or unsteady flows in 3D using unstructured grids. PHASTA, coupled with the Parallel Unstructured Mesh Infrastructure (PUMI), supports parallel, automated, adaptive simulation workflows. Researchers can easily execute these workflows on the TACC Stampede Xeon and Knights Landing nodes without being burdened by the details of each system using the PHASTA science gateway (created with Apache Airavata). In addition to abstracting away job execution and filesystem details, the gateway creates a searchable archive of past jobs to support reproducibility. Our poster presents the construction of the PHASTA gateway, the workflows it currently supports, and our ongoing efforts to expand functionality and the user base.

@article{
 title = {Diversity and universality of endosymbiotic Rickettsia in the fish parasite Ichthyophthirius multifiliis},
 type = {article},
 year = {2017},
 keywords = {Article,Bayesian learning,biodiversity,bootstra},
 pages = {189},
 volume = {8},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85014553705&doi=10.3389%2Ffmicb.2017.00189&partnerID=40&md5=f5b72ffd87ab95121f40b8b97320b41f},
 publisher = {Frontiers Research Foundation},
 id = {f8040c08-6492-3da2-8181-4b167838d64d},
 created = {2019-10-01T17:20:30.660Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:30.660Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zaila2017},
 source_type = {article},
 notes = {cited By 1},
 private_publication = {false},
 abstract = {Although the presence of endosymbiotic rickettsial bacteria, specifically Candidatus Megaira, has been reported in diverse habitats and a wide range of eukaryotic hosts, it remains unclear how broadly Ca. Megaira are distributed in a single host species. In this study we seek to address whether Ca. Megaira are present in most, if not all isolates, of the parasitic ciliate Ichthyophthirius multifiliis. Conserved regions of bacterial 16S rRNA genes were either PCR amplified, or assembled from deep sequencing data, from 18 isolates/populations of I. multifiliis sampled worldwide (Brazil, Taiwan, and USA). We found that rickettsial rRNA sequences belonging to three out of four Ca. Megaira subclades could be consistently detected in all I. multifiliis samples. I. multifiliis collected from local fish farms tend to be inhabited by the same subclade of Ca. Megaira, whereas those derived from pet fish are often inhabited by more than one subclade of Ca. Megaira. Distributions of Ca. Megaira in I. multifiliis thus better reflect the travel history, but not the phylogeny, of I. multifiliis. In summary, our results suggest that I. multifiliis may be dependent on this endosymbiotic relationship, and the association between Ca. Megaira and I. multifiliis is more diverse than previously thought. © 2017 Zaila, Doak, Ellerbrock, Tung, Martins, Kolbin, Yao, Cassidy-Hanley, Clark and Chang.},
 bibtype = {article},
 author = {Zaila, K E and Doak, T G and Ellerbrock, H and Tung, C.-H. and Martins, M L and Kolbin, D and Yao, M.-C. and Cassidy-Hanley, D M and Clark, T G and Chang, W.-J.},
 doi = {10.3389/fmicb.2017.00189},
 journal = {Frontiers in Microbiology},
 number = {FEB}
}

Although the presence of endosymbiotic rickettsial bacteria, specifically Candidatus Megaira, has been reported in diverse habitats and a wide range of eukaryotic hosts, it remains unclear how broadly Ca. Megaira are distributed in a single host species. In this study we seek to address whether Ca. Megaira are present in most, if not all isolates, of the parasitic ciliate Ichthyophthirius multifiliis. Conserved regions of bacterial 16S rRNA genes were either PCR amplified, or assembled from deep sequencing data, from 18 isolates/populations of I. multifiliis sampled worldwide (Brazil, Taiwan, and USA). We found that rickettsial rRNA sequences belonging to three out of four Ca. Megaira subclades could be consistently detected in all I. multifiliis samples. I. multifiliis collected from local fish farms tend to be inhabited by the same subclade of Ca. Megaira, whereas those derived from pet fish are often inhabited by more than one subclade of Ca. Megaira. Distributions of Ca. Megaira in I. multifiliis thus better reflect the travel history, but not the phylogeny, of I. multifiliis. In summary, our results suggest that I. multifiliis may be dependent on this endosymbiotic relationship, and the association between Ca. Megaira and I. multifiliis is more diverse than previously thought. © 2017 Zaila, Doak, Ellerbrock, Tung, Martins, Kolbin, Yao, Cassidy-Hanley, Clark and Chang.

@article{
 title = {Investigating biogeographic boundaries of the Sunda shelf: A phylogenetic analysis of two island populations of Macaca fascicularis},
 type = {article},
 year = {2017},
 websites = {https://onlinelibrary.wiley.com/doi/full/10.1002/ajpa.23235},
 id = {93dc61cf-6aaa-38fe-85e8-17161b9b2f92},
 created = {2019-10-01T17:20:31.791Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:31.791Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Klegarth2017},
 source_type = {JOUR},
 private_publication = {false},
 bibtype = {article},
 author = {Klegarth, A R and Sanders, S A and Gloss, A D and Lane‐deGraaf, K E and Jones‐Engel, L and Fuentes, A and Hollocher, H},
 doi = {10.1002/ajpa.23235},
 journal = {American Journal of Physical Anthropology},
 keywords = {PY7}
}

@article{
 title = {Population Genomics of Paramecium Species},
 type = {article},
 year = {2017},
 keywords = {Animals; DNA,DNA flanking region; genetic polymorphism; genome;,Mitochondrial; Evolution,Molecular; Genetic Variation; Genome,Protozoan; Genomics; Metagenomics; Mutation; Para,mitochondrial DNA},
 pages = {1194-1216},
 volume = {34},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85019092588&doi=10.1093%2Fmolbev%2Fmsx074&partnerID=40&md5=3f72c45c5cb74e418d2e45c4368ebd32},
 publisher = {Oxford University Press},
 id = {a83d7b81-3841-31dd-95ee-99915fd3b02b},
 created = {2019-10-01T17:20:32.195Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:32.195Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Johri20171194},
 source_type = {article},
 notes = {cited By 0},
 private_publication = {false},
 abstract = {Population-genomic analyses are essential to understanding factors shaping genomic variation and lineage-specific sequence constraints. The dearth of such analyses for unicellular eukaryotes prompted us to assess genomic variation in Paramecium, one of the most well-studied ciliate genera. The Paramecium aurelia complex consists of ∼15 morphologically indistinguishable species that diverged subsequent to two rounds of whole-genome duplications (WGDs, as long as 320 MYA) and possess extremely streamlined genomes. We examine patterns of both nuclear and mitochondrial polymorphism, by sequencing whole genomes of 10-13 worldwide isolates of each of three species belonging to the P. aurelia complex: P. tetraurelia, P. biaurelia, P. sexaurelia, as well as two outgroup species that do not share the WGDs: P. caudatum and P. multimicronucleatum. An apparent absence of global geographic population structure suggests continuous or recent dispersal of Paramecium over long distances. Intergenic regions are highly constrained relative to coding sequences, especially in P. caudatum and P. multimicronucleatum that have shorter intergenic distances. Sequence diversity and divergence are reduced up to ∼100-150 bp both upstream and downstream of genes, suggesting strong constraints imposed by the presence of densely packed regulatory modules. In addition, comparison of sequence variation at non-synonymous and synonymous sites suggests similar recent selective pressures on paralogs within and orthologs across the deeply diverging species. This study presents the first genome-wide population-genomic analysis in ciliates and provides a valuable resource for future studies in evolutionary and functional genetics in Paramecium. © 2017 Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. All rights reserved.},
 bibtype = {article},
 author = {Johri, P and Krenek, S and Marinov, G K and Doak, T G and Berendonk, T U and Lynch, M},
 doi = {10.1093/molbev/msx074},
 journal = {Molecular Biology and Evolution},
 number = {5}
}

Population-genomic analyses are essential to understanding factors shaping genomic variation and lineage-specific sequence constraints. The dearth of such analyses for unicellular eukaryotes prompted us to assess genomic variation in Paramecium, one of the most well-studied ciliate genera. The Paramecium aurelia complex consists of ∼15 morphologically indistinguishable species that diverged subsequent to two rounds of whole-genome duplications (WGDs, as long as 320 MYA) and possess extremely streamlined genomes. We examine patterns of both nuclear and mitochondrial polymorphism, by sequencing whole genomes of 10-13 worldwide isolates of each of three species belonging to the P. aurelia complex: P. tetraurelia, P. biaurelia, P. sexaurelia, as well as two outgroup species that do not share the WGDs: P. caudatum and P. multimicronucleatum. An apparent absence of global geographic population structure suggests continuous or recent dispersal of Paramecium over long distances. Intergenic regions are highly constrained relative to coding sequences, especially in P. caudatum and P. multimicronucleatum that have shorter intergenic distances. Sequence diversity and divergence are reduced up to ∼100-150 bp both upstream and downstream of genes, suggesting strong constraints imposed by the presence of densely packed regulatory modules. In addition, comparison of sequence variation at non-synonymous and synonymous sites suggests similar recent selective pressures on paralogs within and orthologs across the deeply diverging species. This study presents the first genome-wide population-genomic analysis in ciliates and provides a valuable resource for future studies in evolutionary and functional genetics in Paramecium. © 2017 Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. All rights reserved.

@article{
 title = {Optimizing and evaluating the reconstruction of Metagenome-assembled microbial genomes},
 type = {article},
 year = {2017},
 volume = {18},
 websites = {https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-017-4294-1},
 id = {294679cf-295f-38d2-bab3-7583c907ee17},
 created = {2019-10-01T17:20:32.305Z},
 accessed = {2019-08-09},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.946Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Papudeshi2017},
 private_publication = {false},
 abstract = {Background: Microbiome/host interactions describe characteristics that affect the host's health. Shotgun metagenomics includes sequencing a random subset of the microbiome to analyze its taxonomic and metabolic potential. Reconstruction of DNA fragments into genomes from metagenomes (called metagenome-assembled genomes) assigns unknown fragments to taxa/function and facilitates discovery of novel organisms. Genome reconstruction incorporates sequence assembly and sorting of assembled sequences into bins, characteristic of a genome. However, the microbial community composition, including taxonomic and phylogenetic diversity may influence genome reconstruction. We determine the optimal reconstruction method for four microbiome projects that had variable sequencing platforms (IonTorrent and Illumina), diversity (high or low), and environment (coral reefs and kelp forests), using a set of parameters to select for optimal assembly and binning tools. Methods: We tested the effects of the assembly and binning processes on population genome reconstruction using 105 marine metagenomes from 4 projects. Reconstructed genomes were obtained from each project using 3 assemblers (IDBA, MetaVelvet, and SPAdes) and 2 binning tools (GroopM and MetaBat). We assessed the efficiency of assemblers using statistics that including contig continuity and contig chimerism and the effectiveness of binning tools using genome completeness and taxonomic identification. Results: We concluded that SPAdes, assembled more contigs (143,718 ± 124 contigs) of longer length (N50 = 1632 ± 108 bp), and incorporated the most sequences (sequences-assembled = 19.65%). The microbial richness and evenness were maintained across the assembly, suggesting low contig chimeras. SPAdes assembly was responsive to the biological and technological variations within the project, compared with other assemblers. Among binning tools, we conclude that MetaBat produced bins with less variation in GC content (average standard deviation: 1.49), low species richness (4.91 ± 0.66), and higher genome completeness (40.92 ± 1.75) across all projects. MetaBat extracted 115 bins from the 4 projects of which 66 bins were identified as reconstructed metagenome-assembled genomes with sequences belonging to a specific genus. We identified 13 novel genomes, some of which were 100% complete, but show low similarity to genomes within databases. Conclusions: In conclusion, we present a set of biologically relevant parameters for evaluation to select for optimal assembly and binning tools. For the tools we tested, SPAdes assembler and MetaBat binning tools reconstructed quality metagenome-assembled genomes for the four projects. We also conclude that metagenomes from microbial communities that have high coverage of phylogenetically distinct, and low taxonomic diversity results in highest quality metagenome-assembled genomes.},
 bibtype = {article},
 author = {Papudeshi, Bhavya and Haggerty, J. Matthew and Doane, Michael and Morris, Megan M. and Walsh, Kevin and Beattie, Douglas T. and Pande, Dnyanada and Zaeri, Parisa and Silva, Genivaldo G.Z. and Thompson, Fabiano and Edwards, Robert A. and Dinsdale, Elizabeth A.},
 doi = {10.1186/s12864-017-4294-1},
 journal = {BMC Genomics},
 number = {1}
}

Background: Microbiome/host interactions describe characteristics that affect the host's health. Shotgun metagenomics includes sequencing a random subset of the microbiome to analyze its taxonomic and metabolic potential. Reconstruction of DNA fragments into genomes from metagenomes (called metagenome-assembled genomes) assigns unknown fragments to taxa/function and facilitates discovery of novel organisms. Genome reconstruction incorporates sequence assembly and sorting of assembled sequences into bins, characteristic of a genome. However, the microbial community composition, including taxonomic and phylogenetic diversity may influence genome reconstruction. We determine the optimal reconstruction method for four microbiome projects that had variable sequencing platforms (IonTorrent and Illumina), diversity (high or low), and environment (coral reefs and kelp forests), using a set of parameters to select for optimal assembly and binning tools. Methods: We tested the effects of the assembly and binning processes on population genome reconstruction using 105 marine metagenomes from 4 projects. Reconstructed genomes were obtained from each project using 3 assemblers (IDBA, MetaVelvet, and SPAdes) and 2 binning tools (GroopM and MetaBat). We assessed the efficiency of assemblers using statistics that including contig continuity and contig chimerism and the effectiveness of binning tools using genome completeness and taxonomic identification. Results: We concluded that SPAdes, assembled more contigs (143,718 ± 124 contigs) of longer length (N50 = 1632 ± 108 bp), and incorporated the most sequences (sequences-assembled = 19.65%). The microbial richness and evenness were maintained across the assembly, suggesting low contig chimeras. SPAdes assembly was responsive to the biological and technological variations within the project, compared with other assemblers. Among binning tools, we conclude that MetaBat produced bins with less variation in GC content (average standard deviation: 1.49), low species richness (4.91 ± 0.66), and higher genome completeness (40.92 ± 1.75) across all projects. MetaBat extracted 115 bins from the 4 projects of which 66 bins were identified as reconstructed metagenome-assembled genomes with sequences belonging to a specific genus. We identified 13 novel genomes, some of which were 100% complete, but show low similarity to genomes within databases. Conclusions: In conclusion, we present a set of biologically relevant parameters for evaluation to select for optimal assembly and binning tools. For the tools we tested, SPAdes assembler and MetaBat binning tools reconstructed quality metagenome-assembled genomes for the four projects. We also conclude that metagenomes from microbial communities that have high coverage of phylogenetically distinct, and low taxonomic diversity results in highest quality metagenome-assembled genomes.

@article{
 title = {Estimating seven coefficients of pairwise relatedness using population-genomic data},
 type = {article},
 year = {2017},
 keywords = {Alleles; Animals; Computer Simulation; Daphnia; G,Genetic,Population; Genotype; Heterozygote; Inbreeding; M,animal; Article; consanguinity; Daphnia pulex; gen},
 pages = {105-118},
 volume = {206},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020718072&doi=10.1534%2Fgenetics.116.190660&partnerID=40&md5=74072b13c07af013108eb10cbe219fac},
 publisher = {Genetics},
 id = {478dff57-b5e6-31cc-b469-601d95d86338},
 created = {2019-10-01T17:20:32.463Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:32.463Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ackerman2017105},
 source_type = {article},
 notes = {cited By 0},
 private_publication = {false},
 abstract = {Population structure can be described by genotypic-correlation coefficients between groups of individuals, the most basic of which are the pairwise relatedness coefficients between any two individuals. There are nine pairwise relatedness coefficients in the most general model, and we show that these can be reduced to seven coefficients for biallelic loci. Although all nine coefficients can be estimated from pedigrees, six coefficients have been beyond empirical reach. We provide a numerical optimization procedure that estimates all seven reduced coefficients from population-genomic data. Simulations show that the procedure is nearly unbiased, even at 33 coverage, and errors in five of the seven coefficients are statistically uncorrelated. The remaining two coefficients have a negative correlation of errors, but their sum provides an unbiased assessment of the overall correlation of heterozygosity between two individuals. Application of these new methods to four populations of the freshwater crustacean Daphnia pulex reveal the occurrence of half siblings in our samples, as well as a number of identical individuals that are likely obligately asexual clone mates. Statistically significant negative estimates of these pairwise relatedness coefficients, including inbreeding coefficients that were typically negative, underscore the difficulties that arise when interpreting genotypic correlations as estimations of the probability that alleles are identical by descent. © 2017 by the Genetics Society of America.},
 bibtype = {article},
 author = {Ackerman, M S and Johri, P and Spitze, K and Xu, S and Doak, T G and Young, K and Lynch, M},
 doi = {10.1534/genetics.116.190660},
 journal = {Genetics},
 number = {1}
}

Population structure can be described by genotypic-correlation coefficients between groups of individuals, the most basic of which are the pairwise relatedness coefficients between any two individuals. There are nine pairwise relatedness coefficients in the most general model, and we show that these can be reduced to seven coefficients for biallelic loci. Although all nine coefficients can be estimated from pedigrees, six coefficients have been beyond empirical reach. We provide a numerical optimization procedure that estimates all seven reduced coefficients from population-genomic data. Simulations show that the procedure is nearly unbiased, even at 33 coverage, and errors in five of the seven coefficients are statistically uncorrelated. The remaining two coefficients have a negative correlation of errors, but their sum provides an unbiased assessment of the overall correlation of heterozygosity between two individuals. Application of these new methods to four populations of the freshwater crustacean Daphnia pulex reveal the occurrence of half siblings in our samples, as well as a number of identical individuals that are likely obligately asexual clone mates. Statistically significant negative estimates of these pairwise relatedness coefficients, including inbreeding coefficients that were typically negative, underscore the difficulties that arise when interpreting genotypic correlations as estimations of the probability that alleles are identical by descent. © 2017 by the Genetics Society of America.

@inproceedings{
 title = {A voice for bioinformatics},
 type = {inproceedings},
 year = {2017},
 keywords = {Bioinformatics,Command line,Computer applications,Computer programming,Data movements,Ge,Genome analysis},
 volume = {Part F1287},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85025816701&doi=10.1145%2F3093338.3093374&partnerID=40&md5=f475d6d55b3e3145ad7094dd9005a1f2},
 publisher = {Association for Computing Machinery},
 id = {102d1e0f-27b7-372a-aea4-d3f739a0d1f6},
 created = {2019-10-01T17:20:32.725Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:32.725Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ganote2017},
 source_type = {conference},
 notes = {cited By 0; Conference of 2017 Practice and Experience in Advanced Research Computing, PEARC 2017 ; Conference Date: 9 July 2017 Through 13 July 2017; Conference Code:128771},
 private_publication = {false},
 abstract = {One of the challenges to adoption of HPC is the disjunction between those who need it and those who know it. Biology (specifically, genomics) is a growing field for computational use, but the typical biologist does not have an established informatics background. The National Center for Genome Analysis Support (NCGAS) AIDS users in getting past the initial shock of the command line and guides them toward savvy cluster use. NCGAS is initiating a push to become domain champions alongside Oklahoma State's Brian Cougar. Our position at IU gives us a close relationship with XSEDE and we already fulfill a role in pushing users toward XSEDE resources when our local clusters are ill-suited to the job. We currently act as liaison between biologists and Jetstream, IU and TACC's research computing cloud. Typical issues include: Software installation; Software usage-what parameters do I choose, and how do I interpret the results; Batch job submission; Understanding how queues and job handlers work; Data movement, Spinning up VMs on Jetstream We will discuss how we have structured our support, and illustrate our impact on XSEDE resources. © 2017 ACM.},
 bibtype = {inproceedings},
 author = {Ganote, C L and Sanders, S A and Papudeshi, B N and Blood, P D and Doak, T G},
 doi = {10.1145/3093338.3093374},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

One of the challenges to adoption of HPC is the disjunction between those who need it and those who know it. Biology (specifically, genomics) is a growing field for computational use, but the typical biologist does not have an established informatics background. The National Center for Genome Analysis Support (NCGAS) AIDS users in getting past the initial shock of the command line and guides them toward savvy cluster use. NCGAS is initiating a push to become domain champions alongside Oklahoma State's Brian Cougar. Our position at IU gives us a close relationship with XSEDE and we already fulfill a role in pushing users toward XSEDE resources when our local clusters are ill-suited to the job. We currently act as liaison between biologists and Jetstream, IU and TACC's research computing cloud. Typical issues include: Software installation; Software usage-what parameters do I choose, and how do I interpret the results; Batch job submission; Understanding how queues and job handlers work; Data movement, Spinning up VMs on Jetstream We will discuss how we have structured our support, and illustrate our impact on XSEDE resources. © 2017 ACM.

@inproceedings{
 title = {Scalable photogrammetry with high performance computing},
 type = {inproceedings},
 year = {2017},
 keywords = {Benchmarking,Compute resources,Cultural heritages,Graduate s,Photogrammetry,Students},
 pages = {3},
 volume = {Part F1287},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85025826260&doi=10.1145%2F3093338.3104174&partnerID=40&md5=3d7c365c1289d519e724ab1c35dfad07},
 publisher = {Association for Computing Machinery},
 id = {8658fa51-0528-3810-9931-3d86567e12c6},
 created = {2019-10-01T17:20:36.946Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:26.513Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gniady2017},
 source_type = {conference},
 notes = {cited By 0; Conference of 2017 Practice and Experience in Advanced Research Computing, PEARC 2017 ; Conference Date: 9 July 2017 Through 13 July 2017; Conference Code:128771},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {Photogrammetry is used to build 3-dimensional models of everything from terrains to ancient statues. In the past, the stitching process was done on powerful PCs and could take weeks for large datasets. Even relatively small objects often required several hours to stitch together. With the availability of parallel processing options in the latest release of Agisoft PhotoScan, it is possible to leverage the power of high performance computing on large datasets. This poster presents the results of benchmarking tests for three datasets processed at two different model quality levels, medium and high (there are four times more points in the dense point cloud at the high setting) using 2, 4, 8, and 16 nodes. The purpose of the benchmarking is to determine how to optimize software license usage and compute resources against time and output quality. The poster also details the matrix of user-specified parameters that have been built into the python script that submits the parallel jobs. These parameters have evolved through the assessment of needs of users who are using the HPC deployment of photogrammetry as a service.We are excited by the uptake of this new service around campus in different fields across multiple disciplines. A group of cultural heritage experts will be using the service from Italy this summer, and a graduate student in anthropology will be stitching aerial data sets from Mexico. © 2017 Copyright held by the owner/author(s).},
 bibtype = {inproceedings},
 author = {Gniady, T and Ruan, G and Sherman, W and Tuna, E and Wernert, E},
 doi = {10.1145/3093338.3104174},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

Photogrammetry is used to build 3-dimensional models of everything from terrains to ancient statues. In the past, the stitching process was done on powerful PCs and could take weeks for large datasets. Even relatively small objects often required several hours to stitch together. With the availability of parallel processing options in the latest release of Agisoft PhotoScan, it is possible to leverage the power of high performance computing on large datasets. This poster presents the results of benchmarking tests for three datasets processed at two different model quality levels, medium and high (there are four times more points in the dense point cloud at the high setting) using 2, 4, 8, and 16 nodes. The purpose of the benchmarking is to determine how to optimize software license usage and compute resources against time and output quality. The poster also details the matrix of user-specified parameters that have been built into the python script that submits the parallel jobs. These parameters have evolved through the assessment of needs of users who are using the HPC deployment of photogrammetry as a service.We are excited by the uptake of this new service around campus in different fields across multiple disciplines. A group of cultural heritage experts will be using the service from Italy this summer, and a graduate student in anthropology will be stitching aerial data sets from Mexico. © 2017 Copyright held by the owner/author(s).

@inproceedings{
 title = {Jetstream: A Cloud System Enabling Learning in Higher Education Communities},
 type = {inproceedings},
 year = {2017},
 keywords = {Atmosphere,Cloud,Cyberinfras-tructure,Digital,EOT,Education,Globus,Jetstream,Openstack,Outreach,Research,Training,XD,XSEDE},
 pages = {67-72},
 volume = {Part F1317},
 websites = {http://dl.acm.org/citation.cfm?doid=3123458.3123466,https://dl.acm.org/doi/10.1145/3123458.3123466},
 month = {10},
 publisher = {ACM},
 day = {1},
 city = {New York, NY, USA},
 id = {11e0aced-58d1-301f-9756-abea7781f3f1},
 created = {2019-10-01T17:20:37.392Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:46:04.577Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fischer2017},
 source_type = {CONF},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 Copyright held by the owner/author(s). Jetstream is the frst production cloud funded by the NSF for conducting general-purpose science and engineering research as well as an easy-to-use platform for education activities. Unlike many high-performance computing systems, Jetstream uses the interactive Atmosphere graphical user interface developed as part of the iPlant (now CyVerse) project and focuses on interactive use on uniprocessor or multiprocessor. This interface provides for a lower barrier of entry for use by educators, students, practicing scientists, and engineers. A key part of Jetstream's mission is to extend the reach of the NSF's eXtreme Digital (XD) program to a community of users who have not previously utilized NSF XD program resources, including those communities and institutions that traditionally lack signifcant cyberinfrastructure resources. One manner in which Jetstream eases this access is via virtual desktops facilitating use in education and research at small colleges and universities, including Historically Black Colleges and Universities (HBCUs), Minority Serving Institutions (MSIs), Tribal colleges, and higher education institutions in states designated by the NSF as eligible for funding via the Experimental Program to Stimulate Competitive Research (EPSCoR). Jetstream entered into full production in September 2016 and during the frst six months it has supported more than a dozen educational eforts across the United States. Here, we discuss how educators at institutions of higher education have been using Jetstream in the classroom and at student-focused workshops. Specifcally, we explore success stories, difculties encountered, and everything in between. We also discuss plans for increasing the use of cloud-based systems in higher education. A primary goal in this paper is to spark discussions between educators and information technologists on how to improve using cloud resources in education.},
 bibtype = {inproceedings},
 author = {Fischer, Jeremy and Hancock, David Y. and Lowe, John Michael and Turner, George and Snapp-Childs, Winona and Stewart, Craig A.},
 doi = {10.1145/3123458.3123466},
 booktitle = {Proceedings of the 2017 ACM SIGUCCS Annual Conference}
}

© 2017 Copyright held by the owner/author(s). Jetstream is the frst production cloud funded by the NSF for conducting general-purpose science and engineering research as well as an easy-to-use platform for education activities. Unlike many high-performance computing systems, Jetstream uses the interactive Atmosphere graphical user interface developed as part of the iPlant (now CyVerse) project and focuses on interactive use on uniprocessor or multiprocessor. This interface provides for a lower barrier of entry for use by educators, students, practicing scientists, and engineers. A key part of Jetstream's mission is to extend the reach of the NSF's eXtreme Digital (XD) program to a community of users who have not previously utilized NSF XD program resources, including those communities and institutions that traditionally lack signifcant cyberinfrastructure resources. One manner in which Jetstream eases this access is via virtual desktops facilitating use in education and research at small colleges and universities, including Historically Black Colleges and Universities (HBCUs), Minority Serving Institutions (MSIs), Tribal colleges, and higher education institutions in states designated by the NSF as eligible for funding via the Experimental Program to Stimulate Competitive Research (EPSCoR). Jetstream entered into full production in September 2016 and during the frst six months it has supported more than a dozen educational eforts across the United States. Here, we discuss how educators at institutions of higher education have been using Jetstream in the classroom and at student-focused workshops. Specifcally, we explore success stories, difculties encountered, and everything in between. We also discuss plans for increasing the use of cloud-based systems in higher education. A primary goal in this paper is to spark discussions between educators and information technologists on how to improve using cloud resources in education.

@inproceedings{
 title = {Performance Benchmarking of the R Programming Environment on the Stampede 1 . 5 Supercomputer},
 type = {inproceedings},
 year = {2017},
 keywords = {2017,acm reference format,benchmarking,james r,many-core,mccombs and scott michael,performance benchmarking of,r,scalability,xeon phi,xsede},
 pages = {8},
 volume = {Part F1287},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85025810544&doi=10.1145%2F3093338.3093346&partnerID=40&md5=38a5e9f4f39737f05a2dff38119eff0b},
 publisher = {Association for Computing Machinery},
 id = {885f4790-3653-3b7f-b11f-53832ae6de98},
 created = {2019-10-01T17:20:37.438Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:03.078Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {McCombs2017},
 source_type = {conference},
 notes = {cited By 0; Conference of 2017 Practice and Experience in Advanced Research Computing, PEARC 2017 ; Conference Date: 9 July 2017 Through 13 July 2017; Conference Code:128771},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {We present performance results obtained with a new single-node performance benchmark of the R programming environment on the many-core Xeon Phi Knights Landing and standard Xeonbased compute nodes of the Stampede supercomputer cluster at the Texas Advanced Computing Center. The benchmark consists of microbenchmarks of linear algebra kernels and machine learning functionality that includes clustering and neural network training from the R distribution. The standard Xeon-based nodes outperformed their Xeon Phi counterparts for matrices of small to medium dimensions, performing approximately twice as fast for most of the linear algebra microbenchmarks. For matrices of medium to large dimensions, the Knights Landing nodes were competitive with or outperformed the standard Xeon-based nodes with most of the linear algebra microbenchmarks, executing as much as five times faster than the standard Xeon-based nodes. For the clustering and neural network training microbenchmarks, the standard Xeonbased nodes performed up to four times faster than their Xeon Phi counterparts for many large data sets, indicating that commonly used R packages may need to be reengineered to take advantage of existing optimized, scalable kernels. © 2017 Association for Computing Machinery.},
 bibtype = {inproceedings},
 author = {Mccombs, James R and Michael, Scott},
 doi = {10.1145/3093338.3093346},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

We present performance results obtained with a new single-node performance benchmark of the R programming environment on the many-core Xeon Phi Knights Landing and standard Xeonbased compute nodes of the Stampede supercomputer cluster at the Texas Advanced Computing Center. The benchmark consists of microbenchmarks of linear algebra kernels and machine learning functionality that includes clustering and neural network training from the R distribution. The standard Xeon-based nodes outperformed their Xeon Phi counterparts for matrices of small to medium dimensions, performing approximately twice as fast for most of the linear algebra microbenchmarks. For matrices of medium to large dimensions, the Knights Landing nodes were competitive with or outperformed the standard Xeon-based nodes with most of the linear algebra microbenchmarks, executing as much as five times faster than the standard Xeon-based nodes. For the clustering and neural network training microbenchmarks, the standard Xeonbased nodes performed up to four times faster than their Xeon Phi counterparts for many large data sets, indicating that commonly used R packages may need to be reengineered to take advantage of existing optimized, scalable kernels. © 2017 Association for Computing Machinery.

@inproceedings{
 title = {High performance computing enabled simulation of the food-water-energy system},
 type = {inproceedings},
 year = {2017},
 keywords = {Agricultural machinery,Agriculture,Agro ecosystems,Agro-IBIS,Benchmarking,Climate model,Computer clusters,Fi},
 pages = {10},
 volume = {Part F1287},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85025829960&doi=10.1145%2F3093338.3093381&partnerID=40&md5=69d8a119998a6e0aac5a0a4ac12415ad},
 publisher = {Association for Computing Machinery},
 id = {daec685b-ff26-3d1e-8dee-69e36ead6ddf},
 created = {2019-10-01T17:20:37.493Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:01.286Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Dennis2017},
 source_type = {conference},
 notes = {cited By 0; Conference of 2017 Practice and Experience in Advanced Research Computing, PEARC 2017 ; Conference Date: 9 July 2017 Through 13 July 2017; Conference Code:128771},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {Domain science experts are commonly limited by computational efficiency of their code and hardware resources available for execution of desired simulations. Here, we detail a collaboration between domain scientists focused on simulating an ensemble of climate and human management decisions to drive environmental (e.g., water quality) and economic (e.g., crop yield) outcomes. Brie.y, the domain scientists developed a message passing interface to execute the formerly serial code across a number of processors, anticipating signi.cant performance improvement by moving to a cluster computing environment from their desktop machines. The code is both too complex to efficiently re-code from scratch and has a shared codebase that must continue to function on desktop machines as well as the parallel implementation. However, ineff-ciencies in the code caused the LUSTRE .lesystem to bo.leneck performance for all users. The domain scientists collaborated with Indiana University's Science Applications and Performance Tuning and High Performance File System teams to address the unforeseen performance limitations. The non-linear process of testing so.ware advances and hardware performance is a model of the failures and successes that can be anticipated in similar applications. Ultimately, through a series of iterative so.ware and hardware advances the team worked collaboratively to increase performance of the code, cluster, and .le system to enable more than 100-fold increases in performance. As a result, the domain science is able to assess ensembles of climate and human forcing on the model, and sensitivities of ecologically and economically important outcomes of intensively managed agricultural landscapes. © 2017 Copyright is held by the owner/author(s).},
 bibtype = {inproceedings},
 author = {Dennis, H E B and Ward, A S and Balson, T and Li, Y and Henschel, R and Slavin, S and Simms, S and Brunst, H},
 doi = {10.1145/3093338.3093381},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

Domain science experts are commonly limited by computational efficiency of their code and hardware resources available for execution of desired simulations. Here, we detail a collaboration between domain scientists focused on simulating an ensemble of climate and human management decisions to drive environmental (e.g., water quality) and economic (e.g., crop yield) outcomes. Brie.y, the domain scientists developed a message passing interface to execute the formerly serial code across a number of processors, anticipating signi.cant performance improvement by moving to a cluster computing environment from their desktop machines. The code is both too complex to efficiently re-code from scratch and has a shared codebase that must continue to function on desktop machines as well as the parallel implementation. However, ineff-ciencies in the code caused the LUSTRE .lesystem to bo.leneck performance for all users. The domain scientists collaborated with Indiana University's Science Applications and Performance Tuning and High Performance File System teams to address the unforeseen performance limitations. The non-linear process of testing so.ware advances and hardware performance is a model of the failures and successes that can be anticipated in similar applications. Ultimately, through a series of iterative so.ware and hardware advances the team worked collaboratively to increase performance of the code, cluster, and .le system to enable more than 100-fold increases in performance. As a result, the domain science is able to assess ensembles of climate and human forcing on the model, and sensitivities of ecologically and economically important outcomes of intensively managed agricultural landscapes. © 2017 Copyright is held by the owner/author(s).

@techreport{
 title = {Summary of the 2017 NCGAS User Survey},
 type = {techreport},
 year = {2017},
 id = {01acc1e9-c4cb-30bc-bc9e-d41e19fd9665},
 created = {2019-10-01T17:20:37.770Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:37.770Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Doak2017},
 source_type = {RPRT},
 private_publication = {false},
 bibtype = {techreport},
 author = {Doak, Thomas G and Stewart, Craig A and Michaels, Scott D}
}

@techreport{
 title = {ABI Sustaining: The National Center for Genome Analysis Support 2017 Annual Report},
 type = {techreport},
 year = {2017},
 websites = {http://hdl.handle.net/2022/21613},
 id = {930a5dba-6576-3905-8826-59ef6f599520},
 created = {2019-10-01T17:20:37.917Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T21:04:41.151Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Doak2017a},
 source_type = {RPRT},
 private_publication = {false},
 bibtype = {techreport},
 author = {Doak, Thomas G and Stewart, Craig A and Michaels, Scott D}
}

@article{
 title = {Forward Observer system for radar data workflows: Big data management in the field},
 type = {article},
 year = {2017},
 keywords = {Big data,Computer clusters,D,Data analysis system,Data handling,Data sto,Data storage equipment,Information management},
 pages = {92-97},
 volume = {76},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020697663&doi=10.1016%2Fj.future.2017.05.031&partnerID=40&md5=a5874af9af295c2eb064bc71abf890b2},
 publisher = {Elsevier B.V.},
 id = {e77abbb3-8aa7-3a9a-b844-d8ed28cac198},
 created = {2019-10-01T17:20:38.538Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:48.698Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper201792},
 source_type = {article},
 notes = {cited By 0},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {There are unique challenges in managing data collection and management from instruments in the field in general. These issues become extreme when “in the field” means “in a plane over the Antarctic”. In this paper we present the design and function of the Forward Observer a computer cluster and data analysis system that flies in a plane in the Arctic and Antarctic to collect, analyze in real time, and store Synthetic Aperture Radar (SAR) data. SAR is used to analyze the thickness and structure of polar ice sheets. We also discuss the processing of data once it is returned to the continental US and made available via data grids. The needs for in-flight data analysis and storage in the Antarctic and Arctic are highly unusual, and we have developed a novel system to meet those needs. We describe the constraints and requirements that led to the creation of this system and the general functionality which it applies to any instrument. We discuss the main means for handling replication and creating checksum information to ensure that data collected in polar regions are returned safely to mainland US for analysis. So far, not a single byte of data collected in the field has failed to make it home to the US for analysis (although many particular data storage devices have failed or been damaged due to the challenges of the extreme environments in which this system is used). While the Forward Observer system is developed for the extreme situation of data management in the field in the Antarctic, the technology and solutions we have developed are applicable and potentially usable in many situations where researchers wish to do real time data management in the field in areas that are constrained in terms of electrical supply. © 2017 Elsevier B.V.},
 bibtype = {article},
 author = {Knepper, R and Standish, M},
 doi = {10.1016/j.future.2017.05.031},
 journal = {Future Generation Computer Systems}
}

There are unique challenges in managing data collection and management from instruments in the field in general. These issues become extreme when “in the field” means “in a plane over the Antarctic”. In this paper we present the design and function of the Forward Observer a computer cluster and data analysis system that flies in a plane in the Arctic and Antarctic to collect, analyze in real time, and store Synthetic Aperture Radar (SAR) data. SAR is used to analyze the thickness and structure of polar ice sheets. We also discuss the processing of data once it is returned to the continental US and made available via data grids. The needs for in-flight data analysis and storage in the Antarctic and Arctic are highly unusual, and we have developed a novel system to meet those needs. We describe the constraints and requirements that led to the creation of this system and the general functionality which it applies to any instrument. We discuss the main means for handling replication and creating checksum information to ensure that data collected in polar regions are returned safely to mainland US for analysis. So far, not a single byte of data collected in the field has failed to make it home to the US for analysis (although many particular data storage devices have failed or been damaged due to the challenges of the extreme environments in which this system is used). While the Forward Observer system is developed for the extreme situation of data management in the field in the Antarctic, the technology and solutions we have developed are applicable and potentially usable in many situations where researchers wish to do real time data management in the field in areas that are constrained in terms of electrical supply. © 2017 Elsevier B.V.

@inproceedings{
 title = {The Community Software Repository from XSEDE: A Resource for the National Research Community},
 type = {inproceedings},
 year = {2017},
 pages = {1-8},
 volume = {Part F1287},
 websites = {http://dl.acm.org/citation.cfm?doid=3093338.3093373},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {9193b6bc-abb1-302c-9362-2a41db606105},
 created = {2019-10-01T17:20:38.917Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:33:20.059Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Navarro2017},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 ACM. The Extreme Science and Engineering Discovery Environment (XSEDE) connects cyberinfrastructure (CI) resources, software, and services. One of XSEDE's primary goals in supporting US research generally is to "advance the ecosystem"-making use of XSEDE's leadership position to create software, tools, and services that lead to an effective and efficient national cyberinfrastructure. Software enables this endeavor in two very distinct ways: enabling the operation of XSEDE as a distributed yet integrated cyberinfrastructure resource; and by providing access to a wide variety of software of value to end user researchers and students, operators of campus cyberinfrastructure resources, and to those considering to propose new cyberinfrastructure resources to the National Science Foundation (NSF). The Community Software Repository (CSR) provides transparency about how XSEDE operates and provides access to software of use and value to the US research community generally. The CSR provides access to use cases that describe needs expressed by the research community, capability delivery plans that describe how XSEDE meets those needs, and the actual software that meets those needs. Software is delivered in a variety of forms and formats. The CSR also includes mechanisms for interaction between XSEDE staff, software developers, and the end user community to accelerate meeting of community needs and aid software developers in finding audiences for their software. XCI's long term goal is that the XSEDE Community Software Repository will be widely used and valuable to the national research community.},
 bibtype = {inproceedings},
 author = {Navarro, J. P. and Stewart, Craig A. and Knepper, Richard and Liming, Lee and Lifka, David and Dahan, Maytal},
 doi = {10.1145/3093338.3093373},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact - PEARC17}
}

© 2017 ACM. The Extreme Science and Engineering Discovery Environment (XSEDE) connects cyberinfrastructure (CI) resources, software, and services. One of XSEDE's primary goals in supporting US research generally is to "advance the ecosystem"-making use of XSEDE's leadership position to create software, tools, and services that lead to an effective and efficient national cyberinfrastructure. Software enables this endeavor in two very distinct ways: enabling the operation of XSEDE as a distributed yet integrated cyberinfrastructure resource; and by providing access to a wide variety of software of value to end user researchers and students, operators of campus cyberinfrastructure resources, and to those considering to propose new cyberinfrastructure resources to the National Science Foundation (NSF). The Community Software Repository (CSR) provides transparency about how XSEDE operates and provides access to software of use and value to the US research community generally. The CSR provides access to use cases that describe needs expressed by the research community, capability delivery plans that describe how XSEDE meets those needs, and the actual software that meets those needs. Software is delivered in a variety of forms and formats. The CSR also includes mechanisms for interaction between XSEDE staff, software developers, and the end user community to accelerate meeting of community needs and aid software developers in finding audiences for their software. XCI's long term goal is that the XSEDE Community Software Repository will be widely used and valuable to the national research community.

@inproceedings{
 title = {Campus Compute Co-operative (CCC): A service oriented cloud federation},
 type = {inproceedings},
 year = {2017},
 id = {04e65649-ad79-3048-a49d-15f5d11d19fd},
 created = {2019-10-01T17:20:38.943Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:29.797Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Grimshaw2017},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2016 IEEE. Universities struggle to provide both the quantity and diversity of compute resources that their researchers need when their researchers need them. Purchasing resources to meet peak demand for all resource types is cost prohibitive for all but a few institutions. Renting capacity on commercial clouds is seen as an alternative to owning. Commercial clouds though expect to be paid. The Campus Compute Cooperative (CCC) provides an alternative to purchasing capacity from commercial providers that provides increased value to member institutions at reduced cost. Member institutions trade their resources with one another to meet both local peak demand as well as provide access to resource types not available on the local campus that are available elsewhere. Participating institutions have dual roles. First as consumers of resources when their researchers use CCC machines, and second as producers of resources when CCC users from other institutions use their resources. In order to avoid the tragedy of the commons in which everyone only wants to use resources, the resource providers will receive credit when their resources are used by others. The consumer is charged based on the quality of service (high, medium, low) and the particulars of the resource provided (speed, interconnection network, memory, etc.). Account balances are cleared monthly. This paper describes solutions to both the technical and sociopolitical challenges of federating university resources and early results with the CCC. Technical issues include the security model, accounting, job specification/management and user interfaces. Socio-political issues include institutional risk management, how to manage market forces and incentives to avoid sub-optimal outcomes, and budget predictability.},
 bibtype = {inproceedings},
 author = {Grimshaw, A. and Prodhan, M.A. and Thomas, A. and Stewart, C. and Knepper, R.},
 doi = {10.1109/eScience.2016.7870880},
 booktitle = {Proceedings of the 2016 IEEE 12th International Conference on e-Science, e-Science 2016}
}

© 2016 IEEE. Universities struggle to provide both the quantity and diversity of compute resources that their researchers need when their researchers need them. Purchasing resources to meet peak demand for all resource types is cost prohibitive for all but a few institutions. Renting capacity on commercial clouds is seen as an alternative to owning. Commercial clouds though expect to be paid. The Campus Compute Cooperative (CCC) provides an alternative to purchasing capacity from commercial providers that provides increased value to member institutions at reduced cost. Member institutions trade their resources with one another to meet both local peak demand as well as provide access to resource types not available on the local campus that are available elsewhere. Participating institutions have dual roles. First as consumers of resources when their researchers use CCC machines, and second as producers of resources when CCC users from other institutions use their resources. In order to avoid the tragedy of the commons in which everyone only wants to use resources, the resource providers will receive credit when their resources are used by others. The consumer is charged based on the quality of service (high, medium, low) and the particulars of the resource provided (speed, interconnection network, memory, etc.). Account balances are cleared monthly. This paper describes solutions to both the technical and sociopolitical challenges of federating university resources and early results with the CCC. Technical issues include the security model, accounting, job specification/management and user interfaces. Socio-political issues include institutional risk management, how to manage market forces and incentives to avoid sub-optimal outcomes, and budget predictability.

@article{
 title = {STAR-Fusion: Fast and Accurate Fusion Transcript Detection from RNA-Seq},
 type = {article},
 year = {2017},
 pages = {120295},
 publisher = {Cold Spring Harbor Labs Journals},
 id = {8b2012ca-32d4-3067-929b-ad6ae32a71aa},
 created = {2019-10-01T17:20:39.619Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:39.619Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Haas2017d},
 source_type = {JOUR},
 private_publication = {false},
 bibtype = {article},
 author = {Haas, Brian and Dobin, Alexander and Stransky, Nicolas and Li, Bo and Yang, Xiao and Tickle, Timothy and Bankapur, Asma and Ganote, Carrie and Doak, Thomas and Pochet, Natalie},
 journal = {bioRxiv}
}

@inproceedings{
 title = {Jetstream: Early Operations Performance, Adoption, and Impacts},
 type = {inproceedings},
 year = {2017},
 keywords = {architecture,atmosphere,cloud,cyberinfrastructure,digital,education,globus,identity,jetstream,openstack,outreach,research,storage,training},
 pages = {7},
 websites = {http://doi.acm.org/10.1145/3147213.3155104},
 publisher = {ACM},
 city = {New York, NY, USA},
 series = {UCC '17},
 id = {df8c6344-7624-3c2f-b823-d55011116fa9},
 created = {2019-10-01T17:20:39.863Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:39.863Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hancock:2017:JEO:3147213.3155104},
 source_type = {inproceedings},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Hancock, David},
 doi = {10.1145/3147213.3155104},
 booktitle = {Proceedings of the10th International Conference on Utility and Cloud Computing}
}

@techreport{
 title = {Addressing the national need for more diversity and aggregate investment in cyberinfrastructure supporting open science and engineering research},
 type = {techreport},
 year = {2017},
 id = {c48a404d-56d6-3341-b282-b3e5e1542ccc},
 created = {2019-10-01T17:20:42.452Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:13.545Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2017a},
 source_type = {JOUR},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A}
}

@inproceedings{
 title = {XDMoD value analytics: A tool for measuring the financial and intellectual ROI of your campus cyberinfrastructure facilities},
 type = {inproceedings},
 year = {2017},
 keywords = {Funding,HPC,HPC monitoring,ROI,XDMoD},
 pages = {7},
 volume = {Part F1287},
 websites = {http://doi.acm.org/10.1145/3093338.3093358},
 month = {7},
 publisher = {Association for Computing Machinery},
 day = {9},
 id = {49a7fd03-fd5b-312f-abf5-2074e33eb8e6},
 created = {2019-10-01T17:20:43.617Z},
 accessed = {2019-08-26},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.133Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fulton2017},
 private_publication = {false},
 abstract = {Understanding the financial and intellectual value of campus-based cyberinfrastructure (CI) to the institutions that invest in such CI is intrinsically dificult. Given today's financial pressures, there is o.en administrative pressure questioning the value of campus-based and campus-funded CI resources. In this paper we describe new financial analytics capabilities being added to the widely used system analysis tool Open XDMoD (XSEDE Metrics on Demand) to create a new realm of metrics that will allow us to correlate usage of high performance computing with funding and publications. The capabilities to be added will eventually allow CI centers to view metrics relevant to both scientific output in terms of publications, and financial data in terms of awarded grants. The creation of Open XDMoD Value Analytics was funded by the National Science Foundation as a two year project. We are now nearing the end of the first year of this award, during which we focused on financial analytics. During the second year of this project we will focus on analytics of intellectual output.This module will allow the same sorts of analyses about systems and users as the financial analytics module, but in terms of intellectual outputs such as number of publications, citations to publications, and H indices. This module will also have capabilities to visualize such data, integrated with financial data. We plan to present these tools at PEARC '18.},
 bibtype = {inproceedings},
 author = {Fulton, Ben and Gallo, Steven and Henschel, Robert and Yearke, Tom and Börner, Katy and DeLeon, Robert L. and Furlani, Thomas and Stewart, Craig A. and Link, Matt},
 doi = {10.1145/3093338.3093358},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17 )}
}

Understanding the financial and intellectual value of campus-based cyberinfrastructure (CI) to the institutions that invest in such CI is intrinsically dificult. Given today's financial pressures, there is o.en administrative pressure questioning the value of campus-based and campus-funded CI resources. In this paper we describe new financial analytics capabilities being added to the widely used system analysis tool Open XDMoD (XSEDE Metrics on Demand) to create a new realm of metrics that will allow us to correlate usage of high performance computing with funding and publications. The capabilities to be added will eventually allow CI centers to view metrics relevant to both scientific output in terms of publications, and financial data in terms of awarded grants. The creation of Open XDMoD Value Analytics was funded by the National Science Foundation as a two year project. We are now nearing the end of the first year of this award, during which we focused on financial analytics. During the second year of this project we will focus on analytics of intellectual output.This module will allow the same sorts of analyses about systems and users as the financial analytics module, but in terms of intellectual outputs such as number of publications, citations to publications, and H indices. This module will also have capabilities to visualize such data, integrated with financial data. We plan to present these tools at PEARC '18.

@inproceedings{
 title = {A hybrid approach to population construction for agricultural agent-based simulation},
 type = {inproceedings},
 year = {2017},
 id = {e8af4cb3-dcf6-356d-9d2d-fbd8e1f2f026},
 created = {2019-10-01T17:20:44.847Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:19.294Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2016 IEEE. An Agent Based Model (ABM) is a powerful tool for its ability to represent heterogeneous agents which through their interactions can reveal emergent phenomena. For this to occur though, the set of agents in an ABM has to accurately model a real world population to reflect its heterogeneity. But when studying human behavior in less well developed settings, the availability of the real population data can be limited, making it impossible to create agents directly from the real population. In this paper, we propose a hybrid method to deal with this data scarcity: we first use the available real population data as the baseline to preserve the true heterogeneity, and fill in the missing characteristics based on survey and remote sensing datasets; then for the remaining undetermined agent characteristics, we use the Microbial Genetic Algorithm to search for a set of values that can optimize the replicative validity of the model to match data observed from real world. We apply our method to the creation of a synthetic population of household agents for the simulation of agricultural decision making processes in rural Zambia. The result shows that the synthetic population created from the farmer register can correctly reflect the marginal distributions and the randomness of survey data; and can minimize the difference between the distribution of simulated yield and that of the observed yield in Post Harvest Survey (PHS).},
 bibtype = {inproceedings},
 author = {Chen, P. and Evans, T. and Frisby, M. and Izquierdo, E. and Plale, B.},
 doi = {10.1109/eScience.2016.7870914},
 booktitle = {Proceedings of the 2016 IEEE 12th International Conference on e-Science, e-Science 2016}
}

© 2016 IEEE. An Agent Based Model (ABM) is a powerful tool for its ability to represent heterogeneous agents which through their interactions can reveal emergent phenomena. For this to occur though, the set of agents in an ABM has to accurately model a real world population to reflect its heterogeneity. But when studying human behavior in less well developed settings, the availability of the real population data can be limited, making it impossible to create agents directly from the real population. In this paper, we propose a hybrid method to deal with this data scarcity: we first use the available real population data as the baseline to preserve the true heterogeneity, and fill in the missing characteristics based on survey and remote sensing datasets; then for the remaining undetermined agent characteristics, we use the Microbial Genetic Algorithm to search for a set of values that can optimize the replicative validity of the model to match data observed from real world. We apply our method to the creation of a synthetic population of household agents for the simulation of agricultural decision making processes in rural Zambia. The result shows that the synthetic population created from the farmer register can correctly reflect the marginal distributions and the randomness of survey data; and can minimize the difference between the distribution of simulated yield and that of the observed yield in Post Harvest Survey (PHS).

@inbook{
 type = {inbook},
 year = {2017},
 pages = {91-111},
 publisher = {Elsevier},
 id = {cb43394f-4aac-33e6-b5b7-bfdc0fcbddb4},
 created = {2019-10-01T17:20:47.444Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:42.639Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Plale2017},
 source_type = {CHAP},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {inbook},
 author = {Plale, Beth and Kouper, Inna},
 chapter = {The Centrality of Data: Data Lifecycle and Data Pipelines},
 title = {Data Analytics for Intelligent Transportation Systems}
}

@inproceedings{
 title = {Low latency stream processing: Apache Heron with Infiniband & Intel Omni-Path},
 type = {inproceedings},
 year = {2017},
 id = {d1406119-7e96-3a95-b654-b8236eb8b552},
 created = {2019-10-01T17:20:47.550Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:41.749Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kamburugamuve2017},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Kamburugamuve, Supun and Ramasamy, Karthik and Swany, Martin and rey Fox, Geo},
 doi = {10.1145/3147213.3147232},
 booktitle = {UCC}
}

@inproceedings{
 title = {Enhancing access to digital media: The language application grid in the HTRC Data Capsule},
 type = {inproceedings},
 year = {2017},
 volume = {Part F1287},
 id = {3fa285cc-54ad-3ac8-98e2-4ac34fc0731f},
 created = {2019-10-01T17:20:47.644Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:39.297Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pustejovsky2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,abe5f81c-4f1f-44ec-a798-58d67325f9ec,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 held by the owner/author(s). The project "Workset Creation for Scholarly Analysis and Data Capsules" is building an infrastructure where researchers have access to text processing tools that can then be used on a copyrighted set of digital data. The infrastructure is built on (1) the HathiTrust Research Center (HTRC) Data Capsule services that can be used to access the HathiTrust Digital Library and (2) the language processing services of the Language Application (LAPPS Grid). The main thrust of the work presented here is the integration of the LAPPS Grid workflow infrastructure with the secure data access computing environment provided by the Data Capsules.},
 bibtype = {inproceedings},
 author = {Pustejovsky, J. and Verhagen, M. and Rim, K. and Ma, Y. and Ran, L. and Liyanage, S. and Murdock, J. and McDonald, R.H. and Plale, B.},
 doi = {10.1145/3093338.3104171},
 booktitle = {ACM International Conference Proceeding Series}
}

© 2017 held by the owner/author(s). The project "Workset Creation for Scholarly Analysis and Data Capsules" is building an infrastructure where researchers have access to text processing tools that can then be used on a copyrighted set of digital data. The infrastructure is built on (1) the HathiTrust Research Center (HTRC) Data Capsule services that can be used to access the HathiTrust Digital Library and (2) the language processing services of the Language Application (LAPPS Grid). The main thrust of the work presented here is the integration of the LAPPS Grid workflow infrastructure with the secure data access computing environment provided by the Data Capsules.

@article{
 title = {Pacific Rim Applications and Grid Middleware Assembly (PRAGMA): International clouds for data science},
 type = {article},
 year = {2017},
 volume = {29},
 id = {d8fef7a7-34a2-3c67-b9ea-36e0168a7309},
 created = {2019-10-01T17:20:48.215Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:23.946Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Plale2017a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Plale, B. and Chen, M.},
 doi = {10.1002/cpe.4140},
 journal = {Concurrency Computation},
 number = {13}
}

@inproceedings{
 title = {Towards Publishing Secure Capsule-Based Analysis},
 type = {inproceedings},
 year = {2017},
 id = {95ade062-b125-3dca-9cc9-e4a2ccbc381c},
 created = {2019-10-01T17:20:48.932Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:26.309Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Murdock2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 IEEE. Computational engagement with the HathiTrust Digital Library (HTDL) is confounded by the in- copyright status and licensing restrictions on the majority of the content. Because of these limitations, computational analysis on the HTDL must either be carried out in a secure environment or on derivative datasets. The HathiTrust Research Center (HTRC) Data Capsule service provides researchers with a secure environment through which they invoke tools that create, analyze, and export non-consumptive datasets. These derivative datasets, so long as they do not reproduce the full-text of the original work, are a transformative work protected by Fair Use provisions of United States Copyright Law, and can be published for reuse by other researchers, as the HTRC Extracted Features Dataset has been. Secure environments and derivative datasets enable researchers to engage with restricted data from focused studies of a few dozen volumes to large- scale experiments on millions of volumes. This paper describes advances in the Capsule service through a case study of how the HTRC Data Capsule service has advanced our activities on provenance, workflows, worksets, and non-consumptive exports through a topic modeling example. We also discuss the potential applications of this Capsule-based model to other digital libraries wrestling with research access and copyright restrictions.},
 bibtype = {inproceedings},
 author = {Murdock, J. and Jett, J. and Cole, T. and Ma, Y. and Downie, J.S. and Plale, B.},
 doi = {10.1109/JCDL.2017.7991585},
 booktitle = {Proceedings of the ACM/IEEE Joint Conference on Digital Libraries}
}

© 2017 IEEE. Computational engagement with the HathiTrust Digital Library (HTDL) is confounded by the in- copyright status and licensing restrictions on the majority of the content. Because of these limitations, computational analysis on the HTDL must either be carried out in a secure environment or on derivative datasets. The HathiTrust Research Center (HTRC) Data Capsule service provides researchers with a secure environment through which they invoke tools that create, analyze, and export non-consumptive datasets. These derivative datasets, so long as they do not reproduce the full-text of the original work, are a transformative work protected by Fair Use provisions of United States Copyright Law, and can be published for reuse by other researchers, as the HTRC Extracted Features Dataset has been. Secure environments and derivative datasets enable researchers to engage with restricted data from focused studies of a few dozen volumes to large- scale experiments on millions of volumes. This paper describes advances in the Capsule service through a case study of how the HTRC Data Capsule service has advanced our activities on provenance, workflows, worksets, and non-consumptive exports through a topic modeling example. We also discuss the potential applications of this Capsule-based model to other digital libraries wrestling with research access and copyright restrictions.

@inproceedings{
 title = {Data Capsule Appliance for Research Analysis of Restricted and Sensitive Data in Academic Libraries},
 type = {inproceedings},
 year = {2017},
 websites = {https://www.cni.org/topics/special-collections/data-capsule-appliance-for-research-analysis-of-restricted-and-sensitive-data-in-academic-libraries},
 city = {Washington, DC},
 id = {f89e61ab-9809-3446-8659-8f93fbb435fe},
 created = {2019-10-01T17:20:51.504Z},
 accessed = {2018-01-12},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:51.504Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {McDonald2017},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {McDonald, R. H. and Mitchell, Erik and Unsworth, John and Kouper, Inna},
 booktitle = {Coalition of Networked Information Fall Meeting CNI-2017}
}

@inproceedings{
 title = {Comet - Tales from the Long Tail - Two years in and 10,000 users later},
 type = {inproceedings},
 year = {2017},
 volume = {Part F1287},
 id = {265020d0-3edd-3b79-b1d8-452ef6838eed},
 created = {2019-10-01T17:20:57.550Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:31.385Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Strandea2017},
 folder_uuids = {82975498-107c-4bb3-bb76-f87bce3e9f6e,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 ACM. The Comet petascale supercomputer was put into production as an XSEDE resource in early 2015 with the goal of serving a much larger user community than HPC systems of similar size. The Comet project set an audacious goal of reaching over 10,000 users in its four years of planned operation. That goal was achieved in less than two years, due in large part to the adoption of policies that favor smaller allocations and science gateways. Here we describe our experiences in operating and supporting Comet, highlight some of the important science that it has enabled, and provide some practical lessons that we have learned by operating a system designed for the long tail of science.},
 bibtype = {inproceedings},
 author = {Strandea, S.M. and Caia, H. and Cooper, T. and Flammer, K. and Irving, C. and Von Laszewski, G. and Majumdar, A. and Mishin, D. and Papadopoulos, P. and Pfeiffer, W. and Sinkovits, R.S. and Tatineni, M. and Wagner, R. and Wang, F. and Wilkins-Diehr, N. and Wolter, N. and Norman, M.L.},
 doi = {10.1145/3093338.3093383},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

© 2017 ACM. The Comet petascale supercomputer was put into production as an XSEDE resource in early 2015 with the goal of serving a much larger user community than HPC systems of similar size. The Comet project set an audacious goal of reaching over 10,000 users in its four years of planned operation. That goal was achieved in less than two years, due in large part to the adoption of policies that favor smaller allocations and science gateways. Here we describe our experiences in operating and supporting Comet, highlight some of the important science that it has enabled, and provide some practical lessons that we have learned by operating a system designed for the long tail of science.

@inproceedings{
 title = {Provenance Enriched PID Kernel Information as OAI-ORE Map Replacement for SEAD Research Objects},
 type = {inproceedings},
 year = {2017},
 id = {c4f245c1-f6e8-33ee-b7a7-b8d2cfc85f32},
 created = {2019-10-01T17:20:59.087Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:56.173Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kouper2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 IEEE. PIDs and PID Kernel Information, activities of the Research Data Alliance, have the potential to expand the utility and benefit of data provenance. The poster describes such expansion and outlines a study of the trade-offs of replacing the Research Object (RO) and OAI-ORE map solution of the SEAD publishing services with the PID Kernel Information approach.},
 bibtype = {inproceedings},
 author = {Kouper, I. and Luo, Y. and Suriarachchi, I. and Plale, B.},
 doi = {10.1109/JCDL.2017.7991612},
 booktitle = {Proceedings of the ACM/IEEE Joint Conference on Digital Libraries}
}

© 2017 IEEE. PIDs and PID Kernel Information, activities of the Research Data Alliance, have the potential to expand the utility and benefit of data provenance. The poster describes such expansion and outlines a study of the trade-offs of replacing the Research Object (RO) and OAI-ORE map solution of the SEAD publishing services with the PID Kernel Information approach.

@article{
 title = {Offloading Collective Operations to Programmable Logic},
 type = {article},
 year = {2017},
 volume = {37},
 id = {54f161a7-4c03-3874-a3c3-ef90dbd6a5bd},
 created = {2019-10-01T17:20:59.430Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:06.061Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Arap2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 1981-2012 IEEE. The authors describe their architecture and implementation for offloading collective operations to programmable logic in the communication substrate. Collective operations are widely used in parallel processing. Their design and implementation strategies affect the performance of many high-performance computing applications that utilize them. Collectives are central to the message passing interface (MPI) programming model. The programmable logic provided by field-programmable gate arrays (FPGAS) is a powerful option for creating task-specific logic to aid applications. The authors' approach is applicable in scenarios where there is programmable logic in the communication pipeline and can be used to accelerate various network-based operations. In this article, the authors present a general collective offloading framework for use in applications using MPI. They evaluate their approach on the Xilinx Zynq system on a chip and an FPGA-based network interface card called the NetFPGA. Results are presented both from microbenchmarks and a benchmark scientific application using MPI.},
 bibtype = {article},
 author = {Arap, O. and Brasilino, L.R.B. and Kissel, E. and Shroyer, A. and Swany, M.},
 doi = {10.1109/MM.2017.3711654},
 journal = {IEEE Micro},
 number = {5}
}

© 1981-2012 IEEE. The authors describe their architecture and implementation for offloading collective operations to programmable logic in the communication substrate. Collective operations are widely used in parallel processing. Their design and implementation strategies affect the performance of many high-performance computing applications that utilize them. Collectives are central to the message passing interface (MPI) programming model. The programmable logic provided by field-programmable gate arrays (FPGAS) is a powerful option for creating task-specific logic to aid applications. The authors' approach is applicable in scenarios where there is programmable logic in the communication pipeline and can be used to accelerate various network-based operations. In this article, the authors present a general collective offloading framework for use in applications using MPI. They evaluate their approach on the Xilinx Zynq system on a chip and an FPGA-based network interface card called the NetFPGA. Results are presented both from microbenchmarks and a benchmark scientific application using MPI.

@inproceedings{
 title = {OSiRIS: a distributed Ceph deployment using software defined networking for multi-institutional research},
 type = {inproceedings},
 year = {2017},
 pages = {62045},
 volume = {898},
 issue = {6},
 publisher = {IOP Publishing},
 id = {4689770d-a347-380a-85f7-f0f4b29f879d},
 created = {2019-10-01T17:20:59.733Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:03.329Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {McKee2017},
 source_type = {CONF},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {McKee, Shawn and Kissel, Ezra and Meekhof, Benjeman and Swany, Martin and Miller, Charles and Gregorowicz, Michael},
 booktitle = {Journal of Physics: Conference Series}
}

@article{
 title = {Identification and characterization of information-networks in long-tail data collections},
 type = {article},
 year = {2017},
 volume = {94},
 id = {86c6864e-80c0-3d11-8872-a216ea66f80b},
 created = {2019-10-01T17:21:00.369Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:54.074Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Elag2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 Elsevier Ltd Scientists' ability to synthesize and reuse long-tail scientific data lags far behind their ability to collect and produce these data. Many Earth Science Cyberinfrastructures enable sharing and publishing their data over the web using metadata standards. While profiling data attributes advances the Linked Data approach, it has become clear that building information-networks among distributed data silos is essential to increase their integration and reusability. In this research, we developed a Long-Tail Information-Network (LTIN) model, which uses a metadata-driven approach to build semantic information-networks among datasets published over the web and aggregate them around environmental events. The model identifies and characterizes the spatial and temporal contextual association links and dependencies among datasets. This paper presents the design and application of the LTIN model, and an evaluation of its performance. The model capabilities were demonstrated by inferring the information-network of a stream discharge located at the downstream end of the Illinois River.},
 bibtype = {article},
 author = {Elag, M.M. and Kumar, P. and Marini, L. and Myers, J.D. and Hedstrom, M. and Plale, B.A.},
 doi = {10.1016/j.envsoft.2017.03.032},
 journal = {Environmental Modelling and Software}
}

© 2017 Elsevier Ltd Scientists' ability to synthesize and reuse long-tail scientific data lags far behind their ability to collect and produce these data. Many Earth Science Cyberinfrastructures enable sharing and publishing their data over the web using metadata standards. While profiling data attributes advances the Linked Data approach, it has become clear that building information-networks among distributed data silos is essential to increase their integration and reusability. In this research, we developed a Long-Tail Information-Network (LTIN) model, which uses a metadata-driven approach to build semantic information-networks among datasets published over the web and aggregate them around environmental events. The model identifies and characterizes the spatial and temporal contextual association links and dependencies among datasets. This paper presents the design and application of the LTIN model, and an evaluation of its performance. The model capabilities were demonstrated by inferring the information-network of a stream discharge located at the downstream end of the Illinois River.

@inproceedings{
 title = {High-Performance Massive Subgraph Counting Using Pipelined Adaptive-Group Communication.},
 type = {inproceedings},
 year = {2017},
 pages = {173-197},
 websites = {https://doi.org/10.3233/978-1-61499-882-2-173},
 id = {1ee48d86-fe5d-399b-8ecc-b8e41c276697},
 created = {2019-10-01T17:21:01.407Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.543Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2017b},
 source_type = {CPAPER},
 private_publication = {false},
 abstract = {Subgraph counting aims to count the number of occurrences of a subgraph T (aka as a template) in a given graph G. The basic problem has found applications in diverse domains. The problem is known to be computationally challenging – the complexity grows both as a function of T and G. Recent applications have motivated solving such problems on massive networks with billions of vertices. In this chapter, we study the subgraph counting problem from a parallel computing perspective. We discuss efficient parallel algorithms for approximately resolving subgraph counting problems by using the color-coding technique. We then present several system-level strategies to substantially improve the overall performance of the algorithm in massive subgraph counting problems. We propose: 1) a novel pipelined Adaptive-Group communication pattern to improve inter-node scalability, 2) a fine-grained pipeline design to effectively reduce the memory space of intermediate results, 3) partitioning neighbor lists of subgraph vertices to achieve better thread concurrency and workload balance. Experimentation on an Intel Xeon E5 cluster shows that our implementation achieves 5x speedup of performance compared to the state-of-the-art work while reduces the peak memory utilization by a factor of 2 on large templates of 12 to 15 vertices and input graphs of 2 to 5 billions of edges.},
 bibtype = {inproceedings},
 author = {Chen, Langshi and Peng, Bo and Ossen, Sabra and Vullikanti, Anil and Marathe, Madhav and Jiang, Lei and Qiu, Judy},
 doi = {10.3233/978-1-61499-882-2-173},
 booktitle = {Big Data and HPC: Ecosystem and Convergence, TopHPC 2017}
}

Subgraph counting aims to count the number of occurrences of a subgraph T (aka as a template) in a given graph G. The basic problem has found applications in diverse domains. The problem is known to be computationally challenging – the complexity grows both as a function of T and G. Recent applications have motivated solving such problems on massive networks with billions of vertices. In this chapter, we study the subgraph counting problem from a parallel computing perspective. We discuss efficient parallel algorithms for approximately resolving subgraph counting problems by using the color-coding technique. We then present several system-level strategies to substantially improve the overall performance of the algorithm in massive subgraph counting problems. We propose: 1) a novel pipelined Adaptive-Group communication pattern to improve inter-node scalability, 2) a fine-grained pipeline design to effectively reduce the memory space of intermediate results, 3) partitioning neighbor lists of subgraph vertices to achieve better thread concurrency and workload balance. Experimentation on an Intel Xeon E5 cluster shows that our implementation achieves 5x speedup of performance compared to the state-of-the-art work while reduces the peak memory utilization by a factor of 2 on large templates of 12 to 15 vertices and input graphs of 2 to 5 billions of edges.

@article{
 title = {Mining lake time series using symbolic representation},
 type = {article},
 year = {2017},
 volume = {39},
 id = {fb016cdf-ce82-3c9f-a370-cc423fc408d7},
 created = {2019-10-01T17:21:02.259Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:12.946Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ruan2017},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2017 Elsevier B.V. Sensor networks deployed in lakes and reservoirs, when combined with simulation models and expert knowledge from the global community, are creating deeper understanding of the ecological dynamics of lakes. However, the amount of data and the complex patterns in the data demand substantial compute resources and efficient data mining algorithms, both of which are beyond the realm of traditional limnological research. This paper uniquely adapts methods from computer science for application to data intensive ecological questions, in order to provide ecologists with approachable methodology to facilitate knowledge discovery in lake ecology. We apply a state-of-the-art time series mining technique based on symbolic representation (SAX) to high-frequency time series of phycocyanin (PHYCO) and chlorophyll (CHLORO) fluorescence, both of which are indicators of algal biomass in lakes, as well as model predictions of algal biomass (MODEL). We use data mining techniques to demonstrate that MODEL predicts PHYCO better than it predicts CHLORO. All time series have high redundancy, resulting in a relatively small subset of unique patterns. However, MODEL is much less complex than either PHYCO or CHLORO and fails to reproduce high biomass periods indicative of algal blooms. We develop a set of tools in R to enable motif discovery and anomaly detection within a single lake time series, and relationship study among multiple lake time series through distance metrics, clustering and classification. Furthermore, to improve computation times, we provision web services to launch R tools remotely on high performance computing (HPC) resources. Comprehensive experimental results on observational and simulated lake data demonstrate the effectiveness of our approach.},
 bibtype = {article},
 author = {Ruan, G. and Hanson, P.C. C. and Dugan, H.A. and Plale, Beth A.},
 doi = {10.1016/j.ecoinf.2017.03.001},
 journal = {Ecological Informatics}
}

© 2017 Elsevier B.V. Sensor networks deployed in lakes and reservoirs, when combined with simulation models and expert knowledge from the global community, are creating deeper understanding of the ecological dynamics of lakes. However, the amount of data and the complex patterns in the data demand substantial compute resources and efficient data mining algorithms, both of which are beyond the realm of traditional limnological research. This paper uniquely adapts methods from computer science for application to data intensive ecological questions, in order to provide ecologists with approachable methodology to facilitate knowledge discovery in lake ecology. We apply a state-of-the-art time series mining technique based on symbolic representation (SAX) to high-frequency time series of phycocyanin (PHYCO) and chlorophyll (CHLORO) fluorescence, both of which are indicators of algal biomass in lakes, as well as model predictions of algal biomass (MODEL). We use data mining techniques to demonstrate that MODEL predicts PHYCO better than it predicts CHLORO. All time series have high redundancy, resulting in a relatively small subset of unique patterns. However, MODEL is much less complex than either PHYCO or CHLORO and fails to reproduce high biomass periods indicative of algal blooms. We develop a set of tools in R to enable motif discovery and anomaly detection within a single lake time series, and relationship study among multiple lake time series through distance metrics, clustering and classification. Furthermore, to improve computation times, we provision web services to launch R tools remotely on high performance computing (HPC) resources. Comprehensive experimental results on observational and simulated lake data demonstrate the effectiveness of our approach.

@inproceedings{
 title = {Low Latency Stream Processing: Twitter Heron with Infiniband and Omni-Path},
 type = {inproceedings},
 year = {2017},
 id = {f5d0e355-cb95-3011-89c2-e40f764e1310},
 created = {2019-10-01T17:21:02.526Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:12.327Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kamburugamuve2017a},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Kamburugamuve, Supun and Ramasamy, Karthik and Swany, Martin and Fox, Geoffrey Charles},
 booktitle = {Proceedings of Strata Data Conference}
}

@techreport{
 title = {Introduction to Harp: when Big Data Meets HPC},
 type = {techreport},
 year = {2017},
 pages = {10},
 websites = {https://pdfs.semanticscholar.org/f69e/9f2852c881da4df0142360c745441075a28f.pdf?_ga=2.28162492.1830027813.1567539825-1063534713.1566236187},
 id = {d1ae1abc-8111-3180-bb02-5a653b5af0c6},
 created = {2019-10-01T17:21:02.693Z},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.566Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Zhang2017},
 private_publication = {false},
 abstract = {Harp Data analytics is undergoing a revolution in many scientific domains, demanding cost-effective parallel data analysis techniques. We consider the challenges of creating a high performance data analysis software framework in the context of the current HPC-ABDS software stack (High Performance Computing enhanced Apache Big Data Stack) [1]. We have summarized a list of current data processing software from either HPC or commercial sources [2]. Many critical components of the commodity stack (such as Hadoop) come from Apache open source projects for community usage, while HPC (such as collective communication) is needed to bring performance and other parallel computing capabilities. Many machine learning algorithms are built on iterative computation, which can be formulated as í µí°´"µí°´" = í µí°¹(í µí°·, í µí°´"µí°´"()) (1) where D is the observed dataset, A is model parameters to learn, and F is the model update function. The algorithm keeps updating model A until convergence, either by reaching a threshold criterion or fixed number of iterations. There are several advantages of this iterative procedure as apparently simple functions can iterate and produce complex behavior for interesting problems. The power of iteration and its extensions lies in the approximation or accuracy that can be obtained at each step even if the computation stops abruptly before converges to the final answer. To effectively support large-scale data processing, Twister [3] introduced iterative MapReduce using long-running processes or threads with in-memory caching of invariant data. Harp [4] introduces full collective communication in Table 1 (broadcast, reduce, allgather, allreduce, rotation, regroup or push & pull), adding a separate communication abstraction where the Harp prototype implements the MapCollective concept as a plug-in to Hadoop Ecosystem (see Figure 1 and Figure 2). Instead of using the shuffling phase, Harp uses optimized collective communication operations for data movement since fine-grained data alignment for multiple models is critical for improving performance. It further provides high-level interfaces with various synchronization patterns for parallelizing iterative computation. These enhancements make it possible to exploit HPC capabilities for big data software systems. Figure 1 Map-Collective Model Figure 2 Harp Architecture Shuffle M M M M Collective Communication M M M M R R MapCollective Model MapReduce Model YARN MapReduce V2 Harp MapReduce Applications MapCollective Applications Application Framework Resource Manager},
 bibtype = {techreport},
 author = {Zhang, Bingjing and Peng, Bo and Chen, Langshi and Li, Ethan and Zhou, Yiming and Qiu, Judy}
}

Harp Data analytics is undergoing a revolution in many scientific domains, demanding cost-effective parallel data analysis techniques. We consider the challenges of creating a high performance data analysis software framework in the context of the current HPC-ABDS software stack (High Performance Computing enhanced Apache Big Data Stack) [1]. We have summarized a list of current data processing software from either HPC or commercial sources [2]. Many critical components of the commodity stack (such as Hadoop) come from Apache open source projects for community usage, while HPC (such as collective communication) is needed to bring performance and other parallel computing capabilities. Many machine learning algorithms are built on iterative computation, which can be formulated as í µí°´"µí°´" = í µí°¹(í µí°·, í µí°´"µí°´"()) (1) where D is the observed dataset, A is model parameters to learn, and F is the model update function. The algorithm keeps updating model A until convergence, either by reaching a threshold criterion or fixed number of iterations. There are several advantages of this iterative procedure as apparently simple functions can iterate and produce complex behavior for interesting problems. The power of iteration and its extensions lies in the approximation or accuracy that can be obtained at each step even if the computation stops abruptly before converges to the final answer. To effectively support large-scale data processing, Twister [3] introduced iterative MapReduce using long-running processes or threads with in-memory caching of invariant data. Harp [4] introduces full collective communication in Table 1 (broadcast, reduce, allgather, allreduce, rotation, regroup or push & pull), adding a separate communication abstraction where the Harp prototype implements the MapCollective concept as a plug-in to Hadoop Ecosystem (see Figure 1 and Figure 2). Instead of using the shuffling phase, Harp uses optimized collective communication operations for data movement since fine-grained data alignment for multiple models is critical for improving performance. It further provides high-level interfaces with various synchronization patterns for parallelizing iterative computation. These enhancements make it possible to exploit HPC capabilities for big data software systems. Figure 1 Map-Collective Model Figure 2 Harp Architecture Shuffle M M M M Collective Communication M M M M R R MapCollective Model MapReduce Model YARN MapReduce V2 Harp MapReduce Applications MapCollective Applications Application Framework Resource Manager

@techreport{
 title = {SERVOGrid Technical Documentation},
 type = {techreport},
 year = {2017},
 source = {NASA AIST Technical Report},
 id = {dc1d936b-0852-3de8-b05f-fa478bc8f52d},
 created = {2019-10-01T17:21:04.886Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:28:16.698Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {aktasservogrid},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Aktas, Mehmet and Aydin, Galip and Fox, Geoffrey Charles and Gadgil, Harshawardhan and Pierce, Marlon and Sayar, Ahmet and McLeod, Dennis and Sung, Sang-Soo}
}

@inproceedings{
 title = {Science Gateways: Sustainability via On-Campus Teams},
 type = {inproceedings},
 year = {2017},
 keywords = {On-campus groups,Science gateways,Science gateways community institute,Sustainability},
 pages = {97-102},
 volume = {94},
 websites = {https://linkinghub.elsevier.com/retrieve/pii/S0167739X18315395},
 month = {5},
 publisher = {Elsevier B.V.},
 day = {1},
 id = {de4b11c9-adf6-3bde-9700-8e1775fc99e3},
 created = {2019-10-01T17:21:05.833Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:25:34.283Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {gesingscience},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {The challenges for creators of specific science gateways are manifold, and the expertise needed for well-designed science gateways is very diverse. The sustainability of science gateways is crucial to serve communities effectively, efficiently and reliably. One measure to achieve greater sustainability of science gateways is establishing on-campus teams. Researchers are served more efficiently since the support by experienced developers reduces individual project investments, and a team can bring the diversity of required expertise for a well-designed science gateway. This paper goes into detail about the challenges and the benefits of on-campus groups and of sharing resources across a campus. We provide four successful cases, describe the services of the Science Gateways Community Institute (SGCI) to support the process in building such groups, and recommend strategies for using free campus resources.},
 bibtype = {inproceedings},
 author = {Gesing, Sandra and Lawrence, Katherine and Wilkins-Diehr, Nancy and Dahan, Maytal and Zentner, Michael and Pierce, Marlon E.},
 doi = {10.6084/m9.figshare.5483566.v1},
 booktitle = {Gateways 2017}
}

The challenges for creators of specific science gateways are manifold, and the expertise needed for well-designed science gateways is very diverse. The sustainability of science gateways is crucial to serve communities effectively, efficiently and reliably. One measure to achieve greater sustainability of science gateways is establishing on-campus teams. Researchers are served more efficiently since the support by experienced developers reduces individual project investments, and a team can bring the diversity of required expertise for a well-designed science gateway. This paper goes into detail about the challenges and the benefits of on-campus groups and of sharing resources across a campus. We provide four successful cases, describe the services of the Science Gateways Community Institute (SGCI) to support the process in building such groups, and recommend strategies for using free campus resources.

@inproceedings{
 title = {Science Gateways: The Long Road to the Birth of an Institute},
 type = {inproceedings},
 year = {2017},
 pages = {10 pages},
 websites = {http://hdl.handle.net/10125/41919},
 publisher = {Hawaii International Conference on System Sciences},
 id = {0549159a-2aba-3c5a-83fb-56ab98012b2f},
 created = {2019-10-01T17:21:07.969Z},
 accessed = {2019-08-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:30:51.449Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {gesing2017science},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {Nowadays, research in various disciplines is enhanced via computational methods, cutting-edge technologies and diverse resources including computational infrastructures and instruments. Such infrastructures are often complex and researchers need means to conduct their research in an efficient way without getting distracted with information technology nuances. Science gateways address such demands and offer user interfaces tailored to a specific community. Creators of science gateways face a breadth of topics and manifold challenges, which necessitate close collaboration with the domain specialists but also calling in experts for diverse aspects of a science gateway such as project management, licensing, team composition, sustainability, HPC, visualization, and usability specialists. The Science Gateway Community Institute tackles the challenges around science gateways to support domain specialists and developers via connecting them to diverse experts, offering consultancy as well as providing a software collaborative, which contains ready-to-use science gateway frameworks and science gateway components.},
 bibtype = {inproceedings},
 author = {Gesing, Sandra and Wilkins-Diehr, Nancy and Dahan, Maytal and Lawrence, Katherine and Zentner, Michael and Pierce, Marlon and Hayden, Linda and Marru, Suresh},
 doi = {10.24251/HICSS.2017.755},
 booktitle = {Proceedings of the 50th Hawaii International Conference on System Sciences (2017)}
}

Nowadays, research in various disciplines is enhanced via computational methods, cutting-edge technologies and diverse resources including computational infrastructures and instruments. Such infrastructures are often complex and researchers need means to conduct their research in an efficient way without getting distracted with information technology nuances. Science gateways address such demands and offer user interfaces tailored to a specific community. Creators of science gateways face a breadth of topics and manifold challenges, which necessitate close collaboration with the domain specialists but also calling in experts for diverse aspects of a science gateway such as project management, licensing, team composition, sustainability, HPC, visualization, and usability specialists. The Science Gateway Community Institute tackles the challenges around science gateways to support domain specialists and developers via connecting them to diverse experts, offering consultancy as well as providing a software collaborative, which contains ready-to-use science gateway frameworks and science gateway components.

@techreport{
 title = {Pervasive Technology Institute Annual Report: Research Innovations and Advanced Cyberinfrastructure Services in Support of IU Strategic Goals During FY 2017},
 type = {techreport},
 year = {2017},
 websites = {http://hdl.handle.net/2022/21809},
 id = {a6f9b389-24ea-39ec-acfa-b267b23c4cf8},
 created = {2019-10-01T17:21:11.059Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-10T00:01:46.813Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2017},
 source_type = {RPRT},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig; and Plale, Beth; and Welch, Von; and Pierce, Marlon; and Fox, Geoffrey C.; and Doak, Thomas G.; Hancock, David Y.; Henschel, Robert; and Link, Matthew R.; and Miller, Therese; and Wernert, Eric; and Boyles, Michael J.; and Fulton, Ben; and Weakley, Le Mai; and Ping, Robert; and Gniady, Tassie; and Snapp-Childs, Winona;}
}

@book{
 title = {Bulk Collection: Systematic Government Access to Private-Sector Data},
 type = {book},
 year = {2017},
 publisher = {Oxford University Press},
 id = {92887e88-df84-3bea-86da-f80ffda38419},
 created = {2019-10-01T17:21:13.337Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:49.599Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2017b},
 source_type = {BOOK},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {book},
 author = {Cate, Fred H and Dempsey, James X}
}

@article{
 title = {Machine Learning with Personal Data: Is Data Protection Law Smart Enough to Meet the Challenge?},
 type = {article},
 year = {2017},
 pages = {1-2},
 volume = {7},
 id = {e0fb73b7-3010-3012-abe1-728fc8044026},
 created = {2019-10-01T17:21:13.870Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:27.951Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2017},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla and Millard, Christopher},
 doi = {10.1093/idpl/ipx003},
 journal = {International Data Privacy Law},
 number = {1}
}

@article{
 title = {The medical science DMZ: a network design pattern for data-intensive medical science},
 type = {article},
 year = {2017},
 id = {eacbd033-5e0f-345e-a6f7-42b7247767af},
 created = {2019-10-01T17:21:14.052Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:23.711Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Peisert2017},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Peisert, Sean and Dart, Eli and Barnett, William and Balas, Edward and Cuff, James and Grossman, Robert L and Berman, Ari and Shankar, Anurag and Tierney, Brian},
 journal = {Journal of the American Medical Informatics Association}
}

@article{
 title = {The Rise of Cybersecurity and Its Impact on Data Protection},
 type = {article},
 year = {2017},
 pages = {73-75},
 volume = {7},
 id = {bb9fe027-4ba9-3ae6-a5ba-3e6513c9c908},
 created = {2019-10-01T17:21:15.037Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:29.454Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2017a},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla and Millard, Christopher},
 doi = {https://doi.org/10.1093/idpl/ipw022},
 journal = {International Data Privacy Law},
 number = {2}
}

@inproceedings{
 title = {Apache airavata sharing service: A tool for enabling user collaboration in science gateways},
 type = {inproceedings},
 year = {2017},
 keywords = {Apache Airavata,CIPRES,Collab-oration,Groups,NSG,SEAGrid,SciGaP,Science Gateways,Sharing},
 volume = {Part F1287},
 month = {7},
 publisher = {Association for Computing Machinery},
 day = {9},
 id = {1ed5f759-dd44-3f28-89df-e845edebb792},
 created = {2019-10-01T17:21:23.807Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.668Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Nakandala2017},
 private_publication = {false},
 abstract = {Science Gateways provide user environments and a set of supporting services that help researchers make effective and enhanced use of a diverse set of computing, storage, and related resources. Gateways provide the services and tools users require to enable their scientific exploration, which in-cludes tasks such as running computer simulations or per-forming data analysis. Historically gateways have been con-structed to support the workflow of individual users, but collaboration between users has become an increasingly im-portant part of the discovery process. This trend has created a driving need for gateways to support data sharing between users. For example, a chemistry research group may want to run simulations collaboratively, analyze experimental data or tune parameter studies based on simulation output gen-erated by peers, whether as a default capability, or through explicit creation of sharing privileges. As another example, students in a classroom setting may be required to share their simulation output or data analysis results with the in-structor. However most existing gateways (including the pop-ularly used XSEDE gateways SEAGrid, Ultrascan, CIPRES, and NSG), do not support direct data sharing, so users have to handle these collaborations outside the gateway environ-ment. Given the importance of collaboration in current scien-tific practice, user collaboration should be a prime consider-ation in building science gateways. In this work, we present design considerations and implementation of a generic model that can be used to describe and handle a diverse set of user collaboration use cases that arise in gateways, based on gen-eral requirements gathered from the SEAGrid, CIPRES, and NSG gateways. We then describe the integration of this shar-ing service into these gateways. Though the model and the system were tested and used in the context of Science Gate-ways, the concepts are universally applicable to any domain, and the service can support data sharing in a wide varietyof use cases. © 2017 ACM.},
 bibtype = {inproceedings},
 author = {Nakandala, Supun and Marru, Suresh and Piece, Marlon and Pamidighantam, Sudhakar and Yoshimoto, Kenneth and Schwartz, Terri and Sivagnanam, Subhashini and Majumdar, Amit and Miller, Mark A.},
 doi = {10.1145/3093338.3093359},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

Science Gateways provide user environments and a set of supporting services that help researchers make effective and enhanced use of a diverse set of computing, storage, and related resources. Gateways provide the services and tools users require to enable their scientific exploration, which in-cludes tasks such as running computer simulations or per-forming data analysis. Historically gateways have been con-structed to support the workflow of individual users, but collaboration between users has become an increasingly im-portant part of the discovery process. This trend has created a driving need for gateways to support data sharing between users. For example, a chemistry research group may want to run simulations collaboratively, analyze experimental data or tune parameter studies based on simulation output gen-erated by peers, whether as a default capability, or through explicit creation of sharing privileges. As another example, students in a classroom setting may be required to share their simulation output or data analysis results with the in-structor. However most existing gateways (including the pop-ularly used XSEDE gateways SEAGrid, Ultrascan, CIPRES, and NSG), do not support direct data sharing, so users have to handle these collaborations outside the gateway environ-ment. Given the importance of collaboration in current scien-tific practice, user collaboration should be a prime consider-ation in building science gateways. In this work, we present design considerations and implementation of a generic model that can be used to describe and handle a diverse set of user collaboration use cases that arise in gateways, based on gen-eral requirements gathered from the SEAGrid, CIPRES, and NSG gateways. We then describe the integration of this shar-ing service into these gateways. Though the model and the system were tested and used in the context of Science Gate-ways, the concepts are universally applicable to any domain, and the service can support data sharing in a wide varietyof use cases. © 2017 ACM.

@article{
 title = {A Non-Natural Wurtzite Polymorph of HgSe: A Potential 3D Topological Insulator},
 type = {article},
 year = {2017},
 pages = {6356-6366},
 volume = {29},
 websites = {https://pubs.acs.org/doi/10.1021/acs.chemmater.7b01674},
 month = {8},
 publisher = {American Chemical Society},
 day = {8},
 id = {e5eeed0c-cb9f-3c55-87ba-f6145d2ee88a},
 created = {2019-10-01T17:21:24.980Z},
 accessed = {2019-09-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.187Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {DumettTorres2017},
 private_publication = {false},
 abstract = {This article demonstrates the power of topotactic synthesis coupled with density functional theory (DFT) for accessing and exploring new phases of matter. Naturally occurring HgSe is a semimetal with a zero gap. Unlike this natural zincblende form of HgSe, our DFT investigations predict that wurtzite HgSe has both an inverted band structure and a band gap, making it a 3D topological insulator (TI). Calculated band structures of HgxCd1–xSe alloys containing strongly relativistic Hg and weakly relativistic Cd show that band gap opening is a consequence of symmetry breaking resulting from a combination of crystal anisotropy and the scalar relativistic effect of Hg electrons. The relativistic contribution of Hg is significant enough in alloys with x ≥ 0.33 for achieving 3D TI behavior at room temperature. We experimentally realize the non-natural wurtzite form by topotactic ion exchange of wurtzite CdSe nanocrystals (NCs), which yields alloy NCs in the range x = 0–0.54 whose measured band gaps follow the pred...},
 bibtype = {article},
 author = {Dumett Torres, Daniel and Banerjee, Progna and Pamidighantam, Sudhakar and Jain, Prashant K.},
 doi = {10.1021/acs.chemmater.7b01674},
 journal = {Chemistry of Materials},
 number = {15}
}

This article demonstrates the power of topotactic synthesis coupled with density functional theory (DFT) for accessing and exploring new phases of matter. Naturally occurring HgSe is a semimetal with a zero gap. Unlike this natural zincblende form of HgSe, our DFT investigations predict that wurtzite HgSe has both an inverted band structure and a band gap, making it a 3D topological insulator (TI). Calculated band structures of HgxCd1–xSe alloys containing strongly relativistic Hg and weakly relativistic Cd show that band gap opening is a consequence of symmetry breaking resulting from a combination of crystal anisotropy and the scalar relativistic effect of Hg electrons. The relativistic contribution of Hg is significant enough in alloys with x ≥ 0.33 for achieving 3D TI behavior at room temperature. We experimentally realize the non-natural wurtzite form by topotactic ion exchange of wurtzite CdSe nanocrystals (NCs), which yields alloy NCs in the range x = 0–0.54 whose measured band gaps follow the pred...

@article{
 title = {Radar Determination of Fault Slip and Location in Partially Decorrelated Images},
 type = {article},
 year = {2017},
 pages = {2295-2310},
 volume = {174},
 publisher = {Springer International Publishing},
 id = {2d7cf512-272a-398c-b5fc-16f8a8eb719e},
 created = {2019-10-01T17:21:25.158Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:21:25.158Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {parker2017radar},
 source_type = {article},
 private_publication = {false},
 abstract = {Parker, J., Glasscoe, M., Donnellan, A., Stough, T., Pierce, M., & Wang, J. (2017). Radar Determination of Fault Slip and Location in Partially Decorrelated Images. Pure and Applied Geophysics, 174(6), 2295–2310.},
 bibtype = {article},
 author = {Parker, Jay and Glasscoe, Margaret and Donnellan, Andrea and Stough, Timothy and Pierce, Marlon and Wang, Jun},
 journal = {Pure and Applied Geophysics},
 number = {6}
}

Parker, J., Glasscoe, M., Donnellan, A., Stough, T., Pierce, M., & Wang, J. (2017). Radar Determination of Fault Slip and Location in Partially Decorrelated Images. Pure and Applied Geophysics, 174(6), 2295–2310.

@inproceedings{
 title = {Stampede 2: The Evolution of an XSEDE supercomputer},
 type = {inproceedings},
 year = {2017},
 keywords = {Computational simulation,Computer applications,Computer programming,Exascale computing,Hig,Supercomputers},
 pages = {8},
 volume = {Part F1287},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85025827654&doi=10.1145%2F3093338.3093385&partnerID=40&md5=4ed856ab16c3decf2b1f95272d3fe109},
 publisher = {Association for Computing Machinery},
 id = {7955f357-788c-34c2-9052-ee762ebdd60f},
 created = {2019-10-01T17:21:28.721Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:16.718Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stanzione2017},
 source_type = {conference},
 notes = {cited By 0; Conference of 2017 Practice and Experience in Advanced Research Computing, PEARC 2017 ; Conference Date: 9 July 2017 Through 13 July 2017; Conference Code:128771},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {The Stampede 1 supercomputer was a tremendous success as an XSEDE resource, providing more than eight million successful computational simulations and data analysis jobs to more than ten thousand users. In addition, Stampede 1 introduced new technology that began to move users towards many core processors. As Stampede 1 reaches the end of its production life, it is being replaced in phases by a new supercomputer, Stampede 2, that will not only take up much of the original system's workload, but continue the bridge to technologies on the path to exascale computing. This paper provides a brief summary of the experiences of Stampede 1, and details the design and architecture of Stampede 2. Early results are presented from a subset of Intel Knights Landing nodes that are bridging between the two systems. © 2017 Association for Computing Machinery.},
 bibtype = {inproceedings},
 author = {Stanzione, D and Barth, B and Gaffney, N and Gaither, K and Mehringer, S and Hempel, C and Wernert, E and Minyard, T and Tufo, H and Panda, D and Teller, P},
 doi = {10.1145/3093338.3093385},
 booktitle = {Proceedings of the Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact (PEARC17)}
}

The Stampede 1 supercomputer was a tremendous success as an XSEDE resource, providing more than eight million successful computational simulations and data analysis jobs to more than ten thousand users. In addition, Stampede 1 introduced new technology that began to move users towards many core processors. As Stampede 1 reaches the end of its production life, it is being replaced in phases by a new supercomputer, Stampede 2, that will not only take up much of the original system's workload, but continue the bridge to technologies on the path to exascale computing. This paper provides a brief summary of the experiences of Stampede 1, and details the design and architecture of Stampede 2. Early results are presented from a subset of Intel Knights Landing nodes that are bridging between the two systems. © 2017 Association for Computing Machinery.

@article{
 title = {An archive of spectra from the Mayall Fourier transform spectrometer at Kitt Peak},
 type = {article},
 year = {2017},
 volume = {129},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85011965299&doi=10.1088%2F1538-3873%2F129%2F972%2F024006&partnerID=40&md5=ae36d0681e87052b496e96baed19c7e6},
 publisher = {Institute of Physics Publishing},
 id = {e64f5837-4956-300b-a2f1-f337fb5e284e},
 created = {2019-10-01T17:21:28.746Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:15.528Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pilachowski2017},
 source_type = {article},
 notes = {cited By 0},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {We describe the SpArc science gateway for spectral data obtained using the Fourier Transform Spectrometer (FTS) in operation at the Mayall 4-m telescope at the Kitt Peak National Observatory during the period from 1975 through 1995. SpArc is hosted by Indiana University Bloomington and is available for public access. The archive includes nearly 10,000 individual spectra of more than 800 different astronomical sources including stars, nebulae, galaxies, and solar system objects. We briefly describe the FTS instrument itself and summarize the conversion of the original interferograms into spectral data and the process for recovering the data into FITS files. The architecture of the archive is discussed and the process for retrieving data from the archive is introduced. Sample use cases showing typical FTS spectra are presented. © 2017. The Astronomical Society of the Pacific. All rights reserved.},
 bibtype = {article},
 author = {Pilachowski, C A and Hinkle, K H and Young, M D and Dennis, H B and Gopu, A and Henschel, R and Hayashi, S},
 doi = {10.1088/1538-3873/129/972/024006},
 journal = {Publications of the Astronomical Society of the Pacific},
 number = {972}
}

We describe the SpArc science gateway for spectral data obtained using the Fourier Transform Spectrometer (FTS) in operation at the Mayall 4-m telescope at the Kitt Peak National Observatory during the period from 1975 through 1995. SpArc is hosted by Indiana University Bloomington and is available for public access. The archive includes nearly 10,000 individual spectra of more than 800 different astronomical sources including stars, nebulae, galaxies, and solar system objects. We briefly describe the FTS instrument itself and summarize the conversion of the original interferograms into spectral data and the process for recovering the data into FITS files. The architecture of the archive is discussed and the process for retrieving data from the archive is introduced. Sample use cases showing typical FTS spectra are presented. © 2017. The Astronomical Society of the Pacific. All rights reserved.

@article{
 title = {The Open Science Cyber Risk Profile: The Rosetta Stone for Open Science and Cybersecurity},
 type = {article},
 year = {2017},
 keywords = {Cyber security,Cyber-attacks,Natural sciences computing,OSCRP,Risk profil,Security of data,Security systems},
 pages = {94-95},
 volume = {15},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85031689244&doi=10.1109%2FMSP.2017.3681058&partnerID=40&md5=c8539146c103a7d978d47ba1947849a6},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 id = {9280502c-730a-3403-9173-c5223fb86e4d},
 created = {2019-10-01T18:06:10.908Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:06:37.443Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Peisert201794},
 source_type = {article},
 notes = {cited By 0},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {The Open Science Cyber Risk Profile (OSCRP) working group has created a document that motivates scientists by demonstrating how improving their security posture reduces the risks to their science. This effort aims to bridge the communication gap between scientists and IT security professionals and allows for the effective management of risks to open science caused by IT security threats. © 2003-2012 IEEE.},
 bibtype = {article},
 author = {Peisert, S and Welch, V},
 doi = {10.1109/MSP.2017.3681058},
 journal = {IEEE Security and Privacy},
 number = {5}
}

The Open Science Cyber Risk Profile (OSCRP) working group has created a document that motivates scientists by demonstrating how improving their security posture reduces the risks to their science. This effort aims to bridge the communication gap between scientists and IT security professionals and allows for the effective management of risks to open science caused by IT security threats. © 2003-2012 IEEE.

@inbook{
 type = {inbook},
 year = {2017},
 pages = {1063-1074},
 websites = {http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-5225-2255-3.ch092},
 publisher = {IGI Global},
 city = {Hershey, PA, USA},
 id = {95f16f8a-cf0b-37ef-a827-dd4a88bc643f},
 created = {2019-10-01T18:06:11.140Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:59:21.955Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2017b},
 source_type = {CHAP},
 private_publication = {false},
 abstract = {Computers accelerate our ability to achieve scientific breakthroughs. As technology evolves and new research needs come to light, the role for cyberinfrastructure as “knowledge” infrastructure continues to expand. In essence, cyberinfrastructure can be thought of as the integration of supercomputers, data resources, visualization, and people that extends the impact and utility of information technology. This article discusses cyberinfrastructure, the related topics of science gateways and campus bridging, and identifies future challenges and opportunities in cyberinfrastructure.},
 bibtype = {inbook},
 author = {Stewart, Craig A and Knepper, Richard and Link, Matthew R and Pierce, Marlon and Wernert, Eric and Wilkins-Diehr, Nancy},
 editor = {Mehdi Khosrow-Pour, D B A},
 doi = {10.4018/978-1-5225-2255-3.ch092},
 chapter = {Cyberinfrastructure, Cloud Computing, Science Gateways, Visualization, and Cyberinfrastructure Ease of Use},
 title = {Encyclopedia of Information Science and Technology, Fourth Edition}
}

Computers accelerate our ability to achieve scientific breakthroughs. As technology evolves and new research needs come to light, the role for cyberinfrastructure as “knowledge” infrastructure continues to expand. In essence, cyberinfrastructure can be thought of as the integration of supercomputers, data resources, visualization, and people that extends the impact and utility of information technology. This article discusses cyberinfrastructure, the related topics of science gateways and campus bridging, and identifies future challenges and opportunities in cyberinfrastructure.

@inproceedings{
 title = {Using the Jetstream Research Cloud to provide Science Gateway resources},
 type = {inproceedings},
 year = {2017},
 pages = {753-757},
 websites = {http://ieeexplore.ieee.org/document/7973774/},
 month = {5},
 publisher = {IEEE},
 id = {0f309840-7851-3bc3-b58a-4c557235d27d},
 created = {2019-10-01T18:06:11.141Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.522Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper2017},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {We describe the use of the Jetstream research-cloud, a purpose-built system with the goal of supporting "long-tail" research by providing a flexible, on-demand research infrastructure, to provide scalable back-end resources for science gateways. In addition to providing cloud-like resources for on-demand science, Jetstream offers the capability to instantiate long-running clusters which support science gateways. Science gateways are web-based systems built on computational infrastructure which provide commonly-used tools to a community of users. We created a persistent cluster on the Jetstream system which is connected to the SEAGrid science gateway and provides additional compute resources for a variety of quantum chemistry calculations. We discuss the further application of toolkits provided by the Extreme Science and Engineering Discovery Environment (XSEDE) to build general-purpose clusters on the research cloud.},
 bibtype = {inproceedings},
 author = {Knepper, Richard and Coulter, Eric and Pierce, Marlon and Marru, Suresh and Pamidighantam, Sudhakar},
 doi = {10.1109/CCGRID.2017.121},
 booktitle = {Proceedings of the 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid '17)}
}

We describe the use of the Jetstream research-cloud, a purpose-built system with the goal of supporting "long-tail" research by providing a flexible, on-demand research infrastructure, to provide scalable back-end resources for science gateways. In addition to providing cloud-like resources for on-demand science, Jetstream offers the capability to instantiate long-running clusters which support science gateways. Science gateways are web-based systems built on computational infrastructure which provide commonly-used tools to a community of users. We created a persistent cluster on the Jetstream system which is connected to the SEAGrid science gateway and provides additional compute resources for a variety of quantum chemistry calculations. We discuss the further application of toolkits provided by the Extreme Science and Engineering Discovery Environment (XSEDE) to build general-purpose clusters on the research cloud.

@techreport{
 title = {Indiana University Pervasive Technology Institute},
 type = {techreport},
 year = {2017},
 city = {Bloomington, IN},
 id = {035f0d3f-a20a-3a98-9489-28966a21c617},
 created = {2019-10-01T18:06:11.668Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:06:19.226Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2017a},
 source_type = {misc},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,73f994b4-a3be-4035-a6dd-3802077ce863,9559c0d7-009c-4a6d-b131-a3e3d83d98b8,ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, C A and Welch, V and Plale, B and Fox, Geoffrey Charles and Pierce, M and Sterling, T},
 doi = {10.5072/FK2154N14D}
}

@article{
 title = {One network metric datastore to track them all: the OSG network metric service},
 type = {article},
 year = {2017},
 pages = {82044},
 volume = {898},
 id = {5851db4d-79fe-39d5-a211-33f19410e53c},
 created = {2020-04-22T21:44:56.762Z},
 accessed = {2020-04-22},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.792Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Quick2017},
 private_publication = {false},
 abstract = {The Open Science Grid (OSG) relies upon the network as a critical part of the distributed infrastructures it enables. In 2012, OSG added a new focus area in networking with a goal of becoming the primary source of network information for its members and collaborators. This includes gathering, organizing, and providing network metrics to guarantee effective network usage and prompt detection and resolution of any network issues, including connection failures, congestion, and traffic routing. In September of 2015, this service was deployed into the OSG production environment. We will report on the creation, implementation, testing, and deployment of the OSG Networking Service. Starting from organizing the deployment of perfSONAR toolkits within OSG and its partners, to the challenges of orchestrating regular testing between sites, to reliably gathering the resulting network metrics and making them available for users, virtual organizations, and higher level services, all aspects of implementation will be reviewed. In particular, several higher-level services were developed to bring the OSG network service to its full potential. These include a web-based mesh configuration system, which allows central scheduling and management of all the network tests performed by the instances; a set of probes to continually gather metrics from the remote instances and publish it to different sources; a central network datastore (esmond), which provides interfaces to access the network monitoring information in close to real time and historically (up to a year) giving the state of the tests; and a perfSONAR infrastructure monitor system, ensuring the current perfSONAR instances are correctly configured and operating as intended. We will also describe the challenges we encountered in ongoing operations of the network service and how we have evolved our procedures to address those challenges. Finally we will describe our plans for future extensions and improvements to the service.},
 bibtype = {article},
 author = {Quick, Robert and Babik, Marian and Fajardo, Edgar M and Gross, Kyle and Hayashi, Soichi and Krenz, Marina and Lee, Thomas and Mckee, Shawn and Pipes, Christopher and Teige, Scott},
 doi = {10.1088/1742-6596/898/8/082044},
 journal = {Journal of Physics: Conf. Series}
}

The Open Science Grid (OSG) relies upon the network as a critical part of the distributed infrastructures it enables. In 2012, OSG added a new focus area in networking with a goal of becoming the primary source of network information for its members and collaborators. This includes gathering, organizing, and providing network metrics to guarantee effective network usage and prompt detection and resolution of any network issues, including connection failures, congestion, and traffic routing. In September of 2015, this service was deployed into the OSG production environment. We will report on the creation, implementation, testing, and deployment of the OSG Networking Service. Starting from organizing the deployment of perfSONAR toolkits within OSG and its partners, to the challenges of orchestrating regular testing between sites, to reliably gathering the resulting network metrics and making them available for users, virtual organizations, and higher level services, all aspects of implementation will be reviewed. In particular, several higher-level services were developed to bring the OSG network service to its full potential. These include a web-based mesh configuration system, which allows central scheduling and management of all the network tests performed by the instances; a set of probes to continually gather metrics from the remote instances and publish it to different sources; a central network datastore (esmond), which provides interfaces to access the network monitoring information in close to real time and historically (up to a year) giving the state of the tests; and a perfSONAR infrastructure monitor system, ensuring the current perfSONAR instances are correctly configured and operating as intended. We will also describe the challenges we encountered in ongoing operations of the network service and how we have evolved our procedures to address those challenges. Finally we will describe our plans for future extensions and improvements to the service.

@inproceedings{
 title = {Benchmarking Harp-DAAL: High Performance Hadoop on KNL Clusters},
 type = {inproceedings},
 year = {2017},
 keywords = {BigData,HPC,Xeon Phi},
 pages = {82-89},
 volume = {2017-June},
 websites = {http://ieeexplore.ieee.org/document/8030575/},
 month = {6},
 publisher = {IEEE},
 day = {8},
 id = {711b172a-6620-334b-96d4-aa9c1dd94ceb},
 created = {2020-09-09T19:45:43.999Z},
 accessed = {2020-09-09},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.120Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2017c},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {Data analytics is undergoing a revolution in many scientific domains, and demands cost-effective parallel data analysis techniques. Traditional Java-based Big Data processing tools like Hadoop MapReduce are designed for commodity CPUs. In contrast, emerging manycore processors like the Xeon Phi have an order of magnitude greater computation power and memory bandwidth. To harness their computing capabilities, we propose the Harp-DAAL framework. We show that enhanced versions of MapReduce can be replaced by Harp, a Hadoop plug-in, that offers useful data abstractions for both high-performance iterative computation and MPI-quality communication, as well as drive Intel's native DAAL library. We select a subset of three machine learning algorithms and implement them within Harp-DAAL. Our scalability benchmarks ran on Knights Landing (KNL) clusters and achieved up to 2.5 times speedup of performance over the HPC solution in NOMAD and 15 to 40 times speedup over Java-based solutions in Spark. We further quantify the workloads on single node KNL with a performance breakdown at the micro-architecture level.},
 bibtype = {inproceedings},
 author = {Chen, Langshi and Peng, Bo and Zhang, Bingjing and Liu, Tony and Zou, Yiming and Jiang, Lei and Henschel, Robert and Stewart, Craig and Zhang, Zhang and McCallum, Emily and Tom, Zahniser and Jon, Omer and Qiu, Judy},
 doi = {10.1109/CLOUD.2017.19},
 booktitle = {2017 IEEE 10th International Conference on Cloud Computing (CLOUD)}
}

Data analytics is undergoing a revolution in many scientific domains, and demands cost-effective parallel data analysis techniques. Traditional Java-based Big Data processing tools like Hadoop MapReduce are designed for commodity CPUs. In contrast, emerging manycore processors like the Xeon Phi have an order of magnitude greater computation power and memory bandwidth. To harness their computing capabilities, we propose the Harp-DAAL framework. We show that enhanced versions of MapReduce can be replaced by Harp, a Hadoop plug-in, that offers useful data abstractions for both high-performance iterative computation and MPI-quality communication, as well as drive Intel's native DAAL library. We select a subset of three machine learning algorithms and implement them within Harp-DAAL. Our scalability benchmarks ran on Knights Landing (KNL) clusters and achieved up to 2.5 times speedup of performance over the HPC solution in NOMAD and 15 to 40 times speedup over Java-based solutions in Spark. We further quantify the workloads on single node KNL with a performance breakdown at the micro-architecture level.

@techreport{
 title = {Jetstream Stakeholder Advisory Board Meeting February 2017: Presenters' Report},
 type = {techreport},
 year = {2017},
 keywords = {Jetstream,Technical Report,cloud computing},
 websites = {http://hdl.handle.net/2022/21247},
 month = {2},
 day = {28},
 id = {11c3daf1-43d2-37be-b2a5-6440bd5dbb1d},
 created = {2020-09-09T23:14:56.405Z},
 accessed = {2020-09-09},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.374Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2017c},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A.; and Hancock, David Y.; and Vaughn, Matthew; and Merchant, Nirav; and Lowe, John Michael; and Fischer, Jeremy; and Liming, Lee; and Taylor, James; and Turner, George; and Hammond, C. Bret; and Skidmore, Edwin; and Packard, Michael; and Miller, Therese; and Foster, Ian; and Rad, Paul; and Mehringer, Susan;}
}

@techreport{
 title = {Jetstream (NSF Award 1445604) Year Program Year 3 Annual Report (December 1, 2016 – November 28, 2017)},
 type = {techreport},
 year = {2017},
 keywords = {Technical Report},
 websites = {http://hdl.handle.net/2022/21854.},
 month = {12},
 day = {15},
 id = {e7d5c6c3-153a-3db5-a915-814e40974aa0},
 created = {2020-09-11T15:16:02.597Z},
 accessed = {2020-09-11},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:00.839Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2017d},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Hancock, David Y and Vaughn, Matthew and Merchant, Nirav and Fischer, Jeremy and Michael Lowe, J and Liming, Lee and Taylor, James and Afgan, Enis and Turner, George and Skidmore, Edwin and Packard, Michael and Beck, Brian W and Foster, Ian}
}

@techreport{
 title = {Summary of the Survey of Field/Marine Station Directors and Managers},
 type = {techreport},
 year = {2017},
 keywords = {Technical Report},
 websites = {http://hdl.handle.net/2022/21835},
 month = {7},
 day = {31},
 id = {9c70ae58-a072-3abc-a285-f9c810d480bc},
 created = {2020-09-11T16:53:24.775Z},
 accessed = {2020-09-11},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:00.898Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {DoakThomasG.;StewartCraigA.;WernertJulie;HancockDavidY.;Miller2017},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Doak, Thomas G.; Stewart, Craig A.; Wernert, Julie; Hancock, David Y.; Miller, Therese;}
}

@inproceedings{
 title = {Jetstream – performance, early experiences, and early results},
 type = {inproceedings},
 year = {2016},
 pages = {1-8},
 volume = {17-21-July},
 websites = {http://dl.acm.org/citation.cfm?doid=2949550.2949639},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {0315bdc4-e76a-31fe-a248-2568f3bb5455},
 created = {2017-10-24T15:01:35.358Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:33:20.142Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016b},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Jetstream is a first-of-a-kind system for the NSF - a distributed production cloud resource. The NSF awarded funds to create Jetstream in November 2014. Here we review the purpose for creating Jetstream, present the ac ceptance test results that define Jetstream's key characteristics, describe our experiences in standing up an OpenStack-based cloud environment, and share some of the early scientific results that have been obtained by researchers and students using this system. Jetstream offers unique capability within the XSEDE-supported US national cyberinfrastructure, delivering interactive virtual machines (VMs) via the Atmosphere interface developed by the University of Arizona. As a multi-region deployment that operates as a single integrated system, Jetstream is proving effective in supporting modes and disciplines of research traditionally underrepresented on larger XSEDE-supported clusters and supercomputers. Already, researchers in biology, network science, economics, earth science, and computer science have used Jetstream to perform research -much of it research in the "long tail of science.".},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Hancock, David Y. and Vaughn, Matthew and Fischer, Jeremy and Cockerill, Tim and Liming, Lee and Merchant, Nirav and Miller, Therese and Lowe, John Michael Stanzione, Daniel C. and Taylor, James and Skidmore, Edwin},
 doi = {10.1145/2949550.2949639},
 booktitle = {Proceedings of the XSEDE16 on Diversity, Big Data, and Science at Scale - XSEDE16}
}

Jetstream is a first-of-a-kind system for the NSF - a distributed production cloud resource. The NSF awarded funds to create Jetstream in November 2014. Here we review the purpose for creating Jetstream, present the ac ceptance test results that define Jetstream's key characteristics, describe our experiences in standing up an OpenStack-based cloud environment, and share some of the early scientific results that have been obtained by researchers and students using this system. Jetstream offers unique capability within the XSEDE-supported US national cyberinfrastructure, delivering interactive virtual machines (VMs) via the Atmosphere interface developed by the University of Arizona. As a multi-region deployment that operates as a single integrated system, Jetstream is proving effective in supporting modes and disciplines of research traditionally underrepresented on larger XSEDE-supported clusters and supercomputers. Already, researchers in biology, network science, economics, earth science, and computer science have used Jetstream to perform research -much of it research in the "long tail of science.".

@book{
 title = {Big data, simulations and HPC convergence},
 type = {book},
 year = {2016},
 source = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
 volume = {10044},
 id = {aa1607d2-b50b-381f-bc9f-26d6fca9a484},
 created = {2017-11-28T17:32:48.330Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.203Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Fox2016a},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {© Springer International Publishing AG 2016. Two major trends in computing systems are the growth in high performance computing (HPC) with in particular an international exascale initiative, and big data with an accompanying cloud infrastructure of dramatic and increasing size and sophistication. In this paper, we study an approach to convergence for software and applications/algorithms and show what hardware architectures it suggests. We start by dividing applications into data plus model components and classifying each component (whether from Big Data or Big Compute) in the same way. This leads to 64 properties divided into 4 views, which are Problem Architecture (Macro pattern); Execution Features (Micro patterns); Data Source and Style; and finally the Processing (runtime) View. We discuss convergence software built around HPC-ABDS (High Performance Computing enhanced Apache Big Data Stack) and show how one can merge Big Data and HPC (Big Simulation) concepts into a single stack and discuss appropriate hardware.},
 bibtype = {book},
 author = {Fox, Geoffrey Charles and Qiu, J. and Jha, S. and Ekanayake, S. and Kamburugamuve, S.},
 doi = {10.1007/978-3-319-49748-8_1}
}

© Springer International Publishing AG 2016. Two major trends in computing systems are the growth in high performance computing (HPC) with in particular an international exascale initiative, and big data with an accompanying cloud infrastructure of dramatic and increasing size and sophistication. In this paper, we study an approach to convergence for software and applications/algorithms and show what hardware architectures it suggests. We start by dividing applications into data plus model components and classifying each component (whether from Big Data or Big Compute) in the same way. This leads to 64 properties divided into 4 views, which are Problem Architecture (Macro pattern); Execution Features (Micro patterns); Data Source and Style; and finally the Processing (runtime) View. We discuss convergence software built around HPC-ABDS (High Performance Computing enhanced Apache Big Data Stack) and show how one can merge Big Data and HPC (Big Simulation) concepts into a single stack and discuss appropriate hardware.

@inproceedings{
 title = {Model-centric computation abstractions in machine learning applications},
 type = {inproceedings},
 year = {2016},
 pages = {3},
 publisher = {ACM},
 id = {4531a3f1-e263-3db4-9fe3-b72781a9c160},
 created = {2017-11-28T17:33:37.637Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.332Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2016},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Zhang, Bingjing and Peng, Bo and Qiu, Judy},
 booktitle = {Proceedings of the 3rd ACM SIGMOD Workshop on Algorithms and Systems for MapReduce and Beyond}
}

@techreport{
 title = {Software Frameworks for Deep Learning at Scale},
 type = {techreport},
 year = {2016},
 source = {Internal Indiana University Technical Report},
 id = {35faac0d-c465-34e2-bc4c-e997e068c6ef},
 created = {2017-11-28T17:33:38.688Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.299Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2016},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {techreport},
 author = {Fox, James and Zou, Yiming and Qiu, Judy}
}

@inbook{
 type = {inbook},
 year = {2016},
 pages = {37-62},
 publisher = {CRC Press},
 id = {d8d150f8-39c9-3456-b502-d0e3e3c345d1},
 created = {2017-11-28T17:33:40.638Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.057Z},
 read = {false},
 starred = {true},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wu2016},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inbook},
 author = {Wu, Tak-Lon Stephen and Zhang, Bingjing and Clayton Davis, Emilio and Ferrara, Alessandro Flammini and Menczer, Filippo and Qiu, Judy},
 chapter = {Scalable Query and Analysis for Social Networks: An Integrated High-Level Dataflow System with Pig and Harp},
 title = {Big Data in Complex and Social Networks}
}

@inproceedings{
 title = {Facilitating Visualization Capacity Building},
 type = {inproceedings},
 year = {2016},
 pages = {156},
 publisher = {ACM},
 id = {30c386e5-69a2-35a4-9c84-75b9bfef1f8a},
 created = {2017-12-18T18:01:16.866Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:02:56.781Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Byrd2016},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Byrd, Vetria L and Cottam, Joseph A},
 booktitle = {Proceedings of the 9th International Symposium on Visual Information Communication and Interaction}
}

@inproceedings{
 title = {Epoch Persistence: Safe, efficient, on-demand rendering for streaming data},
 type = {inproceedings},
 year = {2016},
 pages = {1730-1739},
 publisher = {IEEE},
 id = {c1f3a3e8-e2f3-3722-a239-50e4515788c5},
 created = {2017-12-18T18:01:19.410Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:15.921Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Cottam2016},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Cottam, Joseph A and Lumsdaine, Andrew},
 booktitle = {System Sciences (HICSS), 2016 49th Hawaii International Conference on}
}

@techreport{
 title = {The Web's PKI: An Expository Review and Certificate Validation Cost Simulation},
 type = {techreport},
 year = {2016},
 id = {e18d0500-3bd4-31c4-ac6a-7569cefe6a40},
 created = {2017-12-18T21:43:37.145Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.840Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Heiland2016a},
 source_type = {RPRT},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {techreport},
 author = {Heiland, Randy and Garrison III, William C and Qiao, Yechen and Lee, Adam J and Welch, Von}
}

@article{
 title = {High performance lda through collective model communication optimization},
 type = {article},
 year = {2016},
 pages = {86-97},
 volume = {80},
 publisher = {Elsevier},
 id = {acb13c9b-3c97-3691-bfa7-5ada1a1da94c},
 created = {2018-01-09T20:30:37.407Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.822Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2016a},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {© The Authors. Published by Elsevier B.V. LDA is a widely used machine learning technique for big data analysis. The application includes an inference algorithm that iteratively updates a model until it converges. A major challenge is the scaling issue in parallelization owing to the fact that the model size is huge and parallel workers need to communicate the model continually. We identify three important features of the model in parallel LDA computation: 1. The volume of model parameters required for local computation is high; 2. The time complexity of local computation is proportional to the required model size; 3. The model size shrinks as it converges. By investigating collective and asynchronous methods for model communication in di erent tools, we discover that optimized collective communication can improve the model update speed, thus allowing the model to converge faster. The performance improvement derives not only from accelerated communication but also from reduced iteration computation time as the model size shrinks during the model convergence. To foster faster model convergence, we design new collective communication abstractions and implement two Harp-LDA applications, lgs and rtt. We compare our new approach with Yahoo! LDA and Petuum LDA, two leading implementations favoring asynchronous communication methods in the eld, on a 100-node, 4000-thread Intel Haswell cluster. The experiments show that lgs can reach higher model likelihood with shorter or similar execution time compared with Yahoo! LDA, while rtt can run up to 3.9 times faster compared with Petuum LDA when achieving similar model likelihood.},
 bibtype = {article},
 author = {Zhang, Bingjing and Peng, B. Bo and Qiu, Judy},
 doi = {10.1016/j.procs.2016.05.300},
 journal = {Procedia Computer Science}
}

© The Authors. Published by Elsevier B.V. LDA is a widely used machine learning technique for big data analysis. The application includes an inference algorithm that iteratively updates a model until it converges. A major challenge is the scaling issue in parallelization owing to the fact that the model size is huge and parallel workers need to communicate the model continually. We identify three important features of the model in parallel LDA computation: 1. The volume of model parameters required for local computation is high; 2. The time complexity of local computation is proportional to the required model size; 3. The model size shrinks as it converges. By investigating collective and asynchronous methods for model communication in di erent tools, we discover that optimized collective communication can improve the model update speed, thus allowing the model to converge faster. The performance improvement derives not only from accelerated communication but also from reduced iteration computation time as the model size shrinks during the model convergence. To foster faster model convergence, we design new collective communication abstractions and implement two Harp-LDA applications, lgs and rtt. We compare our new approach with Yahoo! LDA and Petuum LDA, two leading implementations favoring asynchronous communication methods in the eld, on a 100-node, 4000-thread Intel Haswell cluster. The experiments show that lgs can reach higher model likelihood with shorter or similar execution time compared with Yahoo! LDA, while rtt can run up to 3.9 times faster compared with Petuum LDA when achieving similar model likelihood.

@inproceedings{
 title = {Computational considerations in transcriptome assemblies and their evaluation, using high quality human RNA-Seq data},
 type = {inproceedings},
 year = {2016},
 keywords = {Big data; Information management; RNA,Computational aspects; Computational demands; Cur,Quality control},
 volume = {17-21-July},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84989162194&doi=10.1145%2F2949550.2949572&partnerID=40&md5=7c2901a1f39cfb52b792e1634f14569e},
 publisher = {Association for Computing Machinery},
 id = {6882dd4c-afea-37be-984a-1d14c1f563fe},
 created = {2018-01-09T20:30:40.489Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:17.623Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ghaffari2016},
 source_type = {conference},
 notes = {cited By 0; Conference of Conference on Diversity, Big Data, and Science at Scale, XSEDE 2016 ; Conference Date: 17 July 2016 Through 21 July 2016; Conference Code:123713},
 folder_uuids = {44d48ba6-09b0-4f17-965c-b625aaaba978},
 private_publication = {false},
 abstract = {It is crucial to understand the performance of transcriptome assemblies to improve current practices. Investigating the factors that affect a transcriptome assembly is very important and is the primary goal of our project. To that end, we designed a multi-step pipeline consisting of variety of pre-processing and quality control steps. XSEDE allocations enabled us to achieve the computational demands of the project. The high memory Blacklight and Greenfield systems at Pittsburgh Supercomputing Center were essential to accomplish multiple steps of this project. This paper presents the computational aspects of our comprehensive transcriptome assembly and validation study.},
 bibtype = {inproceedings},
 author = {Ghaffari, N and Abante, J and Singh, R and Blood, P D and Johnson, C D},
 doi = {10.1145/2949550.2949572},
 booktitle = {ACM International Conference Proceeding Series}
}

It is crucial to understand the performance of transcriptome assemblies to improve current practices. Investigating the factors that affect a transcriptome assembly is very important and is the primary goal of our project. To that end, we designed a multi-step pipeline consisting of variety of pre-processing and quality control steps. XSEDE allocations enabled us to achieve the computational demands of the project. The high memory Blacklight and Greenfield systems at Pittsburgh Supercomputing Center were essential to accomplish multiple steps of this project. This paper presents the computational aspects of our comprehensive transcriptome assembly and validation study.

@article{
 title = {OSoMe: the IUNI observatory on social media},
 type = {article},
 year = {2016},
 pages = {e87},
 volume = {2},
 publisher = {PeerJ Inc.},
 id = {cd24ef68-6b09-3ec7-a215-a6f79e4e5f40},
 created = {2018-01-23T20:26:56.472Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:39.609Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Davis2016},
 source_type = {JOUR},
 folder_uuids = {2aba6c14-9027-4f47-8627-0902e1e2342b,971e10ae-b1f7-4c4b-8f99-aa55f77a64f6,36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {article},
 author = {Davis, Clayton A and Ciampaglia, Giovanni Luca and Aiello, Luca Maria and Chung, Keychul and Conover, Michael D and Ferrara, Emilio and Flammini, Alessandro and Fox, Geoffrey Charles and Gao, Xiaoming and Gonçalves, Bruno},
 journal = {PeerJ Computer Science}
}

@article{
 title = {Genomic and metagenomic analysis of diversity-generating retroelements associated with Treponema denticola},
 type = {article},
 year = {2016},
 keywords = {Article,amino acid substitution,bacterial genom,lipoprotein},
 pages = {pp 852},
 volume = {7},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84980008617&doi=10.3389%2Ffmicb.2016.00852&partnerID=40&md5=60c153d3e939419432ed9b21aa570c92,https://www.frontiersin.org/articles/10.3389/fmicb.2016.00852/full},
 publisher = {Frontiers Research Foundation},
 id = {0b924b58-3cbb-3f34-909f-57b61a0dc76f},
 created = {2019-10-01T17:20:30.593Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:30.593Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Nimkulrat2016},
 source_type = {article},
 notes = {cited By 1},
 private_publication = {false},
 abstract = {Diversity-generating retroelements (DGRs) are genetic cassettes that can produce massive protein sequence variation in prokaryotes. Presumably DGRs confer selective advantages to their hosts (bacteria or viruses) by generating variants of target genes-typically resulting in target proteins with altered ligand-binding specificity-through a specialized error-prone reverse transcription process. The only extensively studied DGR system is from the Bordetella phage BPP-1, although DGRs are predicted to exist in other species. Using bioinformatics analysis, we discovered that the DGR system associated with the Treponema denticola species (a human oral-associated periopathogen) is dynamic (with gains/losses of the system found in the isolates) and diverse (with multiple types found in isolated genomes and the human microbiota). The T. denticola DGR is found in only nine of the 17 sequenced T. denticola strains. Analysis of the DGR-associated template regions and reverse transcriptase gene sequences revealed two types of DGR systems in T. denticola: The ATCC35405-type shared by seven isolates including ATCC35405; and the SP32-type shared by two isolates (SP32 and SP33), suggesting multiple DGR acquisitions. We detected additional variants of the T. denticola DGR systems in the human microbiomes, and found that the SP32-type DGR is more abundant than the ATCC35405-type in the healthy human oral microbiome, although the latter is found in more sequenced isolates. This is the first comprehensive study to characterize the DGRs associated with T. denticola in individual genomes as well as human microbiomes, demonstrating the importance of utilizing both individual genomes and metagenomes for characterizing the elements, and for analyzing their diversity and distribution in human populations. © 2016 Nimkulrat, Lee, Doak and Ye.},
 bibtype = {article},
 author = {Nimkulrat, S and Lee, H and Doak, T G and Ye, Y},
 doi = {10.3389/fmicb.2016.00852},
 journal = {Frontiers in Microbiology},
 number = {JUN}
}

Diversity-generating retroelements (DGRs) are genetic cassettes that can produce massive protein sequence variation in prokaryotes. Presumably DGRs confer selective advantages to their hosts (bacteria or viruses) by generating variants of target genes-typically resulting in target proteins with altered ligand-binding specificity-through a specialized error-prone reverse transcription process. The only extensively studied DGR system is from the Bordetella phage BPP-1, although DGRs are predicted to exist in other species. Using bioinformatics analysis, we discovered that the DGR system associated with the Treponema denticola species (a human oral-associated periopathogen) is dynamic (with gains/losses of the system found in the isolates) and diverse (with multiple types found in isolated genomes and the human microbiota). The T. denticola DGR is found in only nine of the 17 sequenced T. denticola strains. Analysis of the DGR-associated template regions and reverse transcriptase gene sequences revealed two types of DGR systems in T. denticola: The ATCC35405-type shared by seven isolates including ATCC35405; and the SP32-type shared by two isolates (SP32 and SP33), suggesting multiple DGR acquisitions. We detected additional variants of the T. denticola DGR systems in the human microbiomes, and found that the SP32-type DGR is more abundant than the ATCC35405-type in the healthy human oral microbiome, although the latter is found in more sequenced isolates. This is the first comprehensive study to characterize the DGRs associated with T. denticola in individual genomes as well as human microbiomes, demonstrating the importance of utilizing both individual genomes and metagenomes for characterizing the elements, and for analyzing their diversity and distribution in human populations. © 2016 Nimkulrat, Lee, Doak and Ye.

@inproceedings{
 title = {The advanced cyberinfrastructure research and education facilitators virtual residency: Toward a national cyberinfrastructure workforce},
 type = {inproceedings},
 year = {2016},
 keywords = {Cyberinfrastructure,Train the trainer,Workforce development},
 volume = {17-21-July},
 month = {7},
 publisher = {Association for Computing Machinery},
 day = {17},
 id = {44c44feb-540c-3336-be39-f766bd630031},
 created = {2019-10-01T17:20:31.774Z},
 accessed = {2019-08-27},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.898Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Neeman2016},
 private_publication = {false},
 abstract = {An Advanced Cyberinfrastructure Research and Education Facilitator (ACI-REF) works directly with researchers to advance the computing- and data-intensive aspects of their research, helping them to make effective use of Cyberinfrastructure (CI). The University of Oklahoma (OU) is leading a national "virtual residency" program to prepare ACI-REFs to provide CI facilitation to the diverse populations of Science, Technology, Engineering and Mathematics (STEM) researchers that they serve. Until recently, CI Facilitators have had no education or training program; the Virtual Residency program addresses this national need by providing: (1) training, specifically (a) summer workshops and (b) third party training opportunity alerts; (2) a community of CI Facilitators, enabled by (c) a biweekly conference call and (d) a mailing list.},
 bibtype = {inproceedings},
 author = {Neeman, Henry and Ganote, Carrie and Kalescky, Robert and Ramadugu, Sai Kumar and Sherman, Andrew H. and Bergstrom, Aaron and Gray, Zane and Lemley, Evan and Romanella, Alana and Stengel, Brian and Brunson, Dana and Guilfoos, Brian and Moore, Brian G. and Rush, Johnathan and Voss, Dan},
 doi = {10.1145/2949550.2949584},
 booktitle = {Proceedings of the XSEDE16 Conference on Diversity, Big Data, and Science at Scale (XSEDE16)}
}

An Advanced Cyberinfrastructure Research and Education Facilitator (ACI-REF) works directly with researchers to advance the computing- and data-intensive aspects of their research, helping them to make effective use of Cyberinfrastructure (CI). The University of Oklahoma (OU) is leading a national "virtual residency" program to prepare ACI-REFs to provide CI facilitation to the diverse populations of Science, Technology, Engineering and Mathematics (STEM) researchers that they serve. Until recently, CI Facilitators have had no education or training program; the Virtual Residency program addresses this national need by providing: (1) training, specifically (a) summer workshops and (b) third party training opportunity alerts; (2) a community of CI Facilitators, enabled by (c) a biweekly conference call and (d) a mailing list.

@article{
 title = {Disentangling the taxonomy of Rickettsiales and description of two novel symbionts ("Candidatus Bealeia paramacronuclearis" and "Candidatus Fokinia cryptica") sharing the cytoplasm of the ciliate protist Paramecium biaurelia},
 type = {article},
 year = {2016},
 keywords = {Alphaproteobacteria; Anaplasmataceae; Bacteria (m,Bacteria,Cells; Cytology; Oligonucleotides; RNA,Endosymbiotic; Gram-negative bacteria; Molecular,bacterium; cells and cell components; cytoplasm;},
 pages = {7236-7247},
 volume = {82},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84997402854&doi=10.1128%2FAEM.02284-16&partnerID=40&md5=73a151d55bc8d5e37294b554f26a6128},
 publisher = {American Society for Microbiology},
 id = {4530a5da-8de9-3d3d-9824-7160e28fe609},
 created = {2019-10-01T17:20:32.047Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:32.047Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Szokoli20167236},
 source_type = {article},
 notes = {cited By 1},
 private_publication = {false},
 abstract = {In the past 10 years, the number of endosymbionts described within the bacterial order Rickettsiales has constantly grown. Since 2006, 18 novel Rickettsiales genera inhabiting protists, such as ciliates and amoebae, have been described. In this work, we characterize two novel bacterial endosymbionts from Paramecium collected near Bloomington, IN. Both endosymbiotic species inhabit the cytoplasm of the same host. The Gram-negative bacterium "Candidatus Bealeia paramacronuclearis" occurs in clumps and is frequently associated with the host macronucleus. With its electron-dense cytoplasm and a distinct halo surrounding the cell, it is easily distinguishable from the second smaller symbiont, "Candidatus Fokinia cryptica," whose cytoplasm is electron lucid, lacks a halo, and is always surrounded by a symbiontophorous vacuole. For molecular characterization, the small-subunit rRNA genes were sequenced and used for taxonomic assignment as well as the design of species-specific oligonucleotide probes. Phylogenetic analyses revealed that "Candidatus Bealeia paramacronuclearis" clusters with the so-called "basal" Rickettsiales, and "Candidatus Fokinia cryptica" belongs to "Candidatus Midichloriaceae." We obtained tree topologies showing a separation of Rickettsiales into at least two groups: one represented by the families Rickettsiaceae, Anaplasmataceae, and "Candidatus Midichloriaceae" (RAM clade), and the other represented by "basal Rickettsiales," including "Candidatus Bealeia paramacronuclearis." Therefore, and in accordance with recent publications, we propose to limit the order Rickettsiales to the RAM clade and to raise "basal Rickettsiales" to an independent order, Holosporales ord. nov., inside Alphaproteobacteria, which presently includes four family-level clades. Additionally, we define the family "Candidatus Hepatincolaceae" and redefine the family Holosporaceae. © 2016, American Society for Microbiology. All Rights Reserved.},
 bibtype = {article},
 author = {Szokoli, F and Castelli, M and Sabaneyeva, E and Schrallhammer, M and Krenek, S and Doak, T G and Berendonk, T U and Petroni, G},
 doi = {10.1128/AEM.02284-16},
 journal = {Applied and Environmental Microbiology},
 number = {24}
}

In the past 10 years, the number of endosymbionts described within the bacterial order Rickettsiales has constantly grown. Since 2006, 18 novel Rickettsiales genera inhabiting protists, such as ciliates and amoebae, have been described. In this work, we characterize two novel bacterial endosymbionts from Paramecium collected near Bloomington, IN. Both endosymbiotic species inhabit the cytoplasm of the same host. The Gram-negative bacterium "Candidatus Bealeia paramacronuclearis" occurs in clumps and is frequently associated with the host macronucleus. With its electron-dense cytoplasm and a distinct halo surrounding the cell, it is easily distinguishable from the second smaller symbiont, "Candidatus Fokinia cryptica," whose cytoplasm is electron lucid, lacks a halo, and is always surrounded by a symbiontophorous vacuole. For molecular characterization, the small-subunit rRNA genes were sequenced and used for taxonomic assignment as well as the design of species-specific oligonucleotide probes. Phylogenetic analyses revealed that "Candidatus Bealeia paramacronuclearis" clusters with the so-called "basal" Rickettsiales, and "Candidatus Fokinia cryptica" belongs to "Candidatus Midichloriaceae." We obtained tree topologies showing a separation of Rickettsiales into at least two groups: one represented by the families Rickettsiaceae, Anaplasmataceae, and "Candidatus Midichloriaceae" (RAM clade), and the other represented by "basal Rickettsiales," including "Candidatus Bealeia paramacronuclearis." Therefore, and in accordance with recent publications, we propose to limit the order Rickettsiales to the RAM clade and to raise "basal Rickettsiales" to an independent order, Holosporales ord. nov., inside Alphaproteobacteria, which presently includes four family-level clades. Additionally, we define the family "Candidatus Hepatincolaceae" and redefine the family Holosporaceae. © 2016, American Society for Microbiology. All Rights Reserved.

@article{
 title = {The Rate and Spectrum of Spontaneous Mutations in Mycobacterium smegmatis, a Bacterium Naturally Devoid of the Postreplicative Mismatch Repair Pathway.},
 type = {article},
 year = {2016},
 keywords = {GC bias,Mycobacteria,mutation accumulation},
 pages = {2157-63},
 volume = {6},
 websites = {http://www.ncbi.nlm.nih.gov/pubmed/27194804,http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC4938668},
 day = {7},
 id = {e6ed6425-76f5-39c1-baa7-7293a64ddcf8},
 created = {2019-10-01T17:20:32.334Z},
 accessed = {2019-08-20},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.969Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kucukyildirim2016},
 private_publication = {false},
 abstract = {Mycobacterium smegmatis is a bacterium that is naturally devoid of known postreplicative DNA mismatch repair (MMR) homologs, mutS and mutL, providing an opportunity to investigate how the mutation rate and spectrum has evolved in the absence of a highly conserved primary repair pathway. Mutation accumulation experiments of M. smegmatis yielded a base-substitution mutation rate of 5.27 × 10(-10) per site per generation, or 0.0036 per genome per generation, which is surprisingly similar to the mutation rate in MMR-functional unicellular organisms. Transitions were found more frequently than transversions, with the A:T→G:C transition rate significantly higher than the G:C→A:T transition rate, opposite to what is observed in most studied bacteria. We also found that the transition-mutation rate of M. smegmatis is significantly lower than that of other naturally MMR-devoid or MMR-knockout organisms. Two possible candidates that could be responsible for maintaining high DNA fidelity in this MMR-deficient organism are the ancestral-like DNA polymerase DnaE1, which contains a highly efficient DNA proofreading histidinol phosphatase (PHP) domain, and/or the existence of a uracil-DNA glycosylase B (UdgB) homolog that might protect the GC-rich M. smegmatis genome against DNA damage arising from oxidation or deamination. Our results suggest that M. smegmatis has a noncanonical Dam (DNA adenine methylase) methylation system, with target motifs differing from those previously reported. The mutation features of M. smegmatis provide further evidence that genomes harbor alternative routes for improving replication fidelity, even in the absence of major repair pathways.},
 bibtype = {article},
 author = {Kucukyildirim, Sibel and Long, Hongan and Sung, Way and Miller, Samuel F and Doak, Thomas G and Lynch, Michael},
 doi = {10.1534/g3.116.030130},
 journal = {G3: Genes, Genomes, Genetics},
 number = {7}
}

Mycobacterium smegmatis is a bacterium that is naturally devoid of known postreplicative DNA mismatch repair (MMR) homologs, mutS and mutL, providing an opportunity to investigate how the mutation rate and spectrum has evolved in the absence of a highly conserved primary repair pathway. Mutation accumulation experiments of M. smegmatis yielded a base-substitution mutation rate of 5.27 × 10(-10) per site per generation, or 0.0036 per genome per generation, which is surprisingly similar to the mutation rate in MMR-functional unicellular organisms. Transitions were found more frequently than transversions, with the A:T→G:C transition rate significantly higher than the G:C→A:T transition rate, opposite to what is observed in most studied bacteria. We also found that the transition-mutation rate of M. smegmatis is significantly lower than that of other naturally MMR-devoid or MMR-knockout organisms. Two possible candidates that could be responsible for maintaining high DNA fidelity in this MMR-deficient organism are the ancestral-like DNA polymerase DnaE1, which contains a highly efficient DNA proofreading histidinol phosphatase (PHP) domain, and/or the existence of a uracil-DNA glycosylase B (UdgB) homolog that might protect the GC-rich M. smegmatis genome against DNA damage arising from oxidation or deamination. Our results suggest that M. smegmatis has a noncanonical Dam (DNA adenine methylase) methylation system, with target motifs differing from those previously reported. The mutation features of M. smegmatis provide further evidence that genomes harbor alternative routes for improving replication fidelity, even in the absence of major repair pathways.

@article{
 title = {Insertion sequence-caused large-scale rearrangements in the genome of Escherichia coli},
 type = {article},
 year = {2016},
 keywords = {Article; bacterial genome; controlled study; DNA f,Bacterial; Mutagenesis,Base Sequence; DNA Transposable Elements; Escheri,Insertional,Molecular; Genome,transposon},
 pages = {7109-7119},
 volume = {44},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84988377984&doi=10.1093%2Fnar%2Fgkw647&partnerID=40&md5=8bf10db5b54055c6f322feff9d7a4dc2},
 publisher = {Oxford University Press},
 id = {36a0068e-1055-3de5-9426-0415bfffdcc8},
 created = {2019-10-01T17:20:34.205Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:34.205Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lee20167109},
 source_type = {article},
 notes = {cited By 7},
 private_publication = {false},
 abstract = {A majority of large-scale bacterial genome rearrangements involve mobile genetic elements such as insertion sequence (IS) elements. Here we report novel insertions and excisions of IS elements and recombination between homologous IS elements identified in a large collection of Escherichia coli mutation accumulation lines by analysis of whole genome shotgun sequencing data. Based on 857 identified events (758 IS insertions, 98 recombinations and 1 excision), we estimate that the rate of IS insertion is 3.5 × 10-4 insertions per genome per generation and the rate of IS homologous recombination is 4.5 × 10-5 recombinations per genome per generation. These events are mostly contributed by the IS elements IS1, IS2, IS5 and IS186. Spatial analysis of new insertions suggest that transposition is biased to proximal insertions, and the length spectrum of IS-caused deletions is largely explained by local hopping. For any of the ISs studied there is no region of the circular genome that is favored or disfavored for new insertions but there are notable hotspots for deletions. Some elements have preferences for non-coding sequence or for the beginning and end of coding regions, largely explained by target site motifs. Interestingly, transposition and deletion rates remain constant across the wild-type and 12 mutant E. coli lines, each deficient in a distinct DNA repair pathway. Finally, we characterized the target sites of four IS families, confirming previous results and characterizing a highly specific pattern at IS186 target-sites, 5′-GGGG(N6/N7)CCCC-3′. We also detected 48 long deletions not involving IS elements. © 2016 The Author(s) 2016. Published by Oxford University Press on behalf of Nucleic Acids Research.},
 bibtype = {article},
 author = {Lee, H and Doak, T G and Popodi, E and Foster, P L and Tang, H},
 doi = {10.1093/nar/gkw647},
 journal = {Nucleic Acids Research},
 number = {15}
}

A majority of large-scale bacterial genome rearrangements involve mobile genetic elements such as insertion sequence (IS) elements. Here we report novel insertions and excisions of IS elements and recombination between homologous IS elements identified in a large collection of Escherichia coli mutation accumulation lines by analysis of whole genome shotgun sequencing data. Based on 857 identified events (758 IS insertions, 98 recombinations and 1 excision), we estimate that the rate of IS insertion is 3.5 × 10-4 insertions per genome per generation and the rate of IS homologous recombination is 4.5 × 10-5 recombinations per genome per generation. These events are mostly contributed by the IS elements IS1, IS2, IS5 and IS186. Spatial analysis of new insertions suggest that transposition is biased to proximal insertions, and the length spectrum of IS-caused deletions is largely explained by local hopping. For any of the ISs studied there is no region of the circular genome that is favored or disfavored for new insertions but there are notable hotspots for deletions. Some elements have preferences for non-coding sequence or for the beginning and end of coding regions, largely explained by target site motifs. Interestingly, transposition and deletion rates remain constant across the wild-type and 12 mutant E. coli lines, each deficient in a distinct DNA repair pathway. Finally, we characterized the target sites of four IS families, confirming previous results and characterizing a highly specific pattern at IS186 target-sites, 5′-GGGG(N6/N7)CCCC-3′. We also detected 48 long deletions not involving IS elements. © 2016 The Author(s) 2016. Published by Oxford University Press on behalf of Nucleic Acids Research.

@techreport{
 title = {System Acceptance Report for NSF award 1445604” High Performance Computing System Acquisition: Jetstream-A Self-Provisioned, Scalable Science and Engineering Cloud Environment”},
 type = {techreport},
 year = {2016},
 id = {7e6d7379-504f-3bd8-b9b9-d6dc3a8d646d},
 created = {2019-10-01T17:20:35.031Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:34.213Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016e},
 source_type = {RPRT},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Hancock, David Y and Vaughn, Matthew and Merchant, Nirav C and Lowe, John Michael and Fischer, Jeremy and Liming, Lee and Taylor, James and Afgan, Enis and Turner, George}
}

@techreport{
 title = {Jetstream (NSF Award 1445604) Year Program Year 2 Annual Report (Dec 1, 2015–Nov 30, 2016)},
 type = {techreport},
 year = {2016},
 websites = {http://hdl.handle.net/2022/21194},
 id = {04e3c85e-d483-347e-aabb-59d9aa1eef0f},
 created = {2019-10-01T17:20:35.499Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-11T15:17:22.594Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016i},
 source_type = {RPRT},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Hancock, David Y and Vaughn, Matthew and Merchant, Nirav and Lowe, John Michael and Fischer, Jeremy and Liming, Lee and Taylor, James and Afgan, Enis and Turner, George}
}

@article{
 title = {From describing to prescribing parallelism: Translating the SPEC ACCEL OpenACC suite to OpenMP target directives},
 type = {article},
 year = {2016},
 keywords = {Application programming interfaces (API),Benchmarking,Codes (s,Offloading,OpenMP,Openacc,SPEC,SPEC ACCEL},
 pages = {470-488},
 volume = {9945 LNCS},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84992549244&doi=10.1007%2F978-3-319-46079-6_33&partnerID=40&md5=9655cf681ac699c57e5a2bdef2a48e0c},
 publisher = {Springer Verlag},
 id = {61bf5bfe-ccf6-3949-93c5-87e33f167000},
 created = {2019-10-01T17:20:36.936Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:25.871Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Juckeland2016470},
 source_type = {article},
 notes = {cited By 0; Conference of International Workshops on High Performance Computing, ISC High Performance 2016 and Workshop on 2nd International Workshop on Communication Architectures at Extreme Scale, ExaComm 2016, Workshop on Exascale Multi/Many Core Computing Systems, E-MuCoCoS 2016, HPC I/O in the Data Center, HPC-IODC 2016, Application Performance on Intel Xeon Phi – Being Prepared for KNL and Beyond, IXPUG 2016, International Workshop on OpenPOWER for HPC, IWOPH 2016, International Workshop on Performance Portable Programming Models for Accelerators, P^3MA 2016, Workshop on Virtualization in High-Performance Cloud Computing, VHPC 2016, Workshop on Performance and Scalability of Storage Systems, WOPSSS 2016 ; Conference Date: 19 June 2016 Through 23 June 2016; Conference Code:185039},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Current and next generation HPC systems will exploit accelerators and self-hosting devices within their compute nodes to accelerate applications. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. One of the goals of OpenMP and OpenACC is to allow the user to specify parallelism via directives so that compilers can generate device specific code and optimizations. However, the challenge of porting codes becomes more complex because of the different types of parallelism and memory hierarchies available on different architectures. In this paper we discuss our experience with porting the SPEC ACCEL benchmarks from OpenACC to OpenMP 4.5 using a performance portable style that lets the compiler make platform-specific optimizations to achieve good performance on a variety of systems. The ported SPEC ACCEL OpenMP benchmarks were validated on different platforms including Xeon Phi, GPUs and CPUs. We believe that this experience can help the community and compiler vendors understand how users plan to write OpenMP 4.5 applications in a performance portable style. © Springer International Publishing AG 2016.},
 bibtype = {article},
 author = {Juckeland, G and Hernandez, O and Jacob, A C and Neilson, D and Larrea, V G V and Wienke, S and Bobyr, A and Brantley, W C and Chandrasekaran, S and Colgrove, M and Grund, A and Henschel, R and Joubert, W and Müller, M S and Raddatz, D and Shelepugin, P and Whitney, B and Wang, B and Kumaran, K},
 editor = {Mohr B. Kunkel J.M., Taufer M},
 doi = {10.1007/978-3-319-46079-6_33},
 journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}
}

Current and next generation HPC systems will exploit accelerators and self-hosting devices within their compute nodes to accelerate applications. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. One of the goals of OpenMP and OpenACC is to allow the user to specify parallelism via directives so that compilers can generate device specific code and optimizations. However, the challenge of porting codes becomes more complex because of the different types of parallelism and memory hierarchies available on different architectures. In this paper we discuss our experience with porting the SPEC ACCEL benchmarks from OpenACC to OpenMP 4.5 using a performance portable style that lets the compiler make platform-specific optimizations to achieve good performance on a variety of systems. The ported SPEC ACCEL OpenMP benchmarks were validated on different platforms including Xeon Phi, GPUs and CPUs. We believe that this experience can help the community and compiler vendors understand how users plan to write OpenMP 4.5 applications in a performance portable style. © Springer International Publishing AG 2016.

@inproceedings{
 title = {Situating cyberinfrastructure in the public realm: The teragrid and XSEDE projects},
 type = {inproceedings},
 year = {2016},
 keywords = {C,Case study methodologies,Complex networks,Cy-berinfrastructure,Sustainable development},
 pages = {129-135},
 volume = {08-10-June},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84978763642&doi=10.1145%2F2912160.2912209&partnerID=40&md5=1b96431d39ddd31f8ff8beb57840aa2f},
 publisher = {Association for Computing Machinery},
 id = {fbd2cf7f-8345-3055-8236-4c7676e3f06f},
 created = {2019-10-01T17:20:36.974Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:25.578Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper2016129},
 source_type = {conference},
 notes = {cited By 0; Conference of 17th Annual International Conference on Digital Government Research, dg.o 2016 ; Conference Date: 8 June 2016 Through 10 June 2016; Conference Code:122243},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {We examine the evolution of one NSF-funded cyberinfras-tructure organization, the TeraGrid, into its successor or-ganization, the Extreme Science and Engineering Discovery Environment (XSEDE) via a case study methodology with two di-erent frames of reference: virtual organization lit-erature and public management networks theory. Our aim in examining these two projects is to situate the study of cyberinfrastructure in support of basic science in both the virtual organization and public management networks lit-erature. We -nd that our cases con-rm the utility of hi-erarchy when dealing with complexity and that expert use of ICT for governance tasks improves network capacity, es-pecially when time-Tested technologies are available in the environment. These ICT resources can be used for moni-toring of both service delivery functions as well as network health. We also note that the establishment of common inter-organizational standards facilitates interoperability be-tween organizations, allowing sustainability and modularity of network members. © 2016 ACM.},
 bibtype = {inproceedings},
 author = {Knepper, R and Chen, Y.-C.},
 editor = {Kim Y., Liu S M},
 doi = {10.1145/2912160.2912209},
 booktitle = {ACM International Conference Proceeding Series}
}

We examine the evolution of one NSF-funded cyberinfras-tructure organization, the TeraGrid, into its successor or-ganization, the Extreme Science and Engineering Discovery Environment (XSEDE) via a case study methodology with two di-erent frames of reference: virtual organization lit-erature and public management networks theory. Our aim in examining these two projects is to situate the study of cyberinfrastructure in support of basic science in both the virtual organization and public management networks lit-erature. We -nd that our cases con-rm the utility of hi-erarchy when dealing with complexity and that expert use of ICT for governance tasks improves network capacity, es-pecially when time-Tested technologies are available in the environment. These ICT resources can be used for moni-toring of both service delivery functions as well as network health. We also note that the establishment of common inter-organizational standards facilitates interoperability be-tween organizations, allowing sustainability and modularity of network members. © 2016 ACM.

@article{
 title = {Innovations from the early user phase on the Jetstream Research Cloud},
 type = {article},
 year = {2016},
 id = {d6bead14-bb57-3b04-afc1-3d9a9466f018},
 created = {2019-10-01T17:20:37.213Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:23.744Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper2016a},
 source_type = {JOUR},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {article},
 author = {Knepper, Richard and Fischer, Jeremy and Stewart, Craig A and Hancock, David Y and Link, Matthew R},
 journal = {2016 Federated Conference on Computer Science and Information Systems}
}

@techreport{
 title = {High Performance Computing System Acquisition: Jetstream–A Self-Provisioned, Scalable Science and Engineering Cloud Environment (Year 1 Annual Report)},
 type = {techreport},
 year = {2016},
 websites = {http://hdl.handle.net/2022/20746},
 id = {514245f5-f7f7-3161-9dc7-ae6f1b328806},
 created = {2019-10-01T17:20:38.238Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-11T15:47:58.870Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016f},
 source_type = {RPRT},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Foster, Ian and Merchant, Nirav C and Taylor, James and Vaughn, Matthew W}
}

@inproceedings{
 title = {Improving the scalability of a charge detection mass spectrometry workflow},
 type = {inproceedings},
 year = {2016},
 keywords = {Application performance,Big data,Charge detection,Chemic,Chemical compounds,Chemical detection,Mass spect},
 pages = {8},
 volume = {17-21-July},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84989172207&doi=10.1145%2F2949550.2949563&partnerID=40&md5=72e130c503d903ca3cc63cfe00eb39e1},
 publisher = {Association for Computing Machinery},
 id = {7b7626e9-0a17-39c6-8f73-55a870af8949},
 created = {2019-10-01T17:20:38.466Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:50.103Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {McClary2016},
 source_type = {conference},
 notes = {cited By 0; Conference of Conference on Diversity, Big Data, and Science at Scale, XSEDE 2016 ; Conference Date: 17 July 2016 Through 21 July 2016; Conference Code:123713},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {The Indiana University (IU) Department of Chemistry's Martin F. Jarrold (MFJ) Research Group studies a specialized technique of mass spectrometry called Charge Detection Mass Spectrometry (CDMS). The goal of mass spectrometry is to determine the mass of chemical and biological compounds, and with CDMS, the MFJ Research Group is extending the upper limit of mass detection. These researchers have developed a scientific application, which accurately analyzes raw CDMS data generated from their mass spectrometer. This paper explains the comprehensive process of optimizing the group's workflow by improving both the latency and throughput of their CDMS application. These significant performance improvements enabled high efficiency and scalability across IU's Advanced Cyberinfrastructure; overall, this analysis and development resulted in a 25x speedup of the application. © 2016 ACM.},
 bibtype = {inproceedings},
 author = {McClary, S and Henschel, R and Thota, A and Brunst, H and Draper, B},
 doi = {10.1145/2949550.2949563},
 booktitle = {Proceedings of the XSEDE16 Conference on Diversity, Big Data, and Science at Scale (XSEDE16)}
}

The Indiana University (IU) Department of Chemistry's Martin F. Jarrold (MFJ) Research Group studies a specialized technique of mass spectrometry called Charge Detection Mass Spectrometry (CDMS). The goal of mass spectrometry is to determine the mass of chemical and biological compounds, and with CDMS, the MFJ Research Group is extending the upper limit of mass detection. These researchers have developed a scientific application, which accurately analyzes raw CDMS data generated from their mass spectrometer. This paper explains the comprehensive process of optimizing the group's workflow by improving both the latency and throughput of their CDMS application. These significant performance improvements enabled high efficiency and scalability across IU's Advanced Cyberinfrastructure; overall, this analysis and development resulted in a 25x speedup of the application. © 2016 ACM.

@techreport{
 title = {Jetstream–A Self-Provisioned, Scalable Science and Engineering Cloud Environment-NSF Acceptance Report},
 type = {techreport},
 year = {2016},
 id = {156e9afa-514a-36cf-983f-b67244a26bc5},
 created = {2019-10-01T17:20:38.779Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:47.228Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016h},
 source_type = {RPRT},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Foster, Ian and Merchant, Nirav C and Taylor, James and Vaughn, Matthew W}
}

@techreport{
 title = {Updated Acceptance Test Results for the Jetstream Production Environment},
 type = {techreport},
 year = {2016},
 id = {8c696fb2-c1a3-3ac1-809e-522fe479d134},
 created = {2019-10-01T17:20:38.885Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:45.462Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hancock2016b},
 source_type = {RPRT},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Hancock, David Y and Packard, Michael and Turner, George and Stewart, Craig A}
}

@article{
 title = {Comparing the consumption of CPU hours with scientific output for the extreme science and engineering discovery environment (XSEDE)},
 type = {article},
 year = {2016},
 keywords = {Authorship,Bibliometrics,Co,Computer Terminals,astronomy,atmosphere,biology,chemistry,consume},
 volume = {11},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84976348924&doi=10.1371%2Fjournal.pone.0157628&partnerID=40&md5=f0a4ef8d94ed9e9c752ef2a2a46801ca},
 publisher = {Public Library of Science},
 id = {716aa8a2-b036-3f6e-a7ea-18bb75177f0c},
 created = {2019-10-01T17:20:39.277Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:53.669Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper2016},
 source_type = {article},
 notes = {cited By 0},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {This paper presents the results of a study that compares resource usage with publication output using data about the consumption of CPU cycles from the Extreme Science and Engineering Discovery Environment (XSEDE) and resulting scientific publications for 2,691 institutions/teams. Specifically, the datasets comprise a total of 5,374,032,696 central processing unit (CPU) hours run in XSEDE during July 1, 2011 to August 18, 2015 and 2,882 publications that cite the XSEDE resource. Three types of studies were conducted: a geospatial analysis of XSEDE providers and consumers, co-authorship network analysis of XSEDE publications, and bi-modal network analysis of how XSEDE resources are used by different research fields. Resulting visualizations show that a diverse set of consumers make use of XSEDE resources, that users of XSEDE publish together frequently, and that the users of XSEDE with the highest resource usage tend to be "traditional" high-performance computing (HPC) community members from astronomy, atmospheric science, physics, chemistry, and biology. © 2016 Knepper, Börner.},
 bibtype = {article},
 author = {Knepper, R and Börner, K},
 doi = {10.1371/journal.pone.0157628},
 journal = {PLoS ONE},
 number = {6}
}

This paper presents the results of a study that compares resource usage with publication output using data about the consumption of CPU cycles from the Extreme Science and Engineering Discovery Environment (XSEDE) and resulting scientific publications for 2,691 institutions/teams. Specifically, the datasets comprise a total of 5,374,032,696 central processing unit (CPU) hours run in XSEDE during July 1, 2011 to August 18, 2015 and 2,882 publications that cite the XSEDE resource. Three types of studies were conducted: a geospatial analysis of XSEDE providers and consumers, co-authorship network analysis of XSEDE publications, and bi-modal network analysis of how XSEDE resources are used by different research fields. Resulting visualizations show that a diverse set of consumers make use of XSEDE resources, that users of XSEDE publish together frequently, and that the users of XSEDE with the highest resource usage tend to be "traditional" high-performance computing (HPC) community members from astronomy, atmospheric science, physics, chemistry, and biology. © 2016 Knepper, Börner.

@inproceedings{
 title = {Bongo: A BGP speaker built for defending against bad routes},
 type = {inproceedings},
 year = {2016},
 id = {e6a2effa-6c83-39a7-b2e2-ed64c9f1c8d3},
 created = {2019-10-01T17:20:44.287Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:12.174Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Benton2016a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. Hijacks, outages, route leaks, and AS path spoofing are cases where network operators may want to influence the way routes are accepted and propagated from BGP neighbors in ways not supported by traditional BGP speakers. In this paper, we introduce Bongo, a software-based BGP speaker than can selectively filter out or extend the path of BGP updates received from other peers based on arbitrary operator-defined policies. Additionally, we show how the modularity of this system makes it easy to integrate with existing routers as well as other network devices such as OpenFlow switches or firewalls.},
 bibtype = {inproceedings},
 author = {Benton, K. and Camp, L.J. and Swany, M.},
 doi = {10.1109/MILCOM.2016.7795416},
 booktitle = {Proceedings - IEEE Military Communications Conference MILCOM}
}

© 2016 IEEE. Hijacks, outages, route leaks, and AS path spoofing are cases where network operators may want to influence the way routes are accepted and propagated from BGP neighbors in ways not supported by traditional BGP speakers. In this paper, we introduce Bongo, a software-based BGP speaker than can selectively filter out or extend the path of BGP updates received from other peers based on arbitrary operator-defined policies. Additionally, we show how the modularity of this system makes it easy to integrate with existing routers as well as other network devices such as OpenFlow switches or firewalls.

@techreport{
 title = {Provenance as Essential Infrastructure for Data Lakes [Preprint, forthcoming in IPAW 2016]},
 type = {techreport},
 year = {2016},
 id = {2927b89c-5ba8-311e-9d5d-11b55f38571b},
 created = {2019-10-01T17:20:44.346Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:12.777Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Suriarachchi2016a},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {techreport},
 author = {Suriarachchi, Isuru and Plale, Beth}
}

@inproceedings{
 title = {Network-managed virtual global address space for message-driven runtimes},
 type = {inproceedings},
 year = {2016},
 id = {eaa07e41-70a8-3b3c-b68b-ee4b7b7e6c9c},
 created = {2019-10-01T17:20:44.581Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:10.673Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kulkarni2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {Copyright © 2016 by the Association for Computing Machinery, Inc. (ACM). Maintaining a scalable high-performance virtual global address space using distributed memory hardware has proven to be challenging. In this paper we evaluate a new approach for such an active global address space that leverages the capabilities of the network fabric to manage addressing, rather than software at the endpoint hosts. We describe our overall approach, design alternatives, and present initial experimental results that demonstrate the effectiveness and limitations of existing network hardware.},
 bibtype = {inproceedings},
 author = {Kulkarni, A. and Dalessandro, L. and Kissel, E. and Lumsdaine, A. and Sterling, T. and Swany, M.},
 doi = {10.1145/2907294.2907320},
 booktitle = {HPDC 2016 - Proceedings of the 25th ACM International Symposium on High-Performance Parallel and Distributed Computing}
}

Copyright © 2016 by the Association for Computing Machinery, Inc. (ACM). Maintaining a scalable high-performance virtual global address space using distributed memory hardware has proven to be challenging. In this paper we evaluate a new approach for such an active global address space that leverages the capabilities of the network fabric to manage addressing, rather than software at the endpoint hosts. We describe our overall approach, design alternatives, and present initial experimental results that demonstrate the effectiveness and limitations of existing network hardware.

@article{
 title = {Filtering Source-Spoofed IP Traffic Using Feasible Path Reverse Path Forwarding with SDN},
 type = {article},
 year = {2016},
 pages = {441},
 volume = {5},
 publisher = {IACSIT Press},
 id = {e4c49beb-de42-3bd4-95b6-c5a3d15f27f3},
 created = {2019-10-01T17:20:44.667Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:10.369Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Benton2016},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {article},
 author = {Benton, Kevin and Camp, L Jean and Kelley, Tim and Swany, Martin},
 journal = {International Journal of Computer and Communication Engineering},
 number = {6}
}

@book{
 title = {Provenance as essential infrastructure for Data Lakes},
 type = {book},
 year = {2016},
 source = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
 volume = {9672},
 id = {162a97d5-0181-329d-aadd-3b3b7a78b05b},
 created = {2019-10-01T17:20:44.907Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:19.567Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Suriarachchi2016b},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© Springer International Publishing Switzerland 2016. The Data Lake is emerging as a Big Data storage and management solution which can store any type of data at scale and execute data transformations for analysis. Higher flexibility in storage increases the risk of Data Lakes becoming data swamps. In this paper we show how provenance contributes to data management within a Data Lake infrastructure. We study provenance integration challenges and propose a reference architecture for provenance usage in a Data Lake. Finally we discuss the applicability of our tools in the proposed architecture.},
 bibtype = {book},
 author = {Suriarachchi, I. and Plale, B.},
 doi = {10.1007/978-3-319-40593-3_16}
}

© Springer International Publishing Switzerland 2016. The Data Lake is emerging as a Big Data storage and management solution which can store any type of data at scale and execute data transformations for analysis. Higher flexibility in storage increases the risk of Data Lakes becoming data swamps. In this paper we show how provenance contributes to data management within a Data Lake infrastructure. We study provenance integration challenges and propose a reference architecture for provenance usage in a Data Lake. Finally we discuss the applicability of our tools in the proposed architecture.

@article{
 title = {Argus: A Multi-tenancy NoSQL store with workload-aware resource reservation},
 type = {article},
 year = {2016},
 volume = {58},
 id = {e81d3f9a-ce87-324a-bb1d-6e6062c7fed0},
 created = {2019-10-01T17:20:45.192Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:16.539Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zeng2016a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 Elsevier B.V. Multi-tenancy in cloud hosted NoSQL data stores is favored by cloud providers as it allows more effective resource sharing amongst different tenants thus lowering operating costs. A NoSQL provider will often present to each tenant a dedicated view of the store but then behind the scenes consolidate tenant access into a shared instance. This multi-tenancy approach with tenant data and workloads coexisting in the same infrastructure, under certain conditions can lead to performance degradation of one tenant caused by another as we show experimentally. This paper introduces Argus, a NoSQL store equipped with resource reservation to prevent performance interference across tenants in a multi-tenancy environment. Cache reservation is enforced through partitioning the cache space and disk reservation enforced through scheduling requests to a Distributed File System (DFS). We model the reservation on various workloads as a constrained optimization problem and use the stochastic hill climbing algorithm to find a near-optimum plan for different resource reservations. Empirical results show that Argus is able to prevent interference, adapt to dynamic workloads, and outperform A-Cache, another interference preventing NoSQL solution.},
 bibtype = {article},
 author = {Zeng, J. and Plale, B.},
 doi = {10.1016/j.parco.2016.06.003},
 journal = {Parallel Computing}
}

© 2016 Elsevier B.V. Multi-tenancy in cloud hosted NoSQL data stores is favored by cloud providers as it allows more effective resource sharing amongst different tenants thus lowering operating costs. A NoSQL provider will often present to each tenant a dedicated view of the store but then behind the scenes consolidate tenant access into a shared instance. This multi-tenancy approach with tenant data and workloads coexisting in the same infrastructure, under certain conditions can lead to performance degradation of one tenant caused by another as we show experimentally. This paper introduces Argus, a NoSQL store equipped with resource reservation to prevent performance interference across tenants in a multi-tenancy environment. Cache reservation is enforced through partitioning the cache space and disk reservation enforced through scheduling requests to a Distributed File System (DFS). We model the reservation on various workloads as a constrained optimization problem and use the stochastic hill climbing algorithm to find a near-optimum plan for different resource reservations. Empirical results show that Argus is able to prevent interference, adapt to dynamic workloads, and outperform A-Cache, another interference preventing NoSQL solution.

@inproceedings{
 title = {Horme: Random access big data analytics},
 type = {inproceedings},
 year = {2016},
 id = {cbc012f6-640c-3301-9b84-d22c6e833a97},
 created = {2019-10-01T17:20:47.579Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:40.234Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ruan2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. MapReduce is a parallel framework which has been widely adopted for conducting large-scale data analytics. In cases where analysis of multiple millions of books must be analyzed using federally funded high performance computing (HPC) resources, the framework fails to port directly. We propose a solution that builds off of MapReduce for use on a HPC system that preserves the key-value semantics of map-reduce while supporting the random access of query access for subsetting Big Data datasets, and at same time hosting the service using the storage medium found in HPC architectures (parallel file systems) for reduced latencies. Experimental results demonstrate Horme's good performance in the HPC setting, with up to 41.4% faster than NoSQL based solution in random access scenario.},
 bibtype = {inproceedings},
 author = {Ruan, G. and Plale, B.},
 doi = {10.1109/CLUSTER.2016.27},
 booktitle = {Proceedings - IEEE International Conference on Cluster Computing, ICCC}
}

© 2016 IEEE. MapReduce is a parallel framework which has been widely adopted for conducting large-scale data analytics. In cases where analysis of multiple millions of books must be analyzed using federally funded high performance computing (HPC) resources, the framework fails to port directly. We propose a solution that builds off of MapReduce for use on a HPC system that preserves the key-value semantics of map-reduce while supporting the random access of query access for subsetting Big Data datasets, and at same time hosting the service using the storage medium found in HPC architectures (parallel file systems) for reduced latencies. Experimental results demonstrate Horme's good performance in the HPC setting, with up to 41.4% faster than NoSQL based solution in random access scenario.

@inproceedings{
 title = {Photon: Remote memory access middleware for high-performance runtime systems},
 type = {inproceedings},
 year = {2016},
 id = {83612537-d794-3b0b-909e-64454b9336db},
 created = {2019-10-01T17:20:47.768Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:36.557Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kissel2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. We introduce the Photon RDMA middleware librarythat enables consistent remote memory access semantics overa number of network interconnect technologies. A primarygoal of Photon is to expose a lightweight and flexible network abstraction that minimizes communication and message handlingoverheads for high-performance applications and runtime systems, in particular those that require the manipulation of objectswithin a global address space. Both one-sided and rendezvouscommunication models are supported and asynchronous networkprogress is exposed at a fine granularity. Photon implements anovel communication pattern called put-with-completion (PWC) that optimizes a completion notification path with variable sizedata for realizing active message-driven computation. The resultsof our performance evaluation show that our PWC model iscomparable, and often improves upon, existing one-sided RDMAlibraries in message latency and throughput metrics.},
 bibtype = {inproceedings},
 author = {Kissel, E. and Swany, M.},
 doi = {10.1109/IPDPSW.2016.120},
 booktitle = {Proceedings - 2016 IEEE 30th International Parallel and Distributed Processing Symposium, IPDPS 2016}
}

© 2016 IEEE. We introduce the Photon RDMA middleware librarythat enables consistent remote memory access semantics overa number of network interconnect technologies. A primarygoal of Photon is to expose a lightweight and flexible network abstraction that minimizes communication and message handlingoverheads for high-performance applications and runtime systems, in particular those that require the manipulation of objectswithin a global address space. Both one-sided and rendezvouscommunication models are supported and asynchronous networkprogress is exposed at a fine granularity. Photon implements anovel communication pattern called put-with-completion (PWC) that optimizes a completion notification path with variable sizedata for realizing active message-driven computation. The resultsof our performance evaluation show that our PWC model iscomparable, and often improves upon, existing one-sided RDMAlibraries in message latency and throughput metrics.

@inproceedings{
 title = {KVLight: A Lightweight Key-Value Store for Distributed Access in Cloud},
 type = {inproceedings},
 year = {2016},
 id = {f0873ec4-5214-3910-b864-9e33ecc349a7},
 created = {2019-10-01T17:20:47.809Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:36.854Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zeng2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. Key-value stores (KVS) are finding use in Big Data applications as the store offers a flexible data model, scalability in number of distributed nodes, and high availability. In a cloud environment, a distributed KVS is often deployed over the local file system of the nodes in a cluster of virtual machines (VMs). Parallel file system (PFS) offers an alternate approach to disk storage, however a distributed key value store running over a parallel file system can experience overheads due to its unawareness of the PFS. Additionally, distributed KVS requires persistent running services which is not cost effective under the pay-as-you-go model of cloud computing because resources have to be held even under periods of no workload. We propose KVLight, a lightweight KVS that runs over PFS. It is lightweight in the sense that it shifts the responsibility of reliable data storage to the PFS and focuses on performance. Specifically, KVLight is built on an embedded KVS for high performance but uses novel data structures to support concurrent writes, giving capability that embedded KVSs are not currently designed for. Furthermore, it allows on-demand access without running persistent services in front of the file system. Empirical results show that KVLight outperforms Cassandra and Voldemort, two state-of-the-art KVSs, under both synthetic and realistic workloads.},
 bibtype = {inproceedings},
 author = {Zeng, J. and Plale, B.},
 doi = {10.1109/CCGrid.2016.55},
 booktitle = {Proceedings - 2016 16th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing, CCGrid 2016}
}

© 2016 IEEE. Key-value stores (KVS) are finding use in Big Data applications as the store offers a flexible data model, scalability in number of distributed nodes, and high availability. In a cloud environment, a distributed KVS is often deployed over the local file system of the nodes in a cluster of virtual machines (VMs). Parallel file system (PFS) offers an alternate approach to disk storage, however a distributed key value store running over a parallel file system can experience overheads due to its unawareness of the PFS. Additionally, distributed KVS requires persistent running services which is not cost effective under the pay-as-you-go model of cloud computing because resources have to be held even under periods of no workload. We propose KVLight, a lightweight KVS that runs over PFS. It is lightweight in the sense that it shifts the responsibility of reliable data storage to the PFS and focuses on performance. Specifically, KVLight is built on an embedded KVS for high performance but uses novel data structures to support concurrent writes, giving capability that embedded KVSs are not currently designed for. Furthermore, it allows on-demand access without running persistent services in front of the file system. Empirical results show that KVLight outperforms Cassandra and Voldemort, two state-of-the-art KVSs, under both synthetic and realistic workloads.

@inproceedings{
 title = {SamzaSQL: Scalable fast data management with streaming SQL},
 type = {inproceedings},
 year = {2016},
 id = {a8c6dea7-b780-3df8-99bd-f0801c418154},
 created = {2019-10-01T17:20:48.312Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:21.561Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pathirage2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. As the data-driven economy evolves, enterprises have come to realize a competitive advantage in being able to act on high volume, high velocity streams of data. Technologies such as distributed message queues and streaming processing platforms that can scale to thousands of data stream partitions on commodity hardware are a response. However, the programming API provided by these systems is often low-level, requiring substantial custom code that adds to the programmer learning curve and maintenance overhead. Additionally, these systems often lack SQL querying capabilities that have proven popular on Big Data systems like Hive, Impala or Presto. We define a minimal set of extensions to standard SQL for data stream querying and manipulation. These extensions are prototyped in SamzaSQL, a new tool for streaming SQL that compiles streaming SQL into physical plans that are executed on Samza, an open-source distributed stream processing framework. We compare the performance of streaming SQL queries against native Samza applications and discuss usability improvements. SamzaSQL is a part of the open source Apache Samza project and will be available for general use.},
 bibtype = {inproceedings},
 author = {Pathirage, M. and Hyde, J. and Pan, Y. and Plale, B.},
 doi = {10.1109/IPDPSW.2016.141},
 booktitle = {Proceedings - 2016 IEEE 30th International Parallel and Distributed Processing Symposium, IPDPS 2016}
}

© 2016 IEEE. As the data-driven economy evolves, enterprises have come to realize a competitive advantage in being able to act on high volume, high velocity streams of data. Technologies such as distributed message queues and streaming processing platforms that can scale to thousands of data stream partitions on commodity hardware are a response. However, the programming API provided by these systems is often low-level, requiring substantial custom code that adds to the programmer learning curve and maintenance overhead. Additionally, these systems often lack SQL querying capabilities that have proven popular on Big Data systems like Hive, Impala or Presto. We define a minimal set of extensions to standard SQL for data stream querying and manipulation. These extensions are prototyped in SamzaSQL, a new tool for streaming SQL that compiles streaming SQL into physical plans that are executed on Samza, an open-source distributed stream processing framework. We compare the performance of streaming SQL queries against native Samza applications and discuss usability improvements. SamzaSQL is a part of the open source Apache Samza project and will be available for general use.

@inproceedings{
 title = {User managed virtual clusters in comet},
 type = {inproceedings},
 year = {2016},
 volume = {17-21-July},
 id = {5d5dcc84-8742-3a12-9344-79d4ccb1b80f},
 created = {2019-10-01T17:20:56.718Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:38.888Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wagner2016},
 folder_uuids = {82975498-107c-4bb3-bb76-f87bce3e9f6e},
 private_publication = {false},
 abstract = {© 2016 ACM. Hardware virtualization has been gaining a significant share of computing time in the last years. Using virtual machines (VMs) for parallel computing is an attractive option for many users. A VM gives users a freedom of choosing an operating system, software stack and security policies, leaving the physical hardware, OS management, and billing to physical cluster administrators. The well-known solutions for cloud computing, both commercial (Amazon Cloud, Google Cloud, Yahoo Cloud, etc.) and open-source (OpenStack, Eucalyptus) provide platforms for running a single VM or a group of VMs. With all the benefits, there are also some drawbacks, which include reduced performance when running code inside of a VM, increased complexity of cluster management, as well as the need to learn new tools and protocols to manage the clusters. At SDSC, we have created a novel framework and infrastructure by providing virtual HPC clusters to projects using the NSF sponsored Comet supercomputer. Managing virtual clusters on Comet is similar to managing a baremetal cluster in terms of processes and tools that are employed. This is beneficial because such processes and tools are familiar to cluster administrators. Unlike platforms like AWS, Comet's virtualization capability supports installing VMs from ISOs (i.e., a CD-ROM or DVD image) or via an isolated management VLAN (PXE). At the same time, we're helping projects take advantage of VMs by providing an enhanced client tool for interaction with our management system called Cloudmesh client. Cloudmesh client can also be used to manage virtual machines on OpenStack, AWS, and Azure. The article describes our design and approach to running virtual clusters, the tools we developed, and initial user experience.},
 bibtype = {inproceedings},
 author = {Wagner, R. and Papadopoulos, P. and Mishin, D. and Cooper, T. and Tatineti, M. and Von Laszewski, G. and Wang, F. and Fox, Geoffrey Charles G.C.},
 doi = {10.1145/2949550.2949555},
 booktitle = {ACM International Conference Proceeding Series}
}

© 2016 ACM. Hardware virtualization has been gaining a significant share of computing time in the last years. Using virtual machines (VMs) for parallel computing is an attractive option for many users. A VM gives users a freedom of choosing an operating system, software stack and security policies, leaving the physical hardware, OS management, and billing to physical cluster administrators. The well-known solutions for cloud computing, both commercial (Amazon Cloud, Google Cloud, Yahoo Cloud, etc.) and open-source (OpenStack, Eucalyptus) provide platforms for running a single VM or a group of VMs. With all the benefits, there are also some drawbacks, which include reduced performance when running code inside of a VM, increased complexity of cluster management, as well as the need to learn new tools and protocols to manage the clusters. At SDSC, we have created a novel framework and infrastructure by providing virtual HPC clusters to projects using the NSF sponsored Comet supercomputer. Managing virtual clusters on Comet is similar to managing a baremetal cluster in terms of processes and tools that are employed. This is beneficial because such processes and tools are familiar to cluster administrators. Unlike platforms like AWS, Comet's virtualization capability supports installing VMs from ISOs (i.e., a CD-ROM or DVD image) or via an isolated management VLAN (PXE). At the same time, we're helping projects take advantage of VMs by providing an enhanced client tool for interaction with our management system called Cloudmesh client. Cloudmesh client can also be used to manage virtual machines on OpenStack, AWS, and Azure. The article describes our design and approach to running virtual clusters, the tools we developed, and initial user experience.

@inproceedings{
 title = {Providing Statistical Reliability Guarantee for Cloud Clusters},
 type = {inproceedings},
 year = {2016},
 city = {Washington, DC.},
 id = {3954626d-7fb3-3889-af54-95a40ad38f8a},
 created = {2019-10-01T17:20:57.859Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:27.602Z},
 read = {false},
 starred = {true},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Yang2016},
 source_type = {JOUR},
 folder_uuids = {82975498-107c-4bb3-bb76-f87bce3e9f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Yang, Zhouhan and Du, Anna Ye and Das, Sanjukta and Ramesh, Ram and Furlani, Thomas and von Laszewski, Gregor and Qiao, Chunming},
 booktitle = {Providing Statistical Reliability Guarantee for Cloud Clusters. Submitted to Global Communications Conference}
}

@inproceedings{
 title = {A multi-tenant fair share approach to full-text search engine},
 type = {inproceedings},
 year = {2016},
 pages = {45-50},
 publisher = {IEEE Press},
 id = {36b6f627-29ee-3334-8f64-1f408389326a},
 created = {2019-10-01T17:20:58.701Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:00.284Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Peng2016},
 source_type = {CONF},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. Full text search engines underly the search of major content providers, Google, Bing and Yahoo. Open source search engines, such as Solr and ElasticSearch, are highly scalable and widely used in a Software-as-a-Service (SaaS) manner, in which multiple tenants share a single resource for improved resource utilization and lower management cost. Sharing of a full text search engine can exhibit unfairness in the form of performance interference. We propose a multi-tenancy solution that provides fair share of resource usage of a SaaS hosted search engine. It includes a revised deficit round robin technique for admission control, query resource usage estimation and a deadlock breaking mechanism. Experimental results show that our approach works well for both monolithic and distributed search engines.},
 bibtype = {inproceedings},
 author = {Peng, Zong and Plale, Beth},
 doi = {10.1109/DataCloud.2016.010},
 booktitle = {Proceedings of the 7th International Workshop on Data-Intensive Computing in the Cloud}
}

© 2016 IEEE. Full text search engines underly the search of major content providers, Google, Bing and Yahoo. Open source search engines, such as Solr and ElasticSearch, are highly scalable and widely used in a Software-as-a-Service (SaaS) manner, in which multiple tenants share a single resource for improved resource utilization and lower management cost. Sharing of a full text search engine can exhibit unfairness in the form of performance interference. We propose a multi-tenancy solution that provides fair share of resource usage of a SaaS hosted search engine. It includes a revised deficit round robin technique for admission control, query resource usage estimation and a deadlock breaking mechanism. Experimental results show that our approach works well for both monolithic and distributed search engines.

@article{
 title = {Big Data at Scale for Digital Humanities: An Architecture for the},
 type = {article},
 year = {2016},
 pages = {345},
 publisher = {IGI Global},
 id = {28f9c0de-5e0c-3e82-a829-10c10e3b6202},
 created = {2019-10-01T17:20:58.870Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:57.489Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kowalczyk2016},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {article},
 author = {Kowalczyk, Stacy T and Sun, Yiming and Peng, Zong and Plale, Beth and Willis, Craig and Zeng, Jiaan and Pathirage, Milinda and Liyanage, Samitha and Todd, Aaron and Ruan, Guangchen},
 journal = {Big Data: Concepts, Methodologies, Tools, and Applications: Concepts, Methodologies, Tools, and Applications}
}

@book{
 title = {Analysis of memory constrained live provenance},
 type = {book},
 year = {2016},
 source = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
 volume = {9672},
 id = {7aabf41b-617c-3c37-90d9-0dac98577034},
 created = {2019-10-01T17:20:59.241Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:58.080Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© Springer International Publishing Switzerland 2016. We conjecture that meaningful analysis of large-scale provenance can be preserved by analyzing provenance data in limited memory while the data is still in motion; that the provenance needs not be fully resident before analysis can occur. As a proof of concept, this paper defines a stream model for reasoning about provenance data in motion for Big Data provenance.We propose a novel streaming algorithm for the backward provenance query, and apply it to the live provenance captured from agent-based simulations. The performance test demonstrates high throughput, low latency and good scalability, in a distributed stream processing framework built on Apache Kafka and Spark Streaming.},
 bibtype = {book},
 author = {Chen, P. and Evans, T. and Plale, B.},
 doi = {10.1007/978-3-319-40593-3_4}
}

© Springer International Publishing Switzerland 2016. We conjecture that meaningful analysis of large-scale provenance can be preserved by analyzing provenance data in limited memory while the data is still in motion; that the provenance needs not be fully resident before analysis can occur. As a proof of concept, this paper defines a stream model for reasoning about provenance data in motion for Big Data provenance.We propose a novel streaming algorithm for the backward provenance query, and apply it to the live provenance captured from agent-based simulations. The performance test demonstrates high throughput, low latency and good scalability, in a distributed stream processing framework built on Apache Kafka and Spark Streaming.

@inproceedings{
 title = {Offloading collective operations to programmable logic on a Zynq cluster},
 type = {inproceedings},
 year = {2016},
 id = {dfd65d42-328f-36a0-81c5-678829388424},
 created = {2019-10-01T17:20:59.245Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:55.591Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Arap2016},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2016 IEEE. This paper describes our architecture and implementation for offloading collective operations to programmable logic in the communication substrate. Collective operations - operations that involve communication between groups of co-operating processes - are widely used in parallel processing. The design and implementation strategies of collective operations plays a significant role in their performance and thus affects the performance of many high performance computing applications that utilize them. Collectives are central to the widely used Message Passing Interface (MPI) programming model. The programmable logic provided by FPGAs is a powerful option for creating task-specific logic to aid applications. While our work is evaluated on the Xilinx Zynq SoC, it is generally applicable in scenarios where there is programmable logic in the communication pipeline, including FPGAs on network interface cards like the NetFPGA or new systems like Intel's Xeon with on-die Altera FPGA resources. In this paper we have adapted and generalized our previous work in offloading collective operations to the NetFPGA. Here we present a general collective offloading framework for use in applications using the Message Passing Interface (MPI). The implementation is realized on the Xilinx Zynq reference platform, the Zedboard, using an Ethernet daughter card called EthernetFMC. Results from microbenchmarks are presented as well as from some scientific applications using MPI.},
 bibtype = {inproceedings},
 author = {Arap, O. and Swany, M.},
 doi = {10.1109/HOTI.2016.024},
 booktitle = {Proceedings - 2016 IEEE 24th Annual Symposium on High-Performance Interconnects, HOTI 2016}
}

© 2016 IEEE. This paper describes our architecture and implementation for offloading collective operations to programmable logic in the communication substrate. Collective operations - operations that involve communication between groups of co-operating processes - are widely used in parallel processing. The design and implementation strategies of collective operations plays a significant role in their performance and thus affects the performance of many high performance computing applications that utilize them. Collectives are central to the widely used Message Passing Interface (MPI) programming model. The programmable logic provided by FPGAs is a powerful option for creating task-specific logic to aid applications. While our work is evaluated on the Xilinx Zynq SoC, it is generally applicable in scenarios where there is programmable logic in the communication pipeline, including FPGAs on network interface cards like the NetFPGA or new systems like Intel's Xeon with on-die Altera FPGA resources. In this paper we have adapted and generalized our previous work in offloading collective operations to the NetFPGA. Here we present a general collective offloading framework for use in applications using the Message Passing Interface (MPI). The implementation is realized on the Xilinx Zynq reference platform, the Zedboard, using an Ethernet daughter card called EthernetFMC. Results from microbenchmarks are presented as well as from some scientific applications using MPI.

@misc{
 title = {Evaluating Collectives in Networks of Multicore/Two-level Reduction},
 type = {misc},
 year = {2016},
 id = {dbedbdc8-3b5e-311e-a576-4a71af78be0d},
 created = {2019-10-01T17:20:59.669Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:03.618Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wickramasinghe2016},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {misc},
 author = {Wickramasinghe, U S and D’Alessandro, Luke and Lumsdaine, Andrew and Kissel, Ezra and Swany, Martin and Newton, Ryan}
}

@inproceedings{
 title = {Rapid Monitoring of Drought Impacts on Small-Scale Farms in Africa through Integration of Farmer SMS data and Environmental Sensors},
 type = {inproceedings},
 year = {2016},
 id = {f28c6e30-3321-3154-b07a-dc745844b010},
 created = {2019-10-01T17:21:00.794Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:51.310Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Evans2016},
 source_type = {CONF},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Evans, T P and Caylor, K K and Estes, L D and Plale, B A and Attari, S and Waldman, K},
 booktitle = {AGU Fall Meeting Abstracts}
}

@article{
 title = {NIST Big Data Interoperability Framework: Volume 3, Use Cases and General Requirements (NIST Special Publication 1500-3)},
 type = {article},
 year = {2016},
 keywords = {Big Data,Big Data Application Provider,Big Data Framework Provider,Big Data characteristics,Big Data taxonomy,Data Consumer,Data Provider,Management Fabric,Security and Privacy Fabric,System Orchestrator,data science,reference architecture,use cases.},
 volume = {3},
 websites = {http://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.1500-3.pdf},
 id = {e4463cb3-f000-30a4-a0a4-1fcab30ab226},
 created = {2019-10-01T17:21:01.399Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:31.534Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Subgroup2016},
 private_publication = {false},
 abstract = {Big Data is a term used to describe the large amount of data in the networked, digitized, sensor-laden, information-driven world. While opportunities exist with Big Data, the data can overwhelm traditional technical approaches and the growth of data is outpacing scientific and technological advances in data analytics. To advance progress in Big Data, the NIST Big Data Public Working Group (NBD-PWG) is working to develop consensus on important fundamental concepts related to Big Data. The results are reported in the NIST Big Data Interoperability Framework series of volumes. This volume, Volume 3, contains the original 51 Version 1 use cases gathered by the NBD-PWG Use Cases and Requirements Subgroup and the requirements generated from those use cases. The use cases are presented in their original and summarized form. Requirements, or challenges, were extracted from each use case, and then summarized over all the use cases. These generalized requirements were used in the development of the NIST Big Data Reference Architecture (NBDRA), which is presented in Volume 6. Currently, the subgroup is accepting additional use case submissions using the more detailed Use Case Template 2. The Use Case Template 2 and the two Version 2 use cases collected to date are presented and summarized in this volume.},
 bibtype = {article},
 author = {Subgroup, NIST Big Data Public Working Group: Use Cases and Requirements},
 doi = {http://dx.doi.org/10.6028/NIST.SP.1500-3},
 number = {June}
}

Big Data is a term used to describe the large amount of data in the networked, digitized, sensor-laden, information-driven world. While opportunities exist with Big Data, the data can overwhelm traditional technical approaches and the growth of data is outpacing scientific and technological advances in data analytics. To advance progress in Big Data, the NIST Big Data Public Working Group (NBD-PWG) is working to develop consensus on important fundamental concepts related to Big Data. The results are reported in the NIST Big Data Interoperability Framework series of volumes. This volume, Volume 3, contains the original 51 Version 1 use cases gathered by the NBD-PWG Use Cases and Requirements Subgroup and the requirements generated from those use cases. The use cases are presented in their original and summarized form. Requirements, or challenges, were extracted from each use case, and then summarized over all the use cases. These generalized requirements were used in the development of the NIST Big Data Reference Architecture (NBDRA), which is presented in Volume 6. Currently, the subgroup is accepting additional use case submissions using the more detailed Use Case Template 2. The Use Case Template 2 and the two Version 2 use cases collected to date are presented and summarized in this volume.

@inproceedings{
 title = {Crossing analytics systems: A case for integrated provenance in data lakes},
 type = {inproceedings},
 year = {2016},
 pages = {349-354},
 websites = {http://ieeexplore.ieee.org/document/7870919/},
 month = {10},
 publisher = {IEEE},
 id = {b11147d2-bd28-3963-8f74-71d4058411e6},
 created = {2019-10-01T17:21:02.856Z},
 accessed = {2018-09-05},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:23.340Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Suriarachchi2017},
 source_type = {CONF},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 abstract = {© 2016 IEEE. The volumes of data in Big Data, their variety and unstructured nature, have had researchers looking beyond the data warehouse. The data warehouse, among other features, requires mapping data to a schema upon ingest, an approach seen as inflexible for the massive variety of Big Data. The Data Lake is emerging as an alternate solution for storing data of widely divergent types and scales. Designed for high flexibility, the Data Lake follows a schema-on-read philosophy and data transformations are assumed to be performed within the Data Lake. During its lifecycle in a Data Lake, a data product may undergo numerous transformations performed by any number of Big Data processing engines leading to questions of traceability. In this paper we argue that provenance contributes to easier data management and traceability within a Data Lake infrastructure. We discuss the challenges in provenance integration in a Data Lake and propose a reference architecture to overcome the challenges. We evaluate our architecture through a prototype implementation built using our distributed provenance collection tools.},
 bibtype = {inproceedings},
 author = {Suriarachchi, Isuru and Plale, Beth A.},
 doi = {10.1109/eScience.2016.7870919},
 booktitle = {Proceedings of the 2016 IEEE 12th International Conference on e-Science, e-Science 2016}
}

© 2016 IEEE. The volumes of data in Big Data, their variety and unstructured nature, have had researchers looking beyond the data warehouse. The data warehouse, among other features, requires mapping data to a schema upon ingest, an approach seen as inflexible for the massive variety of Big Data. The Data Lake is emerging as an alternate solution for storing data of widely divergent types and scales. Designed for high flexibility, the Data Lake follows a schema-on-read philosophy and data transformations are assumed to be performed within the Data Lake. During its lifecycle in a Data Lake, a data product may undergo numerous transformations performed by any number of Big Data processing engines leading to questions of traceability. In this paper we argue that provenance contributes to easier data management and traceability within a Data Lake infrastructure. We discuss the challenges in provenance integration in a Data Lake and propose a reference architecture to overcome the challenges. We evaluate our architecture through a prototype implementation built using our distributed provenance collection tools.

@inproceedings{
 title = {Anatomy of the SEAGrid Science Gateway},
 type = {inproceedings},
 year = {2016},
 pages = {40},
 websites = {http://dl.acm.org/citation.cfm?doid=2949550.2949591},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 institution = {ACM},
 id = {100d4f29-2f83-3fdb-bedc-a6cb01a4b563},
 created = {2019-10-01T17:21:06.461Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:28:38.317Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {nakandala2016anatomy},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 abstract = {The SEAGrid science gateway provides scientists and educators with user interfaces and tools for conducting computational chemistry, material science, and engineering experiments online using XSEDE and campus computing resources. This paper describes the architecture of the recently completed technology refresh for the gateway, replacing its desktop user interface, adding a web browser user interface, using Apache Airavata middleware for job management, and providing enhanced data search and feature extraction capabilities. These introduce several challenges, particularly in providing unified authentication and authorization mechanisms to middleware services for the desktop and web clients, and in extending Apache Airavata middleware with new components. Access, authentication, and authorization problems were solved by using standard-based approaches (OAuth2, XACML) that were implemented by incorporating WSO2's Identity Server into both SEAGrid and Apache Airavata. SEAGrid-specific data extraction capabilities were added to Airavata middleware using a message-based component approach. This approach is generalizable to other advanced and gateway-specific capabilities and enables Airavata to add additional data analysis components without modifying its core functionality.},
 bibtype = {inproceedings},
 author = {Nakandala, Supun and Pamidighantam, Sudhakar and Yodage, Shameera and Doshi, Nipurn and Abeysinghe, Eroma and Kankanamalage, Chathuri Peli and Marru, Suresh and Pierce, Marlon},
 doi = {10.1145/2949550.2949591},
 booktitle = {Proceedings of the XSEDE16 on Diversity, Big Data, and Science at Scale - XSEDE16}
}

The SEAGrid science gateway provides scientists and educators with user interfaces and tools for conducting computational chemistry, material science, and engineering experiments online using XSEDE and campus computing resources. This paper describes the architecture of the recently completed technology refresh for the gateway, replacing its desktop user interface, adding a web browser user interface, using Apache Airavata middleware for job management, and providing enhanced data search and feature extraction capabilities. These introduce several challenges, particularly in providing unified authentication and authorization mechanisms to middleware services for the desktop and web clients, and in extending Apache Airavata middleware with new components. Access, authentication, and authorization problems were solved by using standard-based approaches (OAuth2, XACML) that were implemented by incorporating WSO2's Identity Server into both SEAGrid and Apache Airavata. SEAGrid-specific data extraction capabilities were added to Airavata middleware using a message-based component approach. This approach is generalizable to other advanced and gateway-specific capabilities and enables Airavata to add additional data analysis components without modifying its core functionality.

@article{
 title = {Integrating Apache Airavata with Docker, Marathon, and Mesos},
 type = {article},
 year = {2016},
 pages = {1952-1959},
 volume = {28},
 id = {7ab3072b-e64b-3455-b1ef-b24494c096c4},
 created = {2019-10-01T17:21:07.542Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:37.930Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {saha2016integrating},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {article},
 author = {Saha, Pankaj and Govindaraju, Madhusudhan and Marru, Suresh and Pierce, Marlon},
 journal = {Concurrency and Computation: Practice and Experience},
 number = {7}
}

@inproceedings{
 title = {GeoGateway: A system for analysis of UAVSAR data products},
 type = {inproceedings},
 year = {2016},
 pages = {210-213},
 institution = {IEEE},
 id = {afc10032-c92c-33e4-a122-665ce510c2ac},
 created = {2019-10-01T17:21:07.627Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:35.845Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {donnellan2016geogateway},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Donnellan, Andrea and Parker, Jay and Glasscoe, Margaret and Granat, Robert and Pierce, Marlon and Wang, Jun and Ma, Yu and Ludwig, Lisa Grant and Rundle, John},
 booktitle = {Geoscience and Remote Sensing Symposium (IGARSS), 2016 IEEE International}
}

@inproceedings{
 title = {Apache Airavata security manager: Authentication and authorization implementations for a multi-tenant escience framework},
 type = {inproceedings},
 year = {2016},
 pages = {287-292},
 institution = {IEEE},
 id = {0087833e-b69f-3529-8b7a-0fc55cc37f58},
 created = {2019-10-01T17:21:09.093Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:20.411Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {nakandala2016apache},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Nakandala, Supun and Gunasinghe, Hasini and Marru, Suresh and Pierce, Marlon},
 booktitle = {e-Science (e-Science), 2016 IEEE 12th International Conference on}
}

@article{
 title = {Community Science Exemplars in SEAGrid Science Gateway: Apache Airavata Based Implementation of Advanced Infrastructure},
 type = {article},
 year = {2016},
 pages = {1927-1939},
 volume = {80},
 websites = {https://linkinghub.elsevier.com/retrieve/pii/S1877050916310262},
 publisher = {Elsevier},
 id = {40f7d463-4d21-3c43-b2ca-e76a7c5a7b75},
 created = {2019-10-01T17:21:09.358Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:30.014Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {pamidighantam2016community},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 abstract = {We describe the science discovered by some of the community of researchers using the SEAGrid Science gateway. Specific science projects to be discussed include calcium carbonate and bicarbonate hydrochemistry, mechanistic studies of redox proteins and diffraction modeling of metal and metal-oxide structures and interfaces. The modeling studies involve a variety of ab initio and molecular dynamics computational techniques and coupled execution of a workflows using specific set of applications enabled in the SEAGrid Science Gateway. The integration of applications and resources that enable workflows that couple empirical, semi-empirical, ab initio DFT, and Moller-Plesset perturbative models and combine computational and visualization modules through a single point of access is now possible through the SEAGrid gateway. Integration with the Apache Airavata infrastructure to gain a sustainable and more easily maintainable set of services is described. As part of this integration we also provide a web browser based SEAGrid Portal in addition to the SEAGrid rich client based on the previous GridChem client. We will elaborate the services and their enhancements in this process to exemplify how the new implementation will enhance the maintainability and sustainability. We will also provide exemplar science workflows and contrast how they are supported in the new deployment to showcase the adoptability and user support for services and resources.},
 bibtype = {article},
 author = {Pamidighantam, Sudhakar and Nakandala, Supun and Abeysinghe, Eroma and Wimalasena, Chathuri and Yodage, Shameera Rathnayaka and Marru, Suresh and Pierce, Marlon},
 doi = {10.1016/j.procs.2016.05.535},
 journal = {Procedia Computer Science},
 number = {C}
}

We describe the science discovered by some of the community of researchers using the SEAGrid Science gateway. Specific science projects to be discussed include calcium carbonate and bicarbonate hydrochemistry, mechanistic studies of redox proteins and diffraction modeling of metal and metal-oxide structures and interfaces. The modeling studies involve a variety of ab initio and molecular dynamics computational techniques and coupled execution of a workflows using specific set of applications enabled in the SEAGrid Science Gateway. The integration of applications and resources that enable workflows that couple empirical, semi-empirical, ab initio DFT, and Moller-Plesset perturbative models and combine computational and visualization modules through a single point of access is now possible through the SEAGrid gateway. Integration with the Apache Airavata infrastructure to gain a sustainable and more easily maintainable set of services is described. As part of this integration we also provide a web browser based SEAGrid Portal in addition to the SEAGrid rich client based on the previous GridChem client. We will elaborate the services and their enhancements in this process to exemplify how the new implementation will enhance the maintainability and sustainability. We will also provide exemplar science workflows and contrast how they are supported in the new deployment to showcase the adoptability and user support for services and resources.

@article{
 title = {GSoC 2015 student contributions to GenApp and Airavata},
 type = {article},
 year = {2016},
 pages = {1960-1970},
 volume = {28},
 id = {9e0cb013-31d6-3b9c-af22-6454e68f405e},
 created = {2019-10-01T17:21:09.738Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:25.822Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {brookes2016gsoc},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,44d48ba6-09b0-4f17-965c-b625aaaba978},
 private_publication = {false},
 bibtype = {article},
 author = {Brookes, Emre H and Kapoor, Abhishek and Patra, Priyanshu and Marru, Suresh and Singh, Raminder and Pierce, Marlon},
 journal = {Concurrency and Computation: Practice and Experience},
 number = {7}
}

@techreport{
 title = {The Report of the 2016 Cybersecurity Summit for Large Facilities and Cyberinfrastructure: Strengthening Trustworthy Science},
 type = {techreport},
 year = {2016},
 id = {5b8da0ad-0f0e-3904-9d8b-01799ce61358},
 created = {2019-10-01T17:21:12.655Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:39.751Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Marsteller2016},
 source_type = {RPRT},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {techreport},
 author = {Marsteller, James and Welch, Von and Starzynski Coddens, Amy}
}

@article{
 title = {A Time of Turmoil},
 type = {article},
 year = {2016},
 pages = {1-3},
 volume = {6},
 id = {9cc51cec-32f0-3526-be53-b91015f280f1},
 created = {2019-10-01T17:21:12.817Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:39.029Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2016a},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Svantesson, Dan Jerker B and Lynsky, Orla and Millard, Christopher},
 doi = {https://doi.org/10.1093/idpl/ipw001},
 journal = {International Data Privacy Law},
 number = {1}
}

@article{
 title = {The Language of Data Privacy Law (and How It Differs from Reality)},
 type = {article},
 year = {2016},
 pages = {259-269},
 volume = {4},
 id = {baac43bf-5412-321d-8f10-72e7f7a47c2d},
 created = {2019-10-01T17:21:13.078Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:52.547Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2016},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla and Millard, Christopher},
 journal = {International Data Privacy Law},
 number = {1}
}

@inproceedings{
 title = {Modeling the Past and Future of Identity Management for Scientific Collaborations},
 type = {inproceedings},
 year = {2016},
 id = {b438e611-7529-3c43-9374-82b2b4a95bff},
 created = {2019-10-01T17:21:13.350Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:48.854Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cowles2016},
 source_type = {CONF},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Cowles, Robert and Jackson, Craig and Welch, Von},
 booktitle = {Proceedings of the International Symposium on Grids and Clouds (ISGC) 2016. 13-18 March 2016. Academia Sinica, Taipei, Taiwan. Online at http://pos. sissa. it/cgi-bin/reader/conf. cgi? confid= 270, id. 14}
}

@article{
 title = {The Global Data Protection Implications of" Brexit"},
 type = {article},
 year = {2016},
 pages = {167-169},
 volume = {6},
 id = {e0c48fdf-fbfe-3693-a1d1-66eb1ab71127},
 created = {2019-10-01T17:21:13.594Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:47.407Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2016b},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla and Millard, Christopher},
 journal = {International Data Privacy Law},
 number = {3}
}

@techreport{
 title = {PKI-ASAF Design Documents},
 type = {techreport},
 year = {2016},
 id = {9eb716ac-677c-3a28-907a-113d522d5ac8},
 created = {2019-10-01T17:21:14.395Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:35.595Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Welch2016},
 source_type = {RPRT},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {techreport},
 author = {Welch, Von and Heiland, Randy and Garrison III, William C and Lee, Adam J}
}

@techreport{
 title = {Center for Trustworthy Scientific Cyberinfrastructure-The NSF Cybersecurity Center of Excellence: Year One Report},
 type = {techreport},
 year = {2016},
 publisher = {CTSC},
 id = {cb30fa5e-2d5a-3b14-9d05-287a314c636b},
 created = {2019-10-01T17:21:14.570Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:34.168Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Welch2016a},
 source_type = {RPRT},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {techreport},
 author = {Welch, Von}
}

@article{
 title = {The Medical Science DMZ},
 type = {article},
 year = {2016},
 pages = {1199-1201},
 volume = {23},
 publisher = {Oxford University Press},
 id = {e9fda94e-300e-32e3-884a-79383a0fddbb},
 created = {2019-10-01T17:21:16.371Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:00.673Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Peisert2016},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Peisert, Sean and Barnett, William and Dart, Eli and Cuff, James and Grossman, Robert L and Balas, Edward and Berman, Ari and Shankar, Anurag and Tierney, Brian},
 journal = {Journal of the American Medical Informatics Association},
 number = {6}
}

@techreport{
 title = {SciGaP-CTSC Engagement Summary},
 type = {techreport},
 year = {2016},
 id = {99770822-dee5-3cfc-ab77-9dd052153f1d},
 created = {2019-10-01T17:21:17.496Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:04.628Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Heiland2016},
 source_type = {RPRT},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {techreport},
 author = {Heiland, Randy and Koranda, Scott and Welch, Von}
}

@inproceedings{
 title = {Implementation of Simple XSEDE-Like Clusters: Science Enabled and Lessons Learned},
 type = {inproceedings},
 year = {2016},
 pages = {1-8},
 volume = {17-21-July},
 websites = {http://dl.acm.org/citation.cfm?doid=2949550.2949570},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {e653125a-91be-3062-be0a-ab848571eb87},
 created = {2019-10-01T17:21:28.003Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:33:19.839Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Coulter2016},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {The Extreme Science and Engineering Discovery Environ-ment (XSEDE) has created a suite of software that is col-lectively known as the XSEDE-Compatible Basic Cluster (XCBC). It is designed to enable smaller, resource-constrained research groups or universities to quickly and easily imple-ment a computing environment similar to XSEDE comput-ing resources. The XCBC system consists of the Rocks Cluster Manager, developed at the San Diego Supercom-puter Center for use on Gordon and Comet, and an XSEDE-specific "Rocks Roll", containing a selection of libraries, compilers, and scientific software curated by the Campus Bridg-ing (CB) group in the XSEDE project, kept current with those implemented on XSEDE resources. The Campus Bridging team has helped several universities implement the XCBC, and finds the design to be extremely useful for resourcelimited (in time, administrator knowledge, or funding) research groups or institutions. Here, we detail our recent experiences in implementing the XCBC design at university campuses across the country. These XCBC implementations were carried out with Campus Bridging staff traveling on-site to the partner institutions to directly assist with the cluster build. In implementing XCBC on campuses, we found that number of the needs described by campus communities as well as the broader cyberinfrastructure community are solved by technical means, although financial issues remain. The remaining issue to be addressed is technical interoperation between systems, and we describe efforts to improve here.},
 bibtype = {inproceedings},
 author = {Coulter, Eric and Fischer, Jeremy and Hallock, Barbara and Knepper, Richard and Stewart, Craig},
 doi = {10.1145/2949550.2949570},
 booktitle = {Proceedings of the XSEDE16 on Diversity, Big Data, and Science at Scale - XSEDE16}
}

The Extreme Science and Engineering Discovery Environ-ment (XSEDE) has created a suite of software that is col-lectively known as the XSEDE-Compatible Basic Cluster (XCBC). It is designed to enable smaller, resource-constrained research groups or universities to quickly and easily imple-ment a computing environment similar to XSEDE comput-ing resources. The XCBC system consists of the Rocks Cluster Manager, developed at the San Diego Supercom-puter Center for use on Gordon and Comet, and an XSEDE-specific "Rocks Roll", containing a selection of libraries, compilers, and scientific software curated by the Campus Bridg-ing (CB) group in the XSEDE project, kept current with those implemented on XSEDE resources. The Campus Bridging team has helped several universities implement the XCBC, and finds the design to be extremely useful for resourcelimited (in time, administrator knowledge, or funding) research groups or institutions. Here, we detail our recent experiences in implementing the XCBC design at university campuses across the country. These XCBC implementations were carried out with Campus Bridging staff traveling on-site to the partner institutions to directly assist with the cluster build. In implementing XCBC on campuses, we found that number of the needs described by campus communities as well as the broader cyberinfrastructure community are solved by technical means, although financial issues remain. The remaining issue to be addressed is technical interoperation between systems, and we describe efforts to improve here.

@inproceedings{
 title = {Evaluation of SMP shared memory machines for use with in-memory and OpenMP big data applications},
 type = {inproceedings},
 year = {2016},
 keywords = {Application programming interfaces (API),Astrophy,Distributed computer systems,SGI UV,ScaleMP,Symmetric multi processing,Virt},
 pages = {1597-1606},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84991677893&doi=10.1109%2FIPDPSW.2016.133&partnerID=40&md5=f5680d15f3df0b6086275edcd9a8457e},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 id = {04133db5-1d44-3127-ad88-37582054a599},
 created = {2019-10-01T17:21:28.041Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:42.831Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Younge20161597},
 source_type = {conference},
 notes = {cited By 0; Conference of 30th IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2016 ; Conference Date: 23 May 2016 Through 27 May 2016; Conference Code:122812},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {While distributed memory systems have shaped the field of distributed systems for decades, the demand for many-core shared memory resources is increasing. Symmetric Multiprocessor Systems (SMPs) have become increasingly important recently among a wide array of disciplines, rangingfrom Bioinformatics to astrophysics, and beyond. With the increase in big data computing, the size and scope of traditional commodity server systems is often outpaced. While some big data applications can be mapped to distributed memory systems found through many cluster and cloud technologies today, this effort represents a large barrier of entry that some projects cannot cross. Shared memory SMP systems look to effectively and efficiently fill this niche within distributed systems by providing high throughput and performance with minimized development effort, as the computing environment often represents what many researchers are already familiar with. In this paper, we look at the use of two common shared memory systems, the ScaleMP vSMP virtualized SMP deployment at Indiana University, and the SGI UV architecture deployed at University of Arizona. While both systems are notably different in their design, their potential impact on computing is remarkably similar. As such, we look to compare each system first under a set of OpenMP threaded benchmarks via the SPEC group, and to follow up with our experience using each machine for Trinity de-novo assembly. We find both SMP systems are well suited to support various big data applications, with the newer vSMP deployment often slightly faster, however, certain caveats and performance considerations are necessary when considering such SMP systems. © 2016 IEEE.},
 bibtype = {inproceedings},
 author = {Younge, A J and Reidy, C and Henschel, R and Fox, Geoffrey Charles},
 doi = {10.1109/IPDPSW.2016.133},
 booktitle = {Proceedings - 2016 IEEE 30th International Parallel and Distributed Processing Symposium, IPDPS 2016}
}

While distributed memory systems have shaped the field of distributed systems for decades, the demand for many-core shared memory resources is increasing. Symmetric Multiprocessor Systems (SMPs) have become increasingly important recently among a wide array of disciplines, rangingfrom Bioinformatics to astrophysics, and beyond. With the increase in big data computing, the size and scope of traditional commodity server systems is often outpaced. While some big data applications can be mapped to distributed memory systems found through many cluster and cloud technologies today, this effort represents a large barrier of entry that some projects cannot cross. Shared memory SMP systems look to effectively and efficiently fill this niche within distributed systems by providing high throughput and performance with minimized development effort, as the computing environment often represents what many researchers are already familiar with. In this paper, we look at the use of two common shared memory systems, the ScaleMP vSMP virtualized SMP deployment at Indiana University, and the SGI UV architecture deployed at University of Arizona. While both systems are notably different in their design, their potential impact on computing is remarkably similar. As such, we look to compare each system first under a set of OpenMP threaded benchmarks via the SPEC group, and to follow up with our experience using each machine for Trinity de-novo assembly. We find both SMP systems are well suited to support various big data applications, with the newer vSMP deployment often slightly faster, however, certain caveats and performance considerations are necessary when considering such SMP systems. © 2016 IEEE.

@techreport{
 title = {NSF High Performance Computing System Acquisition system description: Jetstream-a self-provisioned, scalable science and engineering cloud environment},
 type = {techreport},
 year = {2016},
 id = {b961a973-8d25-3323-ab1d-ac2389e4b0ab},
 created = {2019-10-01T17:21:28.814Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:31.696Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016g},
 source_type = {JOUR},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Stanzione, Daniel and Cockerill, Timothy and Skidmore, Edwin and Fischer, Jeremy and Lowe, John Michael and Hammond, Bret and Turner, George and Hancock, David Y and Miller, Therese}
}

@techreport{
 title = {Trident: Scalable compute archives: Workflows, visualization, and analysis},
 type = {techreport},
 year = {2016},
 source = {Proceedings of SPIE - The International Society for Optical Engineering},
 keywords = {AngularJS,Applicat,Application programming interfaces (API),Big data,Docker,IU trident,Javascript,Micros},
 volume = {9913},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85006489750&doi=10.1117%2F12.2233111&partnerID=40&md5=b9ffe3c3a1a9583114539c136797249f},
 publisher = {SPIE},
 id = {075956db-5e5f-3f6f-a6d6-8e8dc4c7c29a},
 created = {2019-10-01T17:21:28.917Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:28.967Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gopu2016},
 source_type = {conference},
 notes = {cited By 0; Conference of Software and Cyberinfrastructure for Astronomy IV ; Conference Date: 26 June 2016 Through 30 June 2016; Conference Code:125147},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {The Astronomy scientific community has embraced Big Data processing challenges, e.g. Associated with time-domain astronomy, and come up with a variety of novel and efficient data processing solutions. However, data processing is only a small part of the Big Data challenge. Efficient knowledge discovery and scientific advancement in the Big Data era requires new and equally efficient tools: modern user interfaces for searching, identifying and viewing data online without direct access to the data; tracking of data provenance; searching, plotting and analyzing metadata; interactive visual analysis, especially of (time-dependent) image data; and the ability to execute pipelines on supercomputing and cloud resources with minimal user overhead or expertise even to novice computing users. The Trident project at Indiana University offers a comprehensive web and cloud-based microservice software suite that enables the straight forward deployment of highly customized Scalable Compute Archive (SCA) systems; including extensive visualization and analysis capabilities, with minimal amount of additional coding. Trident seamlessly scales up or down in terms of data volumes and computational needs, and allows feature sets within a web user interface to be quickly adapted to meet individual project requirements. Domain experts only have to provide code or business logic about handling/visualizing their domain's data products and about executing their pipelines and application work flows. Trident's microservices architecture is made up of light-weight services connected by a REST API and/or a message bus; a web interface elements are built using NodeJS, AngularJS, and HighCharts JavaScript libraries among others while backend services are written in NodeJS, PHP/Zend, and Python. The software suite currently consists of (1) a simple work flow execution framework to integrate, deploy, and execute pipelines and applications (2) a progress service to monitor work flows and sub-work flows (3) ImageX, an interactive image visualization service (3) an authentication and authorization service (4) a data service that handles archival, staging and serving of data products, and (5) a notification service that serves statistical collation and reporting needs of various projects. Several other additional components are under development. Trident is an umbrella project, that evolved from the One Degree Imager, Portal, Pipeline, and Archive (ODI-PPA) project which we had initially refactored toward (1) a powerful analysis/visualization portal for Globular Cluster System (GCS) survey data collected by IU researchers, 2) a data search and download portal for the IU Electron Microscopy Center's data (EMC-SCA), 3) a prototype archive for the Ludwig Maximilian University's Wide Field Imager. The new Trident software has been used to deploy (1) a metadata quality control and analytics portal (RADY-SCA) for DICOM formatted medical imaging data produced by the IU Radiology Center, 2) Several prototype work flows for different domains, 3) a snapshot tool within IU's Karst Desktop environment, 4) a limited component-set to serve GIS data within the IU GIS web portal. Trident SCA systems leverage supercomputing and storage resources at Indiana University but can be configured to make use of any cloud/grid resource, from local workstations/servers to (inter)national supercomputing facilities such as XSEDE. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.},
 bibtype = {techreport},
 author = {Gopu, A and Hayashi, S and Young, M D and Kotulla, R and Henschel, R and Harbeck, D},
 editor = {Chiozzi G., Guzman J C},
 doi = {10.1117/12.2233111}
}

The Astronomy scientific community has embraced Big Data processing challenges, e.g. Associated with time-domain astronomy, and come up with a variety of novel and efficient data processing solutions. However, data processing is only a small part of the Big Data challenge. Efficient knowledge discovery and scientific advancement in the Big Data era requires new and equally efficient tools: modern user interfaces for searching, identifying and viewing data online without direct access to the data; tracking of data provenance; searching, plotting and analyzing metadata; interactive visual analysis, especially of (time-dependent) image data; and the ability to execute pipelines on supercomputing and cloud resources with minimal user overhead or expertise even to novice computing users. The Trident project at Indiana University offers a comprehensive web and cloud-based microservice software suite that enables the straight forward deployment of highly customized Scalable Compute Archive (SCA) systems; including extensive visualization and analysis capabilities, with minimal amount of additional coding. Trident seamlessly scales up or down in terms of data volumes and computational needs, and allows feature sets within a web user interface to be quickly adapted to meet individual project requirements. Domain experts only have to provide code or business logic about handling/visualizing their domain's data products and about executing their pipelines and application work flows. Trident's microservices architecture is made up of light-weight services connected by a REST API and/or a message bus; a web interface elements are built using NodeJS, AngularJS, and HighCharts JavaScript libraries among others while backend services are written in NodeJS, PHP/Zend, and Python. The software suite currently consists of (1) a simple work flow execution framework to integrate, deploy, and execute pipelines and applications (2) a progress service to monitor work flows and sub-work flows (3) ImageX, an interactive image visualization service (3) an authentication and authorization service (4) a data service that handles archival, staging and serving of data products, and (5) a notification service that serves statistical collation and reporting needs of various projects. Several other additional components are under development. Trident is an umbrella project, that evolved from the One Degree Imager, Portal, Pipeline, and Archive (ODI-PPA) project which we had initially refactored toward (1) a powerful analysis/visualization portal for Globular Cluster System (GCS) survey data collected by IU researchers, 2) a data search and download portal for the IU Electron Microscopy Center's data (EMC-SCA), 3) a prototype archive for the Ludwig Maximilian University's Wide Field Imager. The new Trident software has been used to deploy (1) a metadata quality control and analytics portal (RADY-SCA) for DICOM formatted medical imaging data produced by the IU Radiology Center, 2) Several prototype work flows for different domains, 3) a snapshot tool within IU's Karst Desktop environment, 4) a limited component-set to serve GIS data within the IU GIS web portal. Trident SCA systems leverage supercomputing and storage resources at Indiana University but can be configured to make use of any cloud/grid resource, from local workstations/servers to (inter)national supercomputing facilities such as XSEDE. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

@inproceedings{
 title = {RabbitQR: Fast and flexible Big Data Processing at LSST data rates using existing, shared-use hardware},
 type = {inproceedings},
 year = {2016},
 id = {dc28ea11-9d62-324b-9aec-6ce5bcb5eac8},
 created = {2019-10-01T17:21:29.012Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:29.247Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kotulla2016},
 source_type = {JOUR},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Kotulla, Ralf and Gopu, Arvind and Hayashi, Soichi},
 doi = {10.1117/12.2233527},
 booktitle = {SPIE 9913, Software and Cyberinfrastructure for Astronomy}
}

@inproceedings{
 title = {Cyberinfrastructure as a Platform to Facilitate Effective Collaboration between Institutions and Support Collaboratories},
 type = {inproceedings},
 year = {2016},
 pages = {37-42},
 volume = {06-09-Nove},
 websites = {https://dl.acm.org/doi/10.1145/2974927.2974962},
 month = {11},
 publisher = {ACM},
 city = {New York, NY, USA},
 id = {a52f8404-2d93-3ac9-bcc6-34780ef45197},
 created = {2019-10-01T18:06:10.043Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T23:36:28.841Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Coulter2016a},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Researchers, scientists, engineers, granting agencies, and increasingly complex research problems have given rise to the scientific "collaboratory"-large organizations that span many institutions, with individual members working together to explore a particular phenomenon. These organizations require computational resources in order to support analyses and to provide platforms where the collaborators can interact. The XSEDE Community Infrastructure (XCI) group assists campuses in using their own resources and promotes the sharing of those resources in order to create collaboratories improving use of the nation's collective cyberinfrastructure. Currently XCI provides toolkits and training, and collaborates with organizations such as ACI-REF, XSEDE Campus Champions, and the Open Science Grid to identify tools and best practices that support the community. This paper discusses the progress in and barriers to developing a robust collaborative environment where computational resources can be shared.},
 bibtype = {inproceedings},
 author = {Coulter, Eric and Fischer, Jeremy and Hallock, Barbara and Knepper, Richard and Lifka, Dave and Navarro, J.P. and Pierce, Marlon and Stewart, Craig},
 doi = {10.1145/2974927.2974962},
 booktitle = {Proceedings of the 2016 ACM SIGUCCS Annual Conference}
}

Researchers, scientists, engineers, granting agencies, and increasingly complex research problems have given rise to the scientific "collaboratory"-large organizations that span many institutions, with individual members working together to explore a particular phenomenon. These organizations require computational resources in order to support analyses and to provide platforms where the collaborators can interact. The XSEDE Community Infrastructure (XCI) group assists campuses in using their own resources and promotes the sharing of those resources in order to create collaboratories improving use of the nation's collective cyberinfrastructure. Currently XCI provides toolkits and training, and collaborates with organizations such as ACI-REF, XSEDE Campus Champions, and the Open Science Grid to identify tools and best practices that support the community. This paper discusses the progress in and barriers to developing a robust collaborative environment where computational resources can be shared.

@techreport{
 title = {Pervasive Technology Institute Annual Report: Research Innovations and Advanced Cyberinfrastructure Services in Support of IU Strategic Goals During FY 2016},
 type = {techreport},
 year = {2016},
 id = {98df9041-2cda-3ad8-9eb4-7797336c67c5},
 created = {2019-10-01T18:06:10.303Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:06:29.379Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2016},
 source_type = {RPRT},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863,ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Plale, Beth and Welch, Von and Fox, Geoffrey Charles and Link, Matthew R and Miller, Therese and Wernert, Eric A and Boyles, Michael J and Fulton, Ben and Hancock, David Y}
}

@article{
 title = {Gathering requirements for advancing simulations in HPC infrastructures via science gateways},
 type = {article},
 year = {2016},
 keywords = {Computation theory,Computational science,Computational simulation,Distributed computer systems,Large scale systems,Web crawl},
 pages = {544-554},
 volume = {82},
 websites = {https://linkinghub.elsevier.com/retrieve/pii/S0167739X17303011,https://www.scopus.com/inward/record.uri?eid=2-s2.0-85015395737&doi=10.1016%2Fj.future.2017.02.042&partnerID=40&md5=083c74b7a31e9af48390ff775de4f7cf},
 month = {5},
 publisher = {North-Holland},
 id = {e20bd754-ab1e-34b1-93fc-c08ec030102d},
 created = {2019-10-01T18:06:11.081Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:06:53.378Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {gesing2017gathering},
 source_type = {article},
 notes = {<b>From Duplicate 2 (<i>Gathering requirements for advancing simulations in HPC infrastructures via science gateways</i> - Gesing, S; Dooley, R; Pierce, M; Krüger, J; Grunzke, R; Herres-Pawlis, S; Hoffmann, A)<br/></b><br/>cited By 0; Article in Press},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 abstract = {Compute-intensive simulations are often based on complex scientific theories and necessitate high-performance computing (HPC) infrastructures to deliver results in reasonable time. While domain researchers are experts in their field and apply sophisticated theoretical models in computational simulations, they are not necessarily also HPC experts or IT specialists in general. Thus, they appreciate easy-to-use solutions tailored to their research, which hide the complex underlying computing and data infrastructures. Science gateways form such end-to-end solutions and their development for compute-intensive simulations necessitates expertise to connect HPC research infrastructures including grid and cloud infrastructures to support with the efficient access to such resources. HPC experts and IT specialists fulfilling this task may have only rudimentary knowledge about the research domain of a simulation. Thus, it is crucial that they gather the requirements of a research use case, which they aim to support efficiently via a science gateway. In the last 10 years quite a few web development frameworks, science gateway frameworks and APIs with different foci and strengths have evolved to support the developers of science gateways in implementing an intuitive solution for a target research domain. The selection of a suitable technology for a specific use case is essential and helps reducing the effort in implementing the science gateway by re-using existing software or frameworks. Thus, a solution for a user community can be provided more efficiently. This paper introduces the general architecture of science gateways, goes into detail for criteria to design science gateways efficiently and gives examples of mature science gateways and science gateway frameworks.},
 bibtype = {article},
 author = {Gesing, Sandra and Dooley, Rion and Pierce, Marlon and Krüger, Jens and Grunzke, Richard and Herres-Pawlis, Sonja and Hoffmann, Alexander},
 doi = {10.1016/j.future.2017.02.042},
 journal = {Future Generation Computer Systems}
}

Compute-intensive simulations are often based on complex scientific theories and necessitate high-performance computing (HPC) infrastructures to deliver results in reasonable time. While domain researchers are experts in their field and apply sophisticated theoretical models in computational simulations, they are not necessarily also HPC experts or IT specialists in general. Thus, they appreciate easy-to-use solutions tailored to their research, which hide the complex underlying computing and data infrastructures. Science gateways form such end-to-end solutions and their development for compute-intensive simulations necessitates expertise to connect HPC research infrastructures including grid and cloud infrastructures to support with the efficient access to such resources. HPC experts and IT specialists fulfilling this task may have only rudimentary knowledge about the research domain of a simulation. Thus, it is crucial that they gather the requirements of a research use case, which they aim to support efficiently via a science gateway. In the last 10 years quite a few web development frameworks, science gateway frameworks and APIs with different foci and strengths have evolved to support the developers of science gateways in implementing an intuitive solution for a target research domain. The selection of a suitable technology for a specific use case is essential and helps reducing the effort in implementing the science gateway by re-using existing software or frameworks. Thus, a solution for a user community can be provided more efficiently. This paper introduces the general architecture of science gateways, goes into detail for criteria to design science gateways efficiently and gives examples of mature science gateways and science gateway frameworks.

@inproceedings{
 title = {A PetaFLOPS Supercomputer as a Campus Resource: Innovation, Impact, and Models for Locally-Owned High Performance Computing at Research Colleges and Universities},
 type = {inproceedings},
 year = {2016},
 keywords = {HPC,Impact,Innovation,Petaops,Supercomputing,University,Value,analytics},
 pages = {61-68},
 volume = {06-09-Nove},
 websites = {https://dl.acm.org/doi/10.1145/2974927.2974956},
 month = {11},
 publisher = {ACM},
 day = {1},
 city = {New York, NY, USA},
 id = {f33173b0-676a-379b-9a54-4513feb85038},
 created = {2019-12-02T16:45:54.317Z},
 accessed = {2019-12-02},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:33:19.837Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Thota2016},
 private_publication = {false},
 abstract = {In 1997, Indiana University (IU) began a purposeful and steady drive to expand the use of supercomputers and what we now call cyberinfrastructure. In 2001, IU implemented the first 1 TFLOPS supercomputer owned by and operated for a single US University. In 2013, IU made an analogous investment and achievement at the 1 PFLOPS level: Big Red II, a Cray XE6/XK7, was the first supercomputer capable of 1 PFLOPS (theoretical) performance that was a dedicated university resource [2]. IU0s high performance computing (HPC) resources have fostered innovation in disciplines from biology to chemistry to medicine. Currently, 185 disciplines and sub disciplines are represented on Big Red II with a wide variety of usage needs. Quantitative data suggest that investment in this supercomputer has been a good value to IU in terms of academic achievement and federal grant income. Here we will discuss how investment in Big Red II has benefited IU, and argue that locally-owned computational resources (scaled appropriately to needs and budgets) may be of benefit to many colleges and universities. We will also discuss software tools under development that will aid others in quantifying the benefit of investment in high performance computing to their campuses.},
 bibtype = {inproceedings},
 author = {Thota, Abhinav S. and Fulton, Ben and Weakley, Le Mai Weakley and Henschel, Robert and Hancock, David Y. and Allen, Matt and Tillotson, Jenett and Link, Matt and Stewart, Craig A.},
 doi = {10.1145/2974927.2974956},
 booktitle = {Proceedings of the 2016 ACM SIGUCCS Annual Conference}
}

In 1997, Indiana University (IU) began a purposeful and steady drive to expand the use of supercomputers and what we now call cyberinfrastructure. In 2001, IU implemented the first 1 TFLOPS supercomputer owned by and operated for a single US University. In 2013, IU made an analogous investment and achievement at the 1 PFLOPS level: Big Red II, a Cray XE6/XK7, was the first supercomputer capable of 1 PFLOPS (theoretical) performance that was a dedicated university resource [2]. IU0s high performance computing (HPC) resources have fostered innovation in disciplines from biology to chemistry to medicine. Currently, 185 disciplines and sub disciplines are represented on Big Red II with a wide variety of usage needs. Quantitative data suggest that investment in this supercomputer has been a good value to IU in terms of academic achievement and federal grant income. Here we will discuss how investment in Big Red II has benefited IU, and argue that locally-owned computational resources (scaled appropriately to needs and budgets) may be of benefit to many colleges and universities. We will also discuss software tools under development that will aid others in quantifying the benefit of investment in high performance computing to their campuses.

@inproceedings{
 title = {Jetstream: a self-provisioned, scalable science and engineering cloud environment},
 type = {inproceedings},
 year = {2015},
 pages = {1-8},
 volume = {2015-July},
 websites = {http://dl.acm.org/citation.cfm?doid=2792745.2792774},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {1590fe21-c3d2-306b-ace5-e830ad6b38bf},
 created = {2017-10-24T15:01:35.592Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T19:33:20.068Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015b},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Copyright © 2015 ACM. Jetstream will be the first production cloud resource supporting general science and engineering research within the XD ecosystem. In this report we describe the motivation for proposing Jetstream, the configuration of the Jetstream system as funded by the NSF, the team that is implementing Jetstream, and the communities we expect to use this new system. Our hope and plan is that Jetstream, which will become available for production use in 2016, will aid thousands of researchers who need modest amounts of computing power interactively. The implementation of Jetstream should increase the size and disciplinary diversity of the US research community that makes use of the resources of the XD ecosystem.},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Cockerill, Timothy M. and Foster, Ian and Hancock, David and Merchant, Nirav and Skidmore, Edwin and Stanzione, Daniel and Taylor, James and Tuecke, Steven and Turner, George and Vaughn, Matthew and Gaffney, Niall I.},
 doi = {10.1145/2792745.2792774},
 booktitle = {Proceedings of the 2015 XSEDE Conference on Scientific Advancements Enabled by Enhanced Cyberinfrastructure - XSEDE '15}
}

Copyright © 2015 ACM. Jetstream will be the first production cloud resource supporting general science and engineering research within the XD ecosystem. In this report we describe the motivation for proposing Jetstream, the configuration of the Jetstream system as funded by the NSF, the team that is implementing Jetstream, and the communities we expect to use this new system. Our hope and plan is that Jetstream, which will become available for production use in 2016, will aid thousands of researchers who need modest amounts of computing power interactively. The implementation of Jetstream should increase the size and disciplinary diversity of the US research community that makes use of the resources of the XD ecosystem.

@inproceedings{
 title = {Dynamic parallelism for simple and efficient gpu graph algorithms},
 type = {inproceedings},
 year = {2015},
 pages = {11},
 publisher = {ACM},
 id = {2fd17b06-35a4-3ad6-9a65-eda55f6871e0},
 created = {2017-11-22T21:02:01.359Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:31.003Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Zhang2015},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Zhang, Peter and Holk, Eric and Matty, John and Misurda, Samantha and Zalewski, Marcin and Chu, Jonathan and McMillan, Scott and Lumsdaine, Andrew},
 booktitle = {Proceedings of the 5th Workshop on Irregular Applications: Architectures and Algorithms}
}

@inproceedings{
 title = {Dynamic Adaptation for Elastic System Services Using Virtual Servers},
 type = {inproceedings},
 year = {2015},
 pages = {125-134},
 publisher = {IEEE},
 id = {c2653407-62c9-35ef-9de4-caaa3689eb71},
 created = {2017-11-22T21:02:48.362Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:31.306Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Kulkarni2015},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Kulkarni, Abhishek and Greenberg, Hugh and Lang, Michael and Lumsdaine, Andrew},
 booktitle = {High Performance Computing (HiPC), 2015 IEEE 22nd International Conference on}
}

@article{
 title = {Fast Data Management with Distributed Streaming SQL},
 type = {article},
 year = {2015},
 id = {4fd582e0-fa3c-3aa6-b751-b895d9c87412},
 created = {2017-11-28T16:29:35.546Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:04.122Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pathirage2015},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {article},
 author = {Pathirage, Milinda and Plale, Beth},
 journal = {arXiv preprint arXiv:1511.03935}
}

@inproceedings{
 title = {Towards an Understanding of Scalable Query and Data Analysis for Social Media Data using High-Level Dataflow Systems},
 type = {inproceedings},
 year = {2015},
 id = {e45f5441-1edb-3e38-87bb-1f7d7514a079},
 created = {2017-11-28T17:33:38.570Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.287Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wu2015a},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Wu, Tak Lon Stephen and Qiu, Judy},
 booktitle = {SC15 27th annual International Conference for High Performance Computing, Networking, Storage and Analysis}
}

@techreport{
 title = {Parallel LDA through synchronized communication optimizations},
 type = {techreport},
 year = {2015},
 publisher = {Technical report},
 id = {b61b2a1f-6e6e-3476-b5df-fa8283cdf236},
 created = {2017-11-28T17:33:38.887Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.777Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2015c},
 source_type = {RPRT},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {techreport},
 author = {Zhang, Bingjing and Peng, Bo and Qiu, Judy}
}

@article{
 title = {Towards a comprehensive set of big data benchmarks},
 type = {article},
 year = {2015},
 pages = {47},
 volume = {26},
 publisher = {IOS Press},
 id = {954d88fc-d02b-3bab-8b64-04ed95cd40c6},
 created = {2017-11-28T17:33:39.319Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.537Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2015},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {article},
 author = {Fox, Geoffrey Charles and Jha, S and Qiu, J and Ekanazake, S and Luckow, André},
 journal = {Big Data and High Performance Computing}
}

@inproceedings{
 title = {Parallel clustering of high-dimensional social media data streams},
 type = {inproceedings},
 year = {2015},
 pages = {323-332},
 publisher = {IEEE},
 id = {4cec1c9a-3570-3bef-b117-84e21d7514eb},
 created = {2017-11-28T17:33:39.330Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:49.941Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gao2015},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Gao, Xiaoming and Ferrara, Emilio and Qiu, Judy},
 booktitle = {Cluster, Cloud and Grid Computing (CCGrid), 2015 15th IEEE/ACM International Symposium on}
}

@techreport{
 title = {A High Order Multi-Scale Numerical Approach for Kinetic Simulations},
 type = {techreport},
 year = {2015},
 publisher = {HOUSTON UNIV TX},
 id = {c43c6c0d-b722-3d6b-baf5-f8c543ca49ed},
 created = {2017-11-28T17:33:40.859Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.360Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Qiu2015},
 source_type = {RPRT},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {techreport},
 author = {Qiu, Jingmei}
}

@techreport{
 title = {Cyberinfrastructure Value Assessment Report},
 type = {techreport},
 year = {2015},
 websites = {http://hdl.handle.net/2022/20568},
 institution = {I-STEM Education Initiative University of Illinois at Urbana-Champaign},
 id = {d1ba6929-a0eb-3f4b-9ef5-394622690329},
 created = {2017-12-18T16:29:17.553Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:46.726Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {DeStafanoL.2015},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {techreport},
 author = {DeStafano, L., L. Rivera}
}

@inproceedings{
 title = {Scalable Relativistic High-Resolution Shock-Capturing for Heterogeneous Computing},
 type = {inproceedings},
 year = {2015},
 pages = {611-618},
 publisher = {IEEE},
 id = {db8ca808-0531-3f3d-9fe7-cf5bd1edf459},
 created = {2017-12-18T17:23:38.146Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.064Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Glines2015},
 source_type = {CONF},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Glines, Forrest Wolfgang and Anderson, Matthew and Neilsen, David},
 booktitle = {Cluster Computing (CLUSTER), 2015 IEEE International Conference on}
}

@article{
 title = {Relativistic Hydrodynamics with Wavelets},
 type = {article},
 year = {2015},
 id = {8a29c344-a12c-3280-ad1a-8dc391bca88b},
 created = {2017-12-18T17:23:38.206Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.504Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {DeBuhr2015},
 source_type = {JOUR},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {article},
 author = {DeBuhr, Jackson and Zhang, Bo and Anderson, Matthew and Neilsen, David and Hirschmann, Eric W},
 journal = {arXiv preprint arXiv:1512.00386}
}

@inproceedings{
 title = {Pixel-oriented techniques for visualizing next-generation HPC systems},
 type = {inproceedings},
 year = {2015},
 pages = {160-164},
 publisher = {IEEE},
 id = {23125774-3da0-390a-ad6b-518392c2fa78},
 created = {2017-12-18T18:01:12.615Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:15.881Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Cottam2015},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Cottam, Joseph and Martin, Ben and Dalessandro, Luke and Lumsdaine, Andrew},
 booktitle = {Software Visualization (VISSOFT), 2015 IEEE 3rd Working Conference on}
}

@article{
 title = {Supervised classification of enzyme residue function using machine learning methods},
 type = {article},
 year = {2015},
 volume = {2},
 id = {0d45045f-f63f-348c-9841-a7f8833f591a},
 created = {2017-12-18T21:43:37.408Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.386Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Youn2015},
 source_type = {JOUR},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {article},
 author = {Youn, Eunseog and Radivojac, Predrag and Peters, Brandon and Moad, Charles and Heiland, Randy and Mooney, Sean D},
 journal = {Machine Learning and Applications: An International Journa}
}

@inproceedings{
 title = {White Paper: Big Data, Simulations and HPC Convergence},
 type = {inproceedings},
 year = {2015},
 pages = {3-17},
 volume = {16},
 publisher = {Springer},
 id = {d7edbcfd-ef87-34d2-947e-df82a7899fd3},
 created = {2017-12-18T21:44:00.976Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.999Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2015b},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Fox, Geoffrey Charles and Qiu, Judy and Jha, Shantenu and Ekanayake, Saliya and Kamburugamuve, Supun},
 booktitle = {Workshop on Big Data Benchmarks}
}

@inproceedings{
 title = {Harp: Collective communication on hadoop},
 type = {inproceedings},
 year = {2015},
 pages = {228-233},
 publisher = {IEEE},
 id = {dbf7b1bb-56b9-33fd-b67c-ab39915211ab},
 created = {2017-12-18T21:44:04.370Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.473Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2015b},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {© 2015 IEEE. Big data processing tools have evolved rapidly in recent years. MapReduce has proven very successful but is not optimized for many important analytics, especially those involving iteration. In this regard, Iterative MapReduce frameworks improve performance of MapReduce job chains through caching. Further, Pregel, Giraph and GraphLab abstract data as a graph and process it in iterations. But all these tools are designed with a fixed data abstraction and have limited collective communication support to synchronize application data and algorithm control states among parallel processes. In this paper, we introduce a collective communication abstraction layer which provides efficient collective communication operations on several common data abstractions such as arrays, key-values and graphs, and define a MapCollective programming model which serves the diverse collective communication demands in different parallel algorithms. We implement a library called Harp to provide the features above and plug it into Hadoop so that applications abstracted in MapCollective model can be easily developed on top of MapReduce framework and conveniently integrated with other tools in Apache Big Data Stack. With improved expressiveness in the abstraction and excellent performance on the implementation, we can simultaneously support various applications from HPC to Cloud systems together with high performance.},
 bibtype = {inproceedings},
 author = {Zhang, Bingjing and Ruan, Yang and Qiu, Judy},
 doi = {10.1109/IC2E.2015.35},
 booktitle = {Cloud Engineering (IC2E), 2015 IEEE International Conference on}
}

© 2015 IEEE. Big data processing tools have evolved rapidly in recent years. MapReduce has proven very successful but is not optimized for many important analytics, especially those involving iteration. In this regard, Iterative MapReduce frameworks improve performance of MapReduce job chains through caching. Further, Pregel, Giraph and GraphLab abstract data as a graph and process it in iterations. But all these tools are designed with a fixed data abstraction and have limited collective communication support to synchronize application data and algorithm control states among parallel processes. In this paper, we introduce a collective communication abstraction layer which provides efficient collective communication operations on several common data abstractions such as arrays, key-values and graphs, and define a MapCollective programming model which serves the diverse collective communication demands in different parallel algorithms. We implement a library called Harp to provide the features above and plug it into Hadoop so that applications abstracted in MapCollective model can be easily developed on top of MapReduce framework and conveniently integrated with other tools in Apache Big Data Stack. With improved expressiveness in the abstraction and excellent performance on the implementation, we can simultaneously support various applications from HPC to Cloud systems together with high performance.

@inproceedings{
 title = {Hpc-abds high performance computing enhanced apache big data stack},
 type = {inproceedings},
 year = {2015},
 pages = {1057-1066},
 publisher = {IEEE},
 id = {669070b7-beab-3f33-ae60-eb0dd6a9af61},
 created = {2017-12-18T21:44:04.474Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.606Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2015a},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {© 2015 IEEE. We review the High Performance Computing Enhanced Apache Big Data Stack HPC-ABDS and summarize the capabilities in 21 identified architecture layers. These cover Message and Data Protocols, Distributed Coordination, Security  &  Privacy, Monitoring, Infrastructure Management, DevOps, Interoperability, File Systems, Cluster  &  Resource management, Data Transport, File management, NoSQL, SQL (NewSQL), Extraction Tools, Object-relational mapping, In-memory caching and databases, Inter-process Communication, Batch Programming model and Runtime, Stream Processing, High-level Programming, Application Hosting and PaaS, Libraries and Applications, Workflow and Orchestration. We summarize status of these layers focusing on issues of importance for data analytics. We highlight areas where HPC and ABDS have good opportunities for integration.},
 bibtype = {inproceedings},
 author = {Fox, G.C. Geoffrey C and Qiu, Judy and Kamburugamuve, Supun and Jha, Shantenu and Luckow, Andre},
 doi = {10.1109/CCGrid.2015.122},
 booktitle = {Cluster, Cloud and Grid Computing (CCGrid), 2015 15th IEEE/ACM International Symposium on}
}

© 2015 IEEE. We review the High Performance Computing Enhanced Apache Big Data Stack HPC-ABDS and summarize the capabilities in 21 identified architecture layers. These cover Message and Data Protocols, Distributed Coordination, Security & Privacy, Monitoring, Infrastructure Management, DevOps, Interoperability, File Systems, Cluster & Resource management, Data Transport, File management, NoSQL, SQL (NewSQL), Extraction Tools, Object-relational mapping, In-memory caching and databases, Inter-process Communication, Batch Programming model and Runtime, Stream Processing, High-level Programming, Application Hosting and PaaS, Libraries and Applications, Workflow and Orchestration. We summarize status of these layers focusing on issues of importance for data analytics. We highlight areas where HPC and ABDS have good opportunities for integration.

@inproceedings{
 title = {Importance of runtime considerations in performance engineering of large-scale distributed graph algorithms},
 type = {inproceedings},
 year = {2015},
 pages = {553-564},
 publisher = {Springer},
 id = {7249519c-034a-37d9-92c4-eede33e6154d},
 created = {2018-01-19T07:02:39.332Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:41.617Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Firoz2015},
 source_type = {CONF},
 folder_uuids = {2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Firoz, Jesun Sahariar and Kanewala, Thejaka Amila and Zalewski, Marcin and Barnas, Martina and Lumsdaine, Andrew},
 booktitle = {European Conference on Parallel Processing}
}

@inproceedings{
 title = {Comparison of single source shortest path algorithms on two recent asynchronous many-task runtime systems},
 type = {inproceedings},
 year = {2015},
 pages = {674-681},
 publisher = {IEEE},
 id = {0fa5e0bb-8879-3c2f-ab23-e3fcb4f922a0},
 created = {2018-01-23T20:26:56.203Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:21.039Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Firoz2015a},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Firoz, Jesun Sahariar and Barnas, Martina and Zalewski, Marcin and Lumsdaine, Andrew},
 booktitle = {Parallel and Distributed Systems (ICPADS), 2015 IEEE 21st International Conference on}
}

@inproceedings{
 title = {Sustained software for cyberinfrastructure: Analyses of successful efforts with a focus on NSF-funded software},
 type = {inproceedings},
 year = {2015},
 pages = {63-72},
 websites = {http://dl.acm.org/citation.cfm?doid=2753524.2753533},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {3a46d027-0880-3c1f-81f8-b9ccd3940f32},
 created = {2019-09-11T16:14:33.675Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:00.948Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015},
 private_publication = {false},
 abstract = {Reliable software that provides needed functionality is clearly essential for an effective distributed cyberinfrastructure (CI) that supports comprehensive, balanced, and flexible distributed CI. Effective distributed cyberinfrastructure, in turn, supports science and engineering applications. The purpose of this study was to understand what factors lead to software projects being well sustained over the long run, focusing on software created with funding from the US National Science Foundation (NSF) and/or used by researchers funded by the NSF. We surveyed NSF-funded researchers and performed in-depth studies of software projects that have been sustained over many years. Successful projects generally used open-source software licenses and employed good software engineering practices and test practices. However, many projects that have not been well sustained over time also met these criteria. The features that stood out about successful projects included deeply committed leadership and some sort of user forum or conference at least annually. In some cases, software project leaders have employed multiple financial strategies over the course of a decades-old software project. Such well-sustained software is used in major distributed CI projects that support thousands of users, and this software is critical to the operation of major distributed CI facilities in the US. The findings of our study identify some characteristics of software that is relevant to the NSF-supported research community, and that has been sustained over many years.},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Barnett, William K. and Wernert, Eric A. and Wernert, Julie A. and Welch, Von and Knepper, Richard},
 doi = {10.1145/2753524.2753533},
 booktitle = {Proceedings of the 1st Workshop on The Science of Cyberinfrastructure Research, Experience, Applications and Models - SCREAM '15}
}

Reliable software that provides needed functionality is clearly essential for an effective distributed cyberinfrastructure (CI) that supports comprehensive, balanced, and flexible distributed CI. Effective distributed cyberinfrastructure, in turn, supports science and engineering applications. The purpose of this study was to understand what factors lead to software projects being well sustained over the long run, focusing on software created with funding from the US National Science Foundation (NSF) and/or used by researchers funded by the NSF. We surveyed NSF-funded researchers and performed in-depth studies of software projects that have been sustained over many years. Successful projects generally used open-source software licenses and employed good software engineering practices and test practices. However, many projects that have not been well sustained over time also met these criteria. The features that stood out about successful projects included deeply committed leadership and some sort of user forum or conference at least annually. In some cases, software project leaders have employed multiple financial strategies over the course of a decades-old software project. Such well-sustained software is used in major distributed CI projects that support thousands of users, and this software is critical to the operation of major distributed CI facilities in the US. The findings of our study identify some characteristics of software that is relevant to the NSF-supported research community, and that has been sustained over many years.

@inproceedings{
 title = {Apache Airavata as a Laboratory - Architecture and Case Study for Component-Based Gateway Middleware},
 type = {inproceedings},
 year = {2015},
 pages = {19-26},
 websites = {http://dl.acm.org/citation.cfm?doid=2753524.2753529},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 institution = {ACM},
 id = {4846c948-96f1-31a8-bc6f-bb893329fdd9},
 created = {2019-10-01T17:20:14.422Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:51.101Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {marru2015apache},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 abstract = {Science gateways are more than user interfaces to computational grids and clouds. Gateways are middleware in their own right, providing flexible, lightweight federations of heterogenous collections of computing resources (such as campus clusters, supercomputers, computational clouds), all of which remain challenges for many alternative middleware approaches. Gateways also are notable for providing science application-centric interfaces to computing resources rather than resource-centric views. An important challenge for science gateway research is to generalize specific science gateway strategies, defining a reference architecture that emcompasses major gateway capabilities while enabling implementation flexibility. Such a reference architecture should also enable "platform as a service" approaches that provide hosted versions of common gateway capabilities. In this paper, we summarize the Apache Airavata software system as a candidate reference architecture for science gateways. We propose the use of a component-based architecture to encompass major gateway capabilities (such as metadata management, meta-scheduling, execution management, and messaging). We examine the messaging system component in this abstract architecture in detail and describe its re-implementation and validation using third party messaging system software to replace a custom-built messaging system. Besides the operational validation of this specific component, we infer a preliminary validation of the overall architecture. The flexibility of component implementations within an overall architecture is essential as it allows gateway middleware to be the subject of distributed computing research for its own sake while also ensuring that we don't get locked into less than optimal implementations for gateway operations.},
 bibtype = {inproceedings},
 author = {Marru, Suresh and Pierce, Marlon and Pamidighantam, Sudhakar and Wimalasena, Chathuri},
 doi = {10.1145/2753524.2753529},
 booktitle = {Proceedings of the 1st Workshop on The Science of Cyberinfrastructure Research, Experience, Applications and Models - SCREAM '15}
}

Science gateways are more than user interfaces to computational grids and clouds. Gateways are middleware in their own right, providing flexible, lightweight federations of heterogenous collections of computing resources (such as campus clusters, supercomputers, computational clouds), all of which remain challenges for many alternative middleware approaches. Gateways also are notable for providing science application-centric interfaces to computing resources rather than resource-centric views. An important challenge for science gateway research is to generalize specific science gateway strategies, defining a reference architecture that emcompasses major gateway capabilities while enabling implementation flexibility. Such a reference architecture should also enable "platform as a service" approaches that provide hosted versions of common gateway capabilities. In this paper, we summarize the Apache Airavata software system as a candidate reference architecture for science gateways. We propose the use of a component-based architecture to encompass major gateway capabilities (such as metadata management, meta-scheduling, execution management, and messaging). We examine the messaging system component in this abstract architecture in detail and describe its re-implementation and validation using third party messaging system software to replace a custom-built messaging system. Besides the operational validation of this specific component, we infer a preliminary validation of the overall architecture. The flexibility of component implementations within an overall architecture is essential as it allows gateway middleware to be the subject of distributed computing research for its own sake while also ensuring that we don't get locked into less than optimal implementations for gateway operations.

@article{
 title = {Potential for a large earthquake near Los Angeles inferred from the 2014 La Habra earthquake},
 type = {article},
 year = {2015},
 pages = {378-385},
 volume = {2},
 id = {a58b4fa2-4e0f-3239-950e-592e73791b22},
 created = {2019-10-01T17:20:16.893Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:19.273Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {donnellan2015potential},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {article},
 author = {Donnellan, Andrea and Grant Ludwig, Lisa and Parker, Jay W and Rundle, John B and Wang, Jun and Pierce, Marlon and Blewitt, Geoffrey and Hensley, Scott},
 journal = {Earth and Space Science},
 number = {9}
}

@article{
 title = {Web Services for Dynamic Coloring of UAVSAR Images},
 type = {article},
 year = {2015},
 pages = {2325-2332},
 volume = {172},
 publisher = {Springer Basel},
 id = {bf961c19-0c7d-3f5a-b944-88b7c6f58566},
 created = {2019-10-01T17:20:17.779Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:00.687Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {wang2015web},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {article},
 author = {Wang, Jun and Pierce, Marlon and Donnellan, Andrea and Parker, Jay},
 journal = {Pure and Applied Geophysics},
 number = {8}
}

@inproceedings{
 title = {Science gateways-leveraging modeling and simulations in HPC infrastructures via increased usability},
 type = {inproceedings},
 year = {2015},
 pages = {19-26},
 institution = {IEEE},
 id = {27222160-d1f4-3c42-8b3b-3c06fbb2a0f2},
 created = {2019-10-01T17:20:18.010Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:58.885Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {gesing2015science},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Gesing, Sandra and Dooley, Rion and Pierce, Marlon and Krüger, Jens and Grunzke, Richard and Herres-Pawlis, Sonja and Hoffmann, Alexander},
 booktitle = {High Performance Computing & Simulation (HPCS), 2015 International Conference on}
}

@article{
 title = {The GenApp framework integrated with Airavata for managed compute resource submissions},
 type = {article},
 year = {2015},
 pages = {4292-4303},
 volume = {27},
 id = {e500a488-c200-32b6-b2a3-0eb98d53198a},
 created = {2019-10-01T17:20:18.047Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:58.595Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {brookes2015genapp},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,44d48ba6-09b0-4f17-965c-b625aaaba978},
 private_publication = {false},
 bibtype = {article},
 author = {Brookes, Emre H and Anjum, Nadeem and Curtis, Joseph E and Marru, Suresh and Singh, Raminder and Pierce, Marlon},
 journal = {Concurrency and Computation: Practice and Experience},
 number = {16}
}

@article{
 title = {Multihazard simulation and cyberinfrastructure},
 type = {article},
 year = {2015},
 pages = {2083-2085},
 volume = {172},
 publisher = {Springer Basel},
 id = {f66b7a59-67b6-3ea5-9e92-7223a6428717},
 created = {2019-10-01T17:20:18.908Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:03.378Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {donnellan2015multihazard},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {article},
 author = {Donnellan, Andrea and Williams, Charles and Pierce, Marlon},
 journal = {Pure and Applied Geophysics},
 number = {8}
}

@inproceedings{
 title = {Publishing and consuming GLUE v2.0 resource information in XSEDE},
 type = {inproceedings},
 year = {2015},
 pages = {1-8},
 volume = {2015-July},
 websites = {http://dl.acm.org/citation.cfm?doid=2792745.2792770},
 month = {7},
 publisher = {ACM Press},
 day = {26},
 city = {New York, New York, USA},
 id = {3afeeb02-cd2e-3d3d-9c97-2a90bd4ff393},
 created = {2019-10-01T17:20:20.616Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.259Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Smith2015},
 private_publication = {false},
 abstract = {XSEDE users, science gateways, and services need a variety of accurate information about XSEDE resources so that they can use those resources effectively. They need information to decide which resources to use, to track their usage of resources, and to provide services to their users. To support this, XSEDE is deploying a new system to gather and publish static and dynamic resource information. This paper gives an overview of the resource information available with this new system, describes the design and performance of the software and services that make up this system, and finally provides examples of how to use this new resource information.},
 bibtype = {inproceedings},
 author = {Smith, Warren and Pamidighantam, Sudhakar and Navarro, John-Paul},
 doi = {10.1145/2792745.2792770},
 booktitle = {Proceedings of the 2015 XSEDE Conference on Scientific Advancements Enabled by Enhanced Cyberinfrastructure - XSEDE '15}
}

XSEDE users, science gateways, and services need a variety of accurate information about XSEDE resources so that they can use those resources effectively. They need information to decide which resources to use, to track their usage of resources, and to provide services to their users. To support this, XSEDE is deploying a new system to gather and publish static and dynamic resource information. This paper gives an overview of the resource information available with this new system, describes the design and performance of the software and services that make up this system, and finally provides examples of how to use this new resource information.

@article{
 title = {Ab Initio Studies of Calcium Carbonate Hydration},
 type = {article},
 year = {2015},
 keywords = {CaCO3,bidentate,cluster,hydration,monodentate},
 pages = {11591-11600},
 volume = {119},
 websites = {https://pubs.acs.org/doi/10.1021/acs.jpca.5b09006},
 month = {11},
 publisher = {American Chemical Society},
 day = {25},
 id = {996d8225-8b63-3254-9171-fe7103c29e8e},
 created = {2019-10-01T17:20:23.450Z},
 accessed = {2019-09-19},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.614Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lopez-Berganza2015},
 private_publication = {false},
 abstract = {Ab initio simulations of large hydrated calcium carbonate clusters are challenging due to the existence of multiple local energy minima. Extensive conformational searches around hydrated calcium carbonate clusters (CaCO3·nH2O for n = 1-18) were performed to find low-energy hydration structures using an efficient combination of Monte Carlo searches, density-functional tight binding (DFTB+) method, and density-functional theory (DFT) at the B3LYP level, or Møller-Plesset perturbation theory at the MP2 level. This multilevel optimization yields several low-energy structures for hydrated calcium carbonate. Structural and energetics analysis of the hydration of these clusters revealed a first hydration shell composed of 12 water molecules. Bond-length and charge densities were also determined for different cluster sizes. The solvation of calcium carbonate in bulk water was investigated by placing the explicitly solvated CaCO3·nH2O clusters in a polarizable continuum model (PCM). The findings of this study provide new insights into the energetics and structure of hydrated calcium carbonate and contribute to the understanding of mechanisms where calcium carbonate formation or dissolution is of relevance. © 2015 American Chemical Society.},
 bibtype = {article},
 author = {Lopez-Berganza, Josue A. and Diao, Yijue and Pamidighantam, Sudhakar and Espinosa-Marzal, Rosa M.},
 doi = {10.1021/acs.jpca.5b09006},
 journal = {The Journal of Physical Chemistry A},
 number = {47}
}

Ab initio simulations of large hydrated calcium carbonate clusters are challenging due to the existence of multiple local energy minima. Extensive conformational searches around hydrated calcium carbonate clusters (CaCO3·nH2O for n = 1-18) were performed to find low-energy hydration structures using an efficient combination of Monte Carlo searches, density-functional tight binding (DFTB+) method, and density-functional theory (DFT) at the B3LYP level, or Møller-Plesset perturbation theory at the MP2 level. This multilevel optimization yields several low-energy structures for hydrated calcium carbonate. Structural and energetics analysis of the hydration of these clusters revealed a first hydration shell composed of 12 water molecules. Bond-length and charge densities were also determined for different cluster sizes. The solvation of calcium carbonate in bulk water was investigated by placing the explicitly solvated CaCO3·nH2O clusters in a polarizable continuum model (PCM). The findings of this study provide new insights into the energetics and structure of hydrated calcium carbonate and contribute to the understanding of mechanisms where calcium carbonate formation or dissolution is of relevance. © 2015 American Chemical Society.

@inproceedings{
 title = {XSEDE value added, cost avoidance, and return on investment},
 type = {inproceedings},
 year = {2015},
 keywords = {Advanced cyberinfrastructure,Cost avoidance,ROI,Return on investment,Value added,XSEDE},
 pages = {1-8},
 volume = {2015-July},
 websites = {http://dl.acm.org/citation.cfm?doid=2792745.2792768},
 month = {7},
 publisher = {ACM Press},
 day = {26},
 city = {New York, New York, USA},
 id = {0d264dc2-ac44-31e4-a624-eccba5b01579},
 created = {2019-10-01T17:20:27.250Z},
 accessed = {2019-08-29},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T18:58:47.252Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015a},
 private_publication = {false},
 abstract = {© 2015 Copyright is held by the owner/author(s). It is difficult for large research facilities to quantify a return on the investments that fund their operations. This is because there can be a time lag of years or decades between an innovation or discovery and the realization of its value through practical application. This report presents a three-part methodology that attempts to assess the value of federal investment in XSEDE: 1) a qualitative examination of the areas where XSEDE adds value to the activities of the open research community, 2) a "thought model" examining the cost avoidance realized by the National Science Foundation (NSF) through the centralization and coordination XSEDE provides, and 3) an assessment of the value XSEDE provides to Service Providers in the XD ecosystem. XSEDE adds significantly to the US research community because it functions as a unified interface to the XD ecosystem and because of its scale. A partly quantitative, partly qualitative analysis suggests the Return on Investment of NSF spending on XSEDE is greater than 1.0, indicating that the aggregate value received by the nation from XSEDE is greater than the cost of direct federal investment in XSEDE.},
 bibtype = {inproceedings},
 author = {Stewart, Craig A. and Roskies, Ralph and Knepper, Richard and Moore, Richard L. and Whitt, Justin and Cockerill, Timothy M.},
 doi = {10.1145/2792745.2792768},
 booktitle = {Proceedings of the 2015 XSEDE Conference on Scientific Advancements Enabled by Enhanced Cyberinfrastructure - XSEDE '15}
}

© 2015 Copyright is held by the owner/author(s). It is difficult for large research facilities to quantify a return on the investments that fund their operations. This is because there can be a time lag of years or decades between an innovation or discovery and the realization of its value through practical application. This report presents a three-part methodology that attempts to assess the value of federal investment in XSEDE: 1) a qualitative examination of the areas where XSEDE adds value to the activities of the open research community, 2) a "thought model" examining the cost avoidance realized by the National Science Foundation (NSF) through the centralization and coordination XSEDE provides, and 3) an assessment of the value XSEDE provides to Service Providers in the XD ecosystem. XSEDE adds significantly to the US research community because it functions as a unified interface to the XD ecosystem and because of its scale. A partly quantitative, partly qualitative analysis suggests the Return on Investment of NSF spending on XSEDE is greater than 1.0, indicating that the aggregate value received by the nation from XSEDE is greater than the cost of direct federal investment in XSEDE.

@article{
 title = {Saccharide breakdown and fermentation by the honey bee gut microbiome},
 type = {article},
 year = {2015},
 pages = {796-815},
 volume = {17},
 month = {3},
 publisher = {Blackwell Publishing Ltd},
 day = {1},
 id = {f1747001-05cf-3eae-853a-160d398a575a},
 created = {2019-10-01T17:20:31.410Z},
 accessed = {2019-08-27},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.836Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Lee2015},
 private_publication = {false},
 abstract = {The honey bee, the world's most important agricultural pollinator, relies exclusively on plant-derived foods for nutrition. Nectar and pollen collected by honey bees are processed and matured within the nest through the activities of honey bee-derived microbes and enzymes. In order to better understand the contribution of the microbial community to food processing in the honey bee, we generated a metatranscriptome of the honey bee gut microbiome. The function of the microbial community in the honey bee, as revealed by metatranscriptome sequencing, resembles that of other animal guts and food-processing environments. We identified three major bacterial classes that are active in the gut (γ-Proteobacteria, Bacilli and Actinobacteria), all of which are predicted to participate in the breakdown of complex macromolecules (e.g. polysaccharides and polypeptides), the fermentation of component parts of these macromolecules, and the generation of various fermentation products, such as short-chain fatty acids and alcohol. The ability of the microbial community to metabolize these carbon-rich food sources was confirmed through the use of community-level physiological profiling. Collectively, these findings suggest that the gut microflora of the honey bee harbours bacterial members with unique roles, which ultimately can contribute to the processing of plant-derived food for colonies.},
 bibtype = {article},
 author = {Lee, Fredrick J. and Rusch, Douglas B. and Stewart, Frank J. and Mattila, Heather R. and Newton, Irene L.G.},
 doi = {10.1111/1462-2920.12526},
 journal = {Environmental Microbiology},
 number = {3}
}

The honey bee, the world's most important agricultural pollinator, relies exclusively on plant-derived foods for nutrition. Nectar and pollen collected by honey bees are processed and matured within the nest through the activities of honey bee-derived microbes and enzymes. In order to better understand the contribution of the microbial community to food processing in the honey bee, we generated a metatranscriptome of the honey bee gut microbiome. The function of the microbial community in the honey bee, as revealed by metatranscriptome sequencing, resembles that of other animal guts and food-processing environments. We identified three major bacterial classes that are active in the gut (γ-Proteobacteria, Bacilli and Actinobacteria), all of which are predicted to participate in the breakdown of complex macromolecules (e.g. polysaccharides and polypeptides), the fermentation of component parts of these macromolecules, and the generation of various fermentation products, such as short-chain fatty acids and alcohol. The ability of the microbial community to metabolize these carbon-rich food sources was confirmed through the use of community-level physiological profiling. Collectively, these findings suggest that the gut microflora of the honey bee harbours bacterial members with unique roles, which ultimately can contribute to the processing of plant-derived food for colonies.

@article{
 title = {Subtractive assembly for comparative metagenomics, and its application to type 2 diabetes metagenomes},
 type = {article},
 year = {2015},
 keywords = {Diabetes Mellitus,Gastrointestinal Tract,Humans,Metagenom,Type 2,gastrointestinal tract,genetics,human,metagenom},
 volume = {16},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84955647881&doi=10.1186%2Fs13059-015-0804-0&partnerID=40&md5=9dff4ceb8c224cd11f569d2f3102ccf2},
 publisher = {BioMed Central Ltd.},
 id = {538fe930-d970-38f1-b337-9e22c5cf6f23},
 created = {2019-10-01T17:20:32.780Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:00.372Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wang2015},
 source_type = {article},
 notes = {cited By 2},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {Comparative metagenomics remains challenging due to the size and complexity of metagenomic datasets. Here we introduce subtractive assembly, a de novo assembly approach for comparative metagenomics that directly assembles only the differential reads that distinguish between two groups of metagenomes. Using simulated datasets, we show it improves both the efficiency of the assembly and the assembly quality of the differential genomes and genes. Further, its application to type 2 diabetes (T2D) metagenomic datasets reveals clear signatures of the T2D gut microbiome, revealing new phylogenetic and functional features of the gut microbial communities associated with T2D. © 2015 Wang et al.},
 bibtype = {article},
 author = {Wang, M and Doak, T G and Ye, Y},
 doi = {10.1186/s13059-015-0804-0},
 journal = {Genome Biology},
 number = {1}
}

Comparative metagenomics remains challenging due to the size and complexity of metagenomic datasets. Here we introduce subtractive assembly, a de novo assembly approach for comparative metagenomics that directly assembles only the differential reads that distinguish between two groups of metagenomes. Using simulated datasets, we show it improves both the efficiency of the assembly and the assembly quality of the differential genomes and genes. Further, its application to type 2 diabetes (T2D) metagenomic datasets reveals clear signatures of the T2D gut microbiome, revealing new phylogenetic and functional features of the gut microbial communities associated with T2D. © 2015 Wang et al.

@article{
 title = {Strand-specific community RNA-seq reveals prevalent and dynamic antisense transcription in human gut microbiota},
 type = {article},
 year = {2015},
 keywords = {Article,antisense transcription,bacterial geneti},
 volume = {6},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84946746270&doi=10.3389%2Ffmicb.2015.00896&partnerID=40&md5=9bf77043bc0a78dec2df20aed340acc6,https://www.frontiersin.org/articles/10.3389/fmicb.2015.00896/full},
 publisher = {Frontiers Research Foundation},
 id = {105b2574-a099-3d15-9b7a-072cefad438b},
 created = {2019-10-01T17:20:33.215Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:56.039Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Bao2015},
 source_type = {article},
 notes = {cited By 4},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {Metagenomics and other meta-omics approaches (including metatranscriptomics) provide insights into the composition and function of microbial communities living in different environments or animal hosts. Metatranscriptomics research provides an unprecedented opportunity to examine gene regulation for many microbial species simultaneously, and more importantly, for the majority that are unculturable microbial species, in their natural environments (or hosts). Current analyses of metatranscriptomic datasets focus on the detection of gene expression levels and the study of the relationship between changes of gene expression and changes of environment. As a demonstration of utilizing metatranscriptomics beyond these common analyses, we developed a computational and statistical procedure to analyze the antisense transcripts in strand-specific metatranscriptomic datasets. Antisense RNAs encoded on the DNA strand opposite a gene's CDS have the potential to form extensive base-pairing interactions with the corresponding sense RNA, and can have important regulatory functions. Most studies of antisense RNAs in bacteria are rather recent, are mostly based on transcriptome analysis, and have been applied mainly to single bacterial species. Application of our approaches to human gut-associated metatranscriptomic datasets allowed us to survey antisense transcription for a large number of bacterial species associated with human beings. The ratio of protein coding genes with antisense transcription ranges from 0 to 35.8% (median = 10.0%) among 47 species. Our results show that antisense transcription is dynamic, varying between human individuals. Functional enrichment analysis revealed a preference of certain gene functions for antisense transcription, and transposase genes are among the most prominent ones (but we also observed antisense transcription in bacterial house-keeping genes). © 2015 Bao, Wang, Doak and Ye.},
 bibtype = {article},
 author = {Bao, G and Wang, M and Doak, T G and Ye, Y},
 doi = {10.3389/fmicb.2015.00896},
 journal = {Frontiers in Microbiology},
 number = {SEP}
}

Metagenomics and other meta-omics approaches (including metatranscriptomics) provide insights into the composition and function of microbial communities living in different environments or animal hosts. Metatranscriptomics research provides an unprecedented opportunity to examine gene regulation for many microbial species simultaneously, and more importantly, for the majority that are unculturable microbial species, in their natural environments (or hosts). Current analyses of metatranscriptomic datasets focus on the detection of gene expression levels and the study of the relationship between changes of gene expression and changes of environment. As a demonstration of utilizing metatranscriptomics beyond these common analyses, we developed a computational and statistical procedure to analyze the antisense transcripts in strand-specific metatranscriptomic datasets. Antisense RNAs encoded on the DNA strand opposite a gene's CDS have the potential to form extensive base-pairing interactions with the corresponding sense RNA, and can have important regulatory functions. Most studies of antisense RNAs in bacteria are rather recent, are mostly based on transcriptome analysis, and have been applied mainly to single bacterial species. Application of our approaches to human gut-associated metatranscriptomic datasets allowed us to survey antisense transcription for a large number of bacterial species associated with human beings. The ratio of protein coding genes with antisense transcription ranges from 0 to 35.8% (median = 10.0%) among 47 species. Our results show that antisense transcription is dynamic, varying between human individuals. Functional enrichment analysis revealed a preference of certain gene functions for antisense transcription, and transposase genes are among the most prominent ones (but we also observed antisense transcription in bacterial house-keeping genes). © 2015 Bao, Wang, Doak and Ye.

@article{
 title = {Draft genome sequence of Caedibacter varicaedens, a kappa killer endosymbiont bacterium of the ciliate Paramecium biaurelia},
 type = {article},
 year = {2015},
 volume = {3},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85009471729&doi=10.1128%2FgenomeA.01310-15&partnerID=40&md5=f7f264dbe6b611814ea0d1ba9b3423e5},
 publisher = {American Society for Microbiology},
 id = {144622e8-e8d9-3c69-9edb-2bc903d4b1a6},
 created = {2019-10-01T17:20:33.243Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:56.331Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Suzuki2015},
 source_type = {article},
 notes = {cited By 0},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {Caedibacter varicaedens is a kappa killer endosymbiont bacterium of the ciliate Paramecium biaurelia. Here, we present the draft genome sequence of C. varicaedens. © 2015 Suzuki et al.},
 bibtype = {article},
 author = {Suzuki, H and Dapper, A L and Jackson, C E and Lee, H and Pejaver, V and Doak, T G and Lynch, M and Preer J.R., Jr.},
 doi = {10.1128/genomeA.01310-15},
 journal = {Genome Announcements},
 number = {6}
}

Caedibacter varicaedens is a kappa killer endosymbiont bacterium of the ciliate Paramecium biaurelia. Here, we present the draft genome sequence of C. varicaedens. © 2015 Suzuki et al.

@article{
 title = {Mutation rate, spectrum, topology and context-dependency in the DNA mismatch repair (MMR) deficient Pseudomonas fluorescens Migula ATCC948},
 type = {article},
 year = {2015},
 pages = {262-271},
 volume = {23},
 websites = {http://gbe.oxfordjournals.org/content/7/1/262.full.pdf+html?sid=e5570fd3-bf72-4c6c-a744-975adcdccdda},
 id = {47e24f67-447f-3617-bd25-0773d329ce11},
 created = {2019-10-01T17:20:33.796Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:03.508Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {lsmadl15},
 source_type = {article},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 bibtype = {article},
 author = {Long, H and Sung, W and Miller, S F and Ackerman, M S and Doak, T G and Lynch, M},
 journal = {GENOME BIOLOGY AND EVOLUTION}
}

@article{
 title = {Background mutational features of the radiation-resistant bacterium deinococcus radiodurans},
 type = {article},
 year = {2015},
 keywords = {Article,Bacterial,Bacterial Proteins,DNA,DNA Damage,Deinococcus,Deinococcus radioduran,Genes,Genetic Drift,Insertional,Mutagenesis,Mutation Rate,Plasmids,Point Mutat,adenine,ba,bacterial strain,cytosine,uracil DNA glycosidase},
 pages = {2383-2392},
 volume = {32},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84943608703&doi=10.1093%2Fmolbev%2Fmsv119&partnerID=40&md5=43f71482fe288975f38f8e279f12baa6},
 publisher = {Oxford University Press},
 id = {fe53c531-a7c4-36a0-9157-6c4ccaf2e7a4},
 created = {2019-10-01T17:20:33.799Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:00.968Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Long20152383},
 source_type = {article},
 notes = {cited By 14},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {Deinococcus bacteria are extremely resistant to radiation, oxidation, and desiccation. Resilience to these factors has been suggested to be due to enhanced damage prevention and repair mechanisms, as well as highly efficient antioxidant protection systems. Here, using mutation-accumulation experiments, we find that the GC-rich Deinococcus radiodurans has an overall background genomicmutation rate similar to that of E. coli, but differs inmutation spectrum, with the A/T to G/C mutation rate (based on a total count of 88 A:T→G:C transitions and 82 A:T→C:G transversions) per site per generation higher than that in the other direction (based on a total count of 157 G:C→A:T transitions and 33 G:C→T:A transversions).We propose that this unique spectrumis shaped mainly by the abundant uracil DNA glycosylases reducing G:C→A:T transitions, adenine methylation elevating A:T→C:G transversions, and absence of cytosine methylation decreasing G:C→A:T transitions. As opposed to the greater than 100 elevation of the mutation rate in MMR (DNA Mismatch Repair deficient) strains of most other organisms,MMR D. radiodurans only exhibits a 4-fold elevation, raising the possibility that other DNA repair mechanisms compensate for a relatively low-efficiency DNA MMR pathway. As D. radiodurans has plentiful insertion sequence (IS) elements in the genome and the activities of IS elements are rarely directly explored, we also estimated the insertion (transposition) rate of the IS elements to be 2.50×103 per genome per generation in the wild-type strain; knocking out MMR did not elevate the IS element insertion rate in this organism. © 2015 The Author.},
 bibtype = {article},
 author = {Long, H and Kucukyildirim, S and Sung, W and Williams, E and Lee, H and Ackerman, M S and Doak, T G and Tang, H and Lynch, M},
 doi = {10.1093/molbev/msv119},
 journal = {Molecular Biology and Evolution},
 number = {9}
}

Deinococcus bacteria are extremely resistant to radiation, oxidation, and desiccation. Resilience to these factors has been suggested to be due to enhanced damage prevention and repair mechanisms, as well as highly efficient antioxidant protection systems. Here, using mutation-accumulation experiments, we find that the GC-rich Deinococcus radiodurans has an overall background genomicmutation rate similar to that of E. coli, but differs inmutation spectrum, with the A/T to G/C mutation rate (based on a total count of 88 A:T→G:C transitions and 82 A:T→C:G transversions) per site per generation higher than that in the other direction (based on a total count of 157 G:C→A:T transitions and 33 G:C→T:A transversions).We propose that this unique spectrumis shaped mainly by the abundant uracil DNA glycosylases reducing G:C→A:T transitions, adenine methylation elevating A:T→C:G transversions, and absence of cytosine methylation decreasing G:C→A:T transitions. As opposed to the greater than 100 elevation of the mutation rate in MMR (DNA Mismatch Repair deficient) strains of most other organisms,MMR D. radiodurans only exhibits a 4-fold elevation, raising the possibility that other DNA repair mechanisms compensate for a relatively low-efficiency DNA MMR pathway. As D. radiodurans has plentiful insertion sequence (IS) elements in the genome and the activities of IS elements are rarely directly explored, we also estimated the insertion (transposition) rate of the IS elements to be 2.50×103 per genome per generation in the wild-type strain; knocking out MMR did not elevate the IS element insertion rate in this organism. © 2015 The Author.

@article{
 title = {The spontaneous mutation rate in the fission yeast Schizosaccharomyces pombe},
 type = {article},
 year = {2015},
 keywords = {5 methylcytosine,Amino Acid Sequence,Article,DNA,DNA methyltransferase,Fungal,Gene Expression Regulat,Haploidy,INDEL Mutation,Mutat,Mutation,allele,budding,cell adhesion,cell div,cytosine},
 pages = {737-744},
 volume = {201},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84941668361&doi=10.1534%2Fgenetics.115.177329&partnerID=40&md5=046524c96924e58d08868a2ec9930ce8},
 publisher = {Genetics},
 id = {608c36af-5061-363b-83cf-493fccccd234},
 created = {2019-10-01T17:20:34.617Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:54.552Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Farlow2015737},
 source_type = {article},
 notes = {cited By 18},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {The rate at which new mutations arise in the genome is a key factor in the evolution and adaptation of species. Here we describe the rate and spectrum of spontaneous mutations for the fission yeast Schizosaccharomyces pombe, a key model organism with many similarities to higher eukaryotes. We undertook an 1700-generation mutation accumulation (MA) experiment with a haploid S. pombe, generating 422 single-base substitutions and 119 insertion-deletion mutations (indels) across the 96 replicates. This equates to a base-substitution mutation rate of 2.00 3 10210 mutations per site per generation, similar to that reported for the distantly related budding yeast Saccharomyces cerevisiae. However, these two yeast species differ dramatically in their spectrum of base substitutions, the types of indels (S. pombe is more prone to insertions), and the pattern of selection required to counteract a strong AT-biased mutation rate. Overall, our results indicate that GC-biased gene conversion does not play a major role in shaping the nucleotide composition of the S. pombe genome and suggest that the mechanisms of DNA maintenance may have diverged significantly between fission and budding yeasts. Unexpectedly, CpG sites appear to be excessively liable to mutation in both species despite the likely absence of DNA methylation. © 2015 by the Genetics Society of America.},
 bibtype = {article},
 author = {Farlow, A and Long, H and Arnoux, S and Sung, W and Doak, T G and Nordborg, M and Lynch, M},
 doi = {10.1534/genetics.115.177329},
 journal = {Genetics},
 number = {2}
}

The rate at which new mutations arise in the genome is a key factor in the evolution and adaptation of species. Here we describe the rate and spectrum of spontaneous mutations for the fission yeast Schizosaccharomyces pombe, a key model organism with many similarities to higher eukaryotes. We undertook an 1700-generation mutation accumulation (MA) experiment with a haploid S. pombe, generating 422 single-base substitutions and 119 insertion-deletion mutations (indels) across the 96 replicates. This equates to a base-substitution mutation rate of 2.00 3 10210 mutations per site per generation, similar to that reported for the distantly related budding yeast Saccharomyces cerevisiae. However, these two yeast species differ dramatically in their spectrum of base substitutions, the types of indels (S. pombe is more prone to insertions), and the pattern of selection required to counteract a strong AT-biased mutation rate. Overall, our results indicate that GC-biased gene conversion does not play a major role in shaping the nucleotide composition of the S. pombe genome and suggest that the mechanisms of DNA maintenance may have diverged significantly between fission and budding yeasts. Unexpectedly, CpG sites appear to be excessively liable to mutation in both species despite the likely absence of DNA methylation. © 2015 by the Genetics Society of America.

@article{
 title = {Describing and predicting developmental profiles of externalizing problems from childhood to adulthood},
 type = {article},
 year = {2015},
 id = {0ee91a63-64fb-379b-abea-fb57d641bdc0},
 created = {2019-10-01T17:20:34.809Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:53.043Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {pbdlp15},
 source_type = {article},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 bibtype = {article},
 author = {Petersen, I T and Bates, J E and Dodge, K A and Lansford, J E and Pettit, G S},
 doi = {10.1017/S0954579414000789.},
 journal = {DEVELOPMENT AND PSYCHOPATHOLOGY}
}

@inproceedings{
 title = {NCBI-BLAST programs optimization on XSEDE resources for sustainable aquaculture},
 type = {inproceedings},
 year = {2015},
 pages = {5},
 id = {10f3c9d6-9fa9-329e-a3b4-2f721ffad56b},
 created = {2019-10-01T17:20:34.992Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:51.580Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {sgphbs15},
 source_type = {inproceedings},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {The development of genomic resources of non-model organisms is now becoming commonplace as the cost of sequencing continues to decrease. The Genome Informatics Facility in collaboration with the Southwest Fisheries Science Center (SWFSC), NOAA is creating these resources for sustainable aquaculture in Seriola lalandi. Gene prediction and annotation are common steps in the pipeline to generate genomic resources, which are computationally intense and time consuming. In our steps to create genomic resources for Seriola lalandi, we found BLAST to be one of our most rate limiting steps. Therefore, we took advantage of our XSEDE Extended Collaborative Support Services (ECSS) to reduce the amount of time required to process our transcriptome data by 300 percent. In this paper, we describe an optimized method for the BLAST tool on the Stampede cluster, which works with any existing datasets or database, without any modification. At modest core counts, our results are similar to the MPI-enabled BLAST algorithm (mpiBLAST), but also allow the much needed and improved flexibility of output formats that the latest versions of BLAST provide. Reducing this time-consuming bottleneck in BLAST will be broadly applicable to the annotation of large sequencing datasets for any organism.},
 bibtype = {inproceedings},
 author = {Seetharam, Arun and Gomez, Antonio and Purcell, Catherine M and Hyde, John R and Blood, Philip D and Severin, Andrew J},
 doi = {http://dx.doi.org/10.1145/2792745.2792749},
 booktitle = {Proceedings of the 2015 XSEDE Conference: Scientific Advancements Enabled by Enhanced Cyberinfrastructure (XSEDE '15)}
}

The development of genomic resources of non-model organisms is now becoming commonplace as the cost of sequencing continues to decrease. The Genome Informatics Facility in collaboration with the Southwest Fisheries Science Center (SWFSC), NOAA is creating these resources for sustainable aquaculture in Seriola lalandi. Gene prediction and annotation are common steps in the pipeline to generate genomic resources, which are computationally intense and time consuming. In our steps to create genomic resources for Seriola lalandi, we found BLAST to be one of our most rate limiting steps. Therefore, we took advantage of our XSEDE Extended Collaborative Support Services (ECSS) to reduce the amount of time required to process our transcriptome data by 300 percent. In this paper, we describe an optimized method for the BLAST tool on the Stampede cluster, which works with any existing datasets or database, without any modification. At modest core counts, our results are similar to the MPI-enabled BLAST algorithm (mpiBLAST), but also allow the much needed and improved flexibility of output formats that the latest versions of BLAST provide. Reducing this time-consuming bottleneck in BLAST will be broadly applicable to the annotation of large sequencing datasets for any organism.

@inproceedings{
 title = {Cyberinfrastructure resources enabling creation of the loblolly pine reference transcriptome},
 type = {inproceedings},
 year = {2015},
 pages = {1-6},
 volume = {2015-July},
 websites = {http://hdl.handle.net/2022/20488,http://dl.acm.org/citation.cfm?doid=2792745.2792748},
 month = {7},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {8a2daf26-2e3e-323b-a0cf-8f045aa304af},
 created = {2019-10-01T17:20:35.154Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:35.154Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wu2015},
 source_type = {inproceedings},
 private_publication = {false},
 abstract = {Today's genomics technologies generate more sequence data than ever before possible, and at substantially lower costs, serving researchers across biological disciplines in transformative ways. Building transcriptome assemblies from RNA sequencing reads is one application of next-generation sequencing (NGS) that has held a central role in biological discovery in both model and non-model organisms, with and without whole genome sequence references. A major limitation in effective building of transcriptome references is no longer the sequencing data generation itself, but the computing infrastructure and expertise needed to assemble, analyze and manage the data. Here we describe a currently available resource dedicated to achieving such goals, and its use for extensive RNA assembly of up to 1.3 billion reads representing the massive transcriptome of loblolly pine, using four major assembly software installations. The Mason cluster, an XSEDE second tier resource at Indiana University, provides the necessary fast CPU cycles, large memory, and high I/O throughput for conducting large-scale genomics research. The National Center for Genome Analysis Support, or NCGAS, provides technical support in using HPC systems, bioinformatic support for determining the appropriate method to analyze a given dataset, and practical assistance in running computations. We demonstrate that a sufficient supercomputing resource and good workflow design are elements that are essential to large eukaryotic genomics and transcriptomics projects such as the complex transcriptome of loblolly pine, gene expression data that inform annotation and functional interpretation of the largest genome sequence reference to date.},
 bibtype = {inproceedings},
 author = {Wu, Le-Shin and Ganote, Carrie L. and Doak, Thomas G and Barnett, William and Mockaitis, Keithanne and Stewart, Craig A},
 doi = {10.1145/2792745.2792748},
 booktitle = {Proceedings of the 2015 XSEDE Conference on Scientific Advancements Enabled by Enhanced Cyberinfrastructure - XSEDE '15}
}

Today's genomics technologies generate more sequence data than ever before possible, and at substantially lower costs, serving researchers across biological disciplines in transformative ways. Building transcriptome assemblies from RNA sequencing reads is one application of next-generation sequencing (NGS) that has held a central role in biological discovery in both model and non-model organisms, with and without whole genome sequence references. A major limitation in effective building of transcriptome references is no longer the sequencing data generation itself, but the computing infrastructure and expertise needed to assemble, analyze and manage the data. Here we describe a currently available resource dedicated to achieving such goals, and its use for extensive RNA assembly of up to 1.3 billion reads representing the massive transcriptome of loblolly pine, using four major assembly software installations. The Mason cluster, an XSEDE second tier resource at Indiana University, provides the necessary fast CPU cycles, large memory, and high I/O throughput for conducting large-scale genomics research. The National Center for Genome Analysis Support, or NCGAS, provides technical support in using HPC systems, bioinformatic support for determining the appropriate method to analyze a given dataset, and practical assistance in running computations. We demonstrate that a sufficient supercomputing resource and good workflow design are elements that are essential to large eukaryotic genomics and transcriptomics projects such as the complex transcriptome of loblolly pine, gene expression data that inform annotation and functional interpretation of the largest genome sequence reference to date.

@inproceedings{
 title = {Research proposal: Barriers to new user and new domain adoption of the XSEDE cyberinfrastructure},
 type = {inproceedings},
 year = {2015},
 keywords = {Barriers to adoption,Computational resources,Cy,Engineering research,Information systems},
 pages = {1023-1028},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84958093783&doi=10.15439%2F2015F244&partnerID=40&md5=4978f6decd6b66dfc9e05cb107ba922a},
 publisher = {Institute of Electrical and Electronics Engineers Inc.},
 id = {eebf7d87-0919-3a56-af7d-7ddd1de1f559},
 created = {2019-10-01T17:20:35.407Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:30.314Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper20151023},
 source_type = {conference},
 notes = {cited By 0; Conference of Federated Conference on Computer Science and Information Systems, FedCSIS 2015 ; Conference Date: 13 September 2015 Through 16 September 2015; Conference Code:117625},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {This research proposal proposes the examination of user attitudes about the Extreme Science and Engineering Discovery Environment (XSEDE). The XSEDE project supports basic research with a common system for making use of national cyberinfrastructure. The systems and infrastructure that make XSEDE useful for researchers are part of an actor network: these systems are socially constructed and they play their own part in the work of XSEDE, and in turn have an effect on the progress of basic research. I have completed previous work on the user relationships in the predecessor to XSEDE, the TeraGrid, and currently carry out participant observation with the management groups of the XSEDE project. By understanding the barriers to adoption of XSEDE by new researchers and new scientific domains, I hope to explore the linkage between resources (in this case computational resources) and scientific outputs. © 2015, IEEE.},
 bibtype = {inproceedings},
 author = {Knepper, R},
 editor = {Paprzycki M. Maciaszek L., Ganzha M Maciaszek L},
 doi = {10.15439/2015F244},
 booktitle = {Proceedings of the 2015 Federated Conference on Computer Science and Information Systems, FedCSIS 2015}
}

This research proposal proposes the examination of user attitudes about the Extreme Science and Engineering Discovery Environment (XSEDE). The XSEDE project supports basic research with a common system for making use of national cyberinfrastructure. The systems and infrastructure that make XSEDE useful for researchers are part of an actor network: these systems are socially constructed and they play their own part in the work of XSEDE, and in turn have an effect on the progress of basic research. I have completed previous work on the user relationships in the predecessor to XSEDE, the TeraGrid, and currently carry out participant observation with the management groups of the XSEDE project. By understanding the barriers to adoption of XSEDE by new researchers and new scientific domains, I hope to explore the linkage between resources (in this case computational resources) and scientific outputs. © 2015, IEEE.

@article{
 title = {Performance Optimization for the Trinity RNA-Seq Assembler},
 type = {article},
 year = {2015},
 websites = {http://hdl.handle.net/2022/20490},
 publisher = {Springer},
 id = {d1e69745-1422-3c38-9c38-74d3a75c815a},
 created = {2019-10-01T17:20:36.303Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:30.530Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wagner2015},
 source_type = {article},
 medium = {(to appear)},
 notes = {Article from presentation at 9th Parallel Tools Workshop, September 2-3, 2015 in Dresden, Germany},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Utilizing the enormous computing resources of high performance computing systems is anything but a trivial task. Performance analysis tools are designed to assist developers in this challenging task by helping to understand the application behavior and identify critical performance issues. In this paper we share our efforts and experiences in analyzing and optimizing Trinity, a well-established framework for the de novo reconstruction of transcriptomes from RNA-seq reads. Thereby, we try to reflect all aspects of the ongoing performance engineering: the identification of optimization targets, the code improvements resulting in 20% overall runtime reduction, as well as the challenges we encountered getting there.},
 bibtype = {article},
 author = {Wagner, M and Fulton, B and Henschel, R},
 journal = {Tools for High Performance Computing}
}

Utilizing the enormous computing resources of high performance computing systems is anything but a trivial task. Performance analysis tools are designed to assist developers in this challenging task by helping to understand the application behavior and identify critical performance issues. In this paper we share our efforts and experiences in analyzing and optimizing Trinity, a well-established framework for the de novo reconstruction of transcriptomes from RNA-seq reads. Thereby, we try to reflect all aspects of the ongoing performance engineering: the identification of optimization targets, the code improvements resulting in 20% overall runtime reduction, as well as the challenges we encountered getting there.

@techreport{
 title = {Use of IU advanced computational systems–parallelism, job mixes, and queue wait times},
 type = {techreport},
 year = {2015},
 id = {77ed0161-5a02-334b-ba4f-1d1c78173069},
 created = {2019-10-01T17:20:36.808Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:28.843Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Link2015a},
 source_type = {RPRT},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Link, Matthew R and Henschel, Robert and Hancock, David Y and Stewart, Craig A}
}

@techreport{
 title = {Results of 2013 Survey of Parallel Computing Needs Focusing on NSF-funded Researchers},
 type = {techreport},
 year = {2015},
 id = {e806c497-08f1-3460-92c3-5faaa6ae439b},
 created = {2019-10-01T17:20:36.835Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:28.250Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015f},
 source_type = {RPRT},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Arenson, Andrew and Fischer, Jeremy and Link, Matthew R and Michael, Scott A and Wernert, Julie A}
}

@article{
 title = {Big Data on Ice: The Forward Observer System for In-flight Synthetic Aperture Radar Processing},
 type = {article},
 year = {2015},
 pages = {1504-1513},
 volume = {51},
 websites = {http://hdl.handle.net/2022/20471,http://www.sciencedirect.com/science/article/pii/S1877050915011485},
 month = {11},
 publisher = {Elsevier},
 id = {82909760-1afb-3e70-9387-b38cad08e11b},
 created = {2019-10-01T17:20:37.135Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:23.184Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Knepper2015a},
 source_type = {article},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {We introduce the Forward Observer system, which is designed to provide data assurance in field data acquisition while receiving significant amounts (several terabytes per flight) of Synthetic Aperture Radar data during flights over the polar regions, which provide unique requirements for developing data collection and processing systems. Under polar conditions in the field and given the difficulty and expense of collecting data, data retention is absolutely critical. Our system provides a storage and analysis cluster with software that connects to field instruments via standard protocols, replicates data to multiple stores automatically as soon as it is written, and provides pre-processing of data so that initial visualizations are available immediately after collection, where they can provide feedback to researchers in the aircraft during the flight.},
 bibtype = {article},
 author = {Knepper, Richard and Link, Matthew R and Standish, Matthew},
 doi = {10.1016/J.PROCS.2015.05.340},
 journal = {Procedia Computer Science}
}

We introduce the Forward Observer system, which is designed to provide data assurance in field data acquisition while receiving significant amounts (several terabytes per flight) of Synthetic Aperture Radar data during flights over the polar regions, which provide unique requirements for developing data collection and processing systems. Under polar conditions in the field and given the difficulty and expense of collecting data, data retention is absolutely critical. Our system provides a storage and analysis cluster with software that connects to field instruments via standard protocols, replicates data to multiple stores automatically as soon as it is written, and provides pre-processing of data so that initial visualizations are available immediately after collection, where they can provide feedback to researchers in the aircraft during the flight.

@article{
 title = {Scalability testing of dne2 in lustre 2.7},
 type = {article},
 year = {2015},
 id = {e76932cd-c730-3c02-b49b-8b9634a18aee},
 created = {2019-10-01T17:20:37.876Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:14.519Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Crowe2015},
 source_type = {JOUR},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {article},
 author = {Crowe, Tom and Lavender, Nathan and Simms, Stephen},
 journal = {Lustre Users Group}
}

@techreport{
 title = {UITS Research Technologies update},
 type = {techreport},
 year = {2015},
 id = {5e6958a0-3fe7-3b9f-8fb7-45908d58745c},
 created = {2019-10-01T17:20:37.965Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:13.650Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015c},
 source_type = {JOUR},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A}
}

@techreport{
 title = {Workshop Report: Campus Bridging: Reducing Obstacles on the Path to Big Answers 2015},
 type = {techreport},
 year = {2015},
 websites = {http://hdl.handle.net/2022/20538},
 month = {9},
 city = {Bloomington, IN},
 institution = {Indiana University},
 id = {d23b8330-4e53-3ce6-9ddc-213ad1e7460f},
 created = {2019-10-01T17:20:38.621Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:29.656Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hallock2015},
 source_type = {techreport},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {For the researcher whose experiments require large-scale cyberinfrastructure, there exists significant challenges to successful completion. These challenges are broad and go far beyond the simple issue that there are not enough large-scale resources available; these solvable issues range from a lack of documentation written for a non-technical audience to a need for greater consistency with regard to system configuration and consistent software configuration and availability on the large-scale resources at national tier supercomputing centers, with a number of other challenges existing alongside the ones mentioned here. Campus Bridging is a relatively young discipline that aims to mitigate these issues for the academic end-user, for whom the entire process can feel like a path comprised entirely of obstacles. The solutions to these problems must by necessity include multiple approaches, with focus not only on the end user but on the system administrators responsible for supporting these resources as well as the systems themselves. These system resources include not only those at the supercomputing centers but also those that exist at the campus or departmental level and even on the personal computing devices the researcher uses to complete his or her work. This workshop report compiles the results of a half-day workshop, held in conjunction with IEEE Cluster 2015 in Chicago, IL.},
 bibtype = {techreport},
 author = {Hallock, Barbara and Knepper, Richard and Stewart, Craig A}
}

For the researcher whose experiments require large-scale cyberinfrastructure, there exists significant challenges to successful completion. These challenges are broad and go far beyond the simple issue that there are not enough large-scale resources available; these solvable issues range from a lack of documentation written for a non-technical audience to a need for greater consistency with regard to system configuration and consistent software configuration and availability on the large-scale resources at national tier supercomputing centers, with a number of other challenges existing alongside the ones mentioned here. Campus Bridging is a relatively young discipline that aims to mitigate these issues for the academic end-user, for whom the entire process can feel like a path comprised entirely of obstacles. The solutions to these problems must by necessity include multiple approaches, with focus not only on the end user but on the system administrators responsible for supporting these resources as well as the systems themselves. These system resources include not only those at the supercomputing centers but also those that exist at the campus or departmental level and even on the personal computing devices the researcher uses to complete his or her work. This workshop report compiles the results of a half-day workshop, held in conjunction with IEEE Cluster 2015 in Chicago, IL.

@techreport{
 title = {Acceptance Test for Jetstream Test Cluster — Jetstream-Arizona (JA) Dell PowerEdge Test and Development Cluster},
 type = {techreport},
 year = {2015},
 websites = {http://hdl.handle.net/2022/20355},
 month = {8},
 institution = {Indiana University},
 id = {2de09be5-9368-3f67-8444-88efac920db4},
 created = {2019-10-01T17:20:39.415Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:29.689Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hancock2015a},
 source_type = {techreport},
 notes = {PTI Technical Report - PTI-TR15-007},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {This paper details the system description and the performance targets, methods used to perform the acceptance tests, and the achieved performance of the Jetstream test cluster.},
 bibtype = {techreport},
 author = {Hancock, D Y and Link, M R and Stewart, C A and Turner, G W}
}

This paper details the system description and the performance targets, methods used to perform the acceptance tests, and the achieved performance of the Jetstream test cluster.

@inproceedings{
 title = {Facilitating scientific collaborations by delegating identity management: Reducing barriers & roadmap for incremental implementation},
 type = {inproceedings},
 year = {2015},
 keywords = {Access control,Cyber security; Delegation; Identity; Identity ma,Risk management},
 pages = {15-19},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84979747691&doi=10.1145%2F2752499.2752501&partnerID=40&md5=83a416d313602ca3c75ba6528f549833},
 publisher = {Association for Computing Machinery, Inc},
 id = {ddb714f8-ab95-3f30-8caa-c89b6ffb09ea},
 created = {2019-10-01T17:20:41.858Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:20:41.858Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cowles201515},
 source_type = {conference},
 notes = {cited By 0; Conference of 2nd Workshop on Changing Landscapes in HPC Security, CLHS 2015 ; Conference Date: 16 June 2015; Conference Code:122511},
 private_publication = {false},
 abstract = {DOE Labs are often presented with conflicting requirements for providing services to scientific collaboratories. An identity management model involving transitive trust is increasingly common. We show how existing policies allow for increased delegation of identity management within an acceptable risk management framework. Specific topics addressed include deemed exports, DOE orders, Inertia and Risk, Traceability, and Technology Limitations. Real life examples of an incremental approach to implementing transitive trust are presented.},
 bibtype = {inproceedings},
 author = {Cowles, R and Jackson, C and Welch, V},
 doi = {10.1145/2752499.2752501},
 booktitle = {CLHS 2015 - Proceedings of the 2015 Workshop on Changing Landscapes in HPC Security, Part of HPDC 2015}
}

DOE Labs are often presented with conflicting requirements for providing services to scientific collaboratories. An identity management model involving transitive trust is increasingly common. We show how existing policies allow for increased delegation of identity management within an acceptable risk management framework. Specific topics addressed include deemed exports, DOE orders, Inertia and Risk, Traceability, and Technology Limitations. Real life examples of an incremental approach to implementing transitive trust are presented.

@book{
 title = {Provenance and Annotation of Data and Processes: 5th International Provenance and Annotation Workshop, IPAW 2014, Cologne, Germany, June 9-13, 2014. Revised Selected Papers},
 type = {book},
 year = {2015},
 volume = {8628},
 publisher = {Springer},
 id = {7ee568d7-909d-3e1c-9d92-a41727b55eb6},
 created = {2019-10-01T17:20:44.985Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:13.069Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ludascher2015},
 source_type = {BOOK},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {book},
 author = {Ludäscher, Bertram and Plale, Beth}
}

@inproceedings{
 title = {Big data provenance analysis and visualization},
 type = {inproceedings},
 year = {2015},
 id = {6c8441cc-01ee-3a04-aa5f-834f250bdbec},
 created = {2019-10-01T17:20:45.071Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:17.159Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. Provenance captured from E-Science experimentation is often large and complex, for instance, from agent-based simulations that have tens of thousands of heterogeneous components interacting over extended time periods. The subject of study of my dissertation is the use of E-Science provenance at scale. My initial research studied the visualization of large provenance graphs and proposed an abstract representation of provenance that supports useful data mining. Recent work involves analyzing large provenance data generated from agent-based simulations on a single machine. In continuation, I propose stream processing techniques to support the continuous and real-time analysis of data provenance, which is captured from agent based simulations on HPC and thus has unprecedented volume and complexity.},
 bibtype = {inproceedings},
 author = {Chen, P. and Plale, B.A.},
 doi = {10.1109/CCGrid.2015.85},
 booktitle = {Proceedings - 2015 IEEE/ACM 15th International Symposium on Cluster, Cloud, and Grid Computing, CCGrid 2015}
}

© 2015 IEEE. Provenance captured from E-Science experimentation is often large and complex, for instance, from agent-based simulations that have tens of thousands of heterogeneous components interacting over extended time periods. The subject of study of my dissertation is the use of E-Science provenance at scale. My initial research studied the visualization of large provenance graphs and proposed an abstract representation of provenance that supports useful data mining. Recent work involves analyzing large provenance data generated from agent-based simulations on a single machine. In continuation, I propose stream processing techniques to support the continuous and real-time analysis of data provenance, which is captured from agent based simulations on HPC and thus has unprecedented volume and complexity.

@book{
 title = {Regenerating and quantifying quality of benchmarking data using static and dynamic provenance},
 type = {book},
 year = {2015},
 source = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
 volume = {8628},
 id = {0af5de89-cee5-3322-a1b8-c1730cfa0f2b},
 created = {2019-10-01T17:20:45.216Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:16.856Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ghoshal2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© Springer International Publishing Switzerland 2015. Application benchmarks are critical to establishing the performance of a new system or library. But benchmarking a system can be tricky and reproducing a benchmark result even trickier. Provenance can help. Referencing benchmarks and their results on similar platforms for collective comparison and evaluation requires capturing provenance related to the process of benchmark execution, programs involved and results generated. In this paper we define a formal model of benchmark applications and required provenance, describe an implementation of the model that employs compile time (static) and runtime provenance capture, and quantify data quality in the context of benchmarks. Our results show that through a mix of compile time and runtime provenance capture, we can enable higher quality benchmark regeneration.},
 bibtype = {book},
 author = {Ghoshal, D. and Chauhan, A. and Plale, B.},
 doi = {10.1007/978-3-319-16462-55}
}

© Springer International Publishing Switzerland 2015. Application benchmarks are critical to establishing the performance of a new system or library. But benchmarking a system can be tricky and reproducing a benchmark result even trickier. Provenance can help. Referencing benchmarks and their results on similar platforms for collective comparison and evaluation requires capturing provenance related to the process of benchmark execution, programs involved and results generated. In this paper we define a formal model of benchmark applications and required provenance, describe an implementation of the model that employs compile time (static) and runtime provenance capture, and quantify data quality in the context of benchmarks. Our results show that through a mix of compile time and runtime provenance capture, we can enable higher quality benchmark regeneration.

@inproceedings{
 title = {Using phoebus data transfer accelerator in cloud environments},
 type = {inproceedings},
 year = {2015},
 volume = {2015-Septe},
 id = {f9c66682-0d2f-3e3e-a3f1-a53247da39a0},
 created = {2019-10-01T17:20:45.370Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:14.542Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2015a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. The quality of data exchange in cloud computing applications relies on the connection performance between user clients and their cloud storage providers, and is often dependent on the wide area network (WAN) properties among data centers. For certain classes of applications, it can be crucial to provide an end-to-end solution that accelerates large data transfers and improves overall user experience.},
 bibtype = {inproceedings},
 author = {Zhang, M. and Kissel, E. and Swany, M.},
 doi = {10.1109/ICC.2015.7248346},
 booktitle = {IEEE International Conference on Communications}
}

© 2015 IEEE. The quality of data exchange in cloud computing applications relies on the connection performance between user clients and their cloud storage providers, and is often dependent on the wide area network (WAN) properties among data centers. For certain classes of applications, it can be crucial to provide an end-to-end solution that accelerates large data transfers and improves overall user experience.

@inproceedings{
 title = {Workload-aware resource reservation for multi-tenant NoSQL},
 type = {inproceedings},
 year = {2015},
 volume = {2015-Octob},
 id = {90e35617-879e-37fc-9caf-5b99a4803a54},
 created = {2019-10-01T17:20:46.828Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:35.879Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zeng2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. Cloud hosted NoSQL data stores are for economic reasons often shared amongst multiple tenants simultaneously. The NoSQL provider consolidates multiple tenants access into a shared NoSQL instance and provides a dedicated view for each tenant. This multi-tenancy has tenants' data and workloads coexisting in the same node, which under certain conditions can lead to performance degradation of one tenant caused by another. In this paper, we investigate the multi-tenant interference in a common NoSQL store, HBase, and propose a resource reservation framework that reserves resources for prevention and dynamically adjusts the reservations according to tenant resource demands. The framework enforces cache reservation by splitting the cache space and disk reservation by scheduling requests to a distributed file system (DFS). A stochastic hill climbing algorithm is used to find a near-optimum plan for different resources reservations. Empirical results show that the framework can prevent interference and adapt to dynamic workloads under multi-tenancy.},
 bibtype = {inproceedings},
 author = {Zeng, J. and Plale, B.},
 doi = {10.1109/CLUSTER.2015.14},
 booktitle = {Proceedings - IEEE International Conference on Cluster Computing, ICCC}
}

© 2015 IEEE. Cloud hosted NoSQL data stores are for economic reasons often shared amongst multiple tenants simultaneously. The NoSQL provider consolidates multiple tenants access into a shared NoSQL instance and provides a dedicated view for each tenant. This multi-tenancy has tenants' data and workloads coexisting in the same node, which under certain conditions can lead to performance degradation of one tenant caused by another. In this paper, we investigate the multi-tenant interference in a common NoSQL store, HBase, and propose a resource reservation framework that reserves resources for prevention and dynamically adjusts the reservations according to tenant resource demands. The framework enforces cache reservation by splitting the cache space and disk reservation by scheduling requests to a distributed file system (DFS). A stochastic hill climbing algorithm is used to find a near-optimum plan for different resources reservations. Empirical results show that the framework can prevent interference and adapt to dynamic workloads under multi-tenancy.

@inproceedings{
 title = {Parallel and quantitative sequential pattern mining for large-scale interval-based temporal data},
 type = {inproceedings},
 year = {2015},
 id = {de801e1d-18aa-3f8a-b4bc-5bda75a5217f},
 created = {2019-10-01T17:20:46.860Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:35.232Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ruan2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2014 IEEE. Mining frequent subsequences of patterns, or sequential pattern mining, has wide application in customer shopping sequence analysis, web log stream analysis, multi-modal behavioral studies, to name a few. To detect unknown, anomalous, and unexpected patterns from large-scale interval-based temporal data without complete a priori knowledge is challenging. In this paper, we present a framework - PESMiner which allows parallel and quantitative mining of sequential patterns at scale. Whereas most existing sequential mining algorithms can only find sequential orders of temporal events, our work presents a novel interactive temporal data mining algorithm capable of extracting precise temporal properties of sequential patterns. Furthermore, our work provides a unified parallel solution that scales our algorithms to larger temporal data sets by exploiting iterative MapReduce tasks. Comprehensive performance evaluations demonstrate that PESMiner significantly outperforms existing interval-based mining algorithms in terms of both quality (i.e. accuracy, precision, and recall) and scalability.},
 bibtype = {inproceedings},
 author = {Ruan, G. and Zhang, H. and Plale, B.},
 doi = {10.1109/BigData.2014.7004410},
 booktitle = {Proceedings - 2014 IEEE International Conference on Big Data, IEEE Big Data 2014}
}

© 2014 IEEE. Mining frequent subsequences of patterns, or sequential pattern mining, has wide application in customer shopping sequence analysis, web log stream analysis, multi-modal behavioral studies, to name a few. To detect unknown, anomalous, and unexpected patterns from large-scale interval-based temporal data without complete a priori knowledge is challenging. In this paper, we present a framework - PESMiner which allows parallel and quantitative mining of sequential patterns at scale. Whereas most existing sequential mining algorithms can only find sequential orders of temporal events, our work presents a novel interactive temporal data mining algorithm capable of extracting precise temporal properties of sequential patterns. Furthermore, our work provides a unified parallel solution that scales our algorithms to larger temporal data sets by exploiting iterative MapReduce tasks. Comprehensive performance evaluations demonstrate that PESMiner significantly outperforms existing interval-based mining algorithms in terms of both quality (i.e. accuracy, precision, and recall) and scalability.

@inproceedings{
 title = {Adaptive Recursive Doubling Algorithm for Collective Communication},
 type = {inproceedings},
 year = {2015},
 id = {9a3bd88c-4f71-3a91-ada7-528341b1e2ec},
 created = {2019-10-01T17:20:47.036Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:33.202Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Arap2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. Process arrival times at MPI collective operations differ significantly. Addressing this fact with special handling for popular collective communication algorithms can yield performance improvements. The recursive doubling algorithm is one of the most efficient techniques for implementing collectives in MPI, especially for short messages and when the number of participating processes is a power of two. In the recursive doubling algorithm, all the processes must complete a given step before the algorithm continues to the next step. In this paper, we present a recursive doubling algorithm that makes use of available data and removes the requirement for each process to arrive at each step before proceeding. Our approach makes use of the multicast feature of the underlying network and progress tagging of messages, describing the currently available partial results. Our approach could be implemented in any parallel execution environment that supports multicasting. Our prototype implementation is based upon a network interface card with an FPGA, the Net FPGA. The Net FPGA provides hardware level programmability to offload processing, precise and controlled timing for accounting for packet and algorithm behavior, allowing classification of skew scenarios. Our algorithm provides up to 10% saving in synchronization delay in the presence of skew and up to 37% saving in number of messages generated, and up to 32% saving in reduction operations performed in MPI Allreduce.},
 bibtype = {inproceedings},
 author = {Arap, O. and Swany, M. and Brown, G. and Himebaugh, B.},
 doi = {10.1109/IPDPSW.2015.82},
 booktitle = {Proceedings - 2015 IEEE 29th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2015}
}

© 2015 IEEE. Process arrival times at MPI collective operations differ significantly. Addressing this fact with special handling for popular collective communication algorithms can yield performance improvements. The recursive doubling algorithm is one of the most efficient techniques for implementing collectives in MPI, especially for short messages and when the number of participating processes is a power of two. In the recursive doubling algorithm, all the processes must complete a given step before the algorithm continues to the next step. In this paper, we present a recursive doubling algorithm that makes use of available data and removes the requirement for each process to arrive at each step before proceeding. Our approach makes use of the multicast feature of the underlying network and progress tagging of messages, describing the currently available partial results. Our approach could be implemented in any parallel execution environment that supports multicasting. Our prototype implementation is based upon a network interface card with an FPGA, the Net FPGA. The Net FPGA provides hardware level programmability to offload processing, precise and controlled timing for accounting for packet and algorithm behavior, allowing classification of skew scenarios. Our algorithm provides up to 10% saving in synchronization delay in the presence of skew and up to 37% saving in number of messages generated, and up to 32% saving in reduction operations performed in MPI Allreduce.

@inproceedings{
 title = {Towards building a lightweight key-value store on parallel file system},
 type = {inproceedings},
 year = {2015},
 volume = {2015-Octob},
 id = {c3d7c6b0-c6bf-3d01-9e3c-46567ffbb1ad},
 created = {2019-10-01T17:20:47.601Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:38.372Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zeng2015a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. As data grows in number and size, big data applications begin to revolutionize the underlying storage system. On one hand, key-value store has prevailed as the back-end storage for big data applications owning to its schema-less data model, high scalability, and etc. On the other hand, parallel file system shared by multiple nodes offers large-capacity, high-throughput, as well as high-bandwidth access and is used widely in high performance computing (HPC) and cloud computing environments. In this paper, we explore the opportunity of building a lightweight key-value store that supports concurrent access over a parallel file system. The key-value store proposed relies on the sharing nature of parallel file system to provide distributed access. Instead of organizing a cluster of nodes with long running services to delegate the access, our key-value store simply embeds itself into applications and requires no long running services neither communication between nodes. Such a design not only simplifies the structure of a distributed key-value store but also avoids overhead introduced by having running services around the file system. We implemented a prototype of this system and compared it against Cassandra, a state-of-art key-value store. Preliminary results are promising.},
 bibtype = {inproceedings},
 author = {Zeng, J. and Plale, B.},
 doi = {10.1109/CLUSTER.2015.100},
 booktitle = {Proceedings - IEEE International Conference on Cluster Computing, ICCC}
}

© 2015 IEEE. As data grows in number and size, big data applications begin to revolutionize the underlying storage system. On one hand, key-value store has prevailed as the back-end storage for big data applications owning to its schema-less data model, high scalability, and etc. On the other hand, parallel file system shared by multiple nodes offers large-capacity, high-throughput, as well as high-bandwidth access and is used widely in high performance computing (HPC) and cloud computing environments. In this paper, we explore the opportunity of building a lightweight key-value store that supports concurrent access over a parallel file system. The key-value store proposed relies on the sharing nature of parallel file system to provide distributed access. Instead of organizing a cluster of nodes with long running services to delegate the access, our key-value store simply embeds itself into applications and requires no long running services neither communication between nodes. Such a design not only simplifies the structure of a distributed key-value store but also avoids overhead introduced by having running services around the file system. We implemented a prototype of this system and compared it against Cassandra, a state-of-art key-value store. Preliminary results are promising.

@inproceedings{
 title = {Filtering IP source spoofing using feasible path reverse path forwarding with SDN},
 type = {inproceedings},
 year = {2015},
 id = {9445c0de-a35c-3540-bba6-4d4ce2db9f5b},
 created = {2019-10-01T17:20:47.712Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:37.776Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Benton2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. Source IP spoofing is still an endemic challenge despite best practices documents being published more than 13 years ago that would prevent it if all ISPs abided by them. We argue that these approaches failed to gain widespread adoption due to fundamental incentive misalignment. We then propose an SDN-based solution designed to be placed at Internet exchange points by ISPs with the incentives to filter spoofed traffic. This solution doesn't replace existing routers and it runs on low-cost, high-speed OpenFlow switches with graceful failure strategies to make adoption simple.},
 bibtype = {inproceedings},
 author = {Benton, K. and Camp, L.J. and Kelley, T. and Swany, M.},
 doi = {10.1109/CNS.2015.7346909},
 booktitle = {2015 IEEE Conference on Communications and NetworkSecurity, CNS 2015}
}

© 2015 IEEE. Source IP spoofing is still an endemic challenge despite best practices documents being published more than 13 years ago that would prevent it if all ISPs abided by them. We argue that these approaches failed to gain widespread adoption due to fundamental incentive misalignment. We then propose an SDN-based solution designed to be placed at Internet exchange points by ISPs with the incentives to filter spoofed traffic. This solution doesn't replace existing routers and it runs on low-cost, high-speed OpenFlow switches with graceful failure strategies to make adoption simple.

@book{
 title = {Preface},
 type = {book},
 year = {2015},
 source = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
 volume = {8628},
 id = {ac2049e4-bdff-39ce-aa98-e5ad76f17365},
 created = {2019-10-01T17:20:48.024Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:24.820Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ludascher2015a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {book},
 author = {Ludäscher, B. and Plale, B.}
}

@inproceedings{
 title = {HELM: Conflict-free active measurement scheduling for shared network resource management},
 type = {inproceedings},
 year = {2015},
 id = {874cfbad-002f-3a03-a643-5775ec1c51ab},
 created = {2019-10-01T17:20:48.498Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:30.548Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. Network resource measurement is a key functionality for large scale network management. Intelligent, network-aware applications may benefit from access to detailed representations of network resources, including multi-layer topologies and real-time traffic measurement, and shared resources may obtain better overall utilization by identifying performance bottlenecks. In this study, we describe a network measurement framework, which includes a network topology analysis mechanism as well as agent tools for running active probes and collecting data from end hosts. The system includes a centralized coordinator, which abstracts network elements into annotated network graphs and applies scheduling algorithms to calculate conflict free measurement probes over shared links. Our evaluation integrated perfSONAR services into our framework and included deployment scenarios on research and education networks such as Internet2 and ESnet. The data presented in this study offers compelling evidence that supports a method by which to measure the performance of real world networks.},
 bibtype = {inproceedings},
 author = {Zhang, M. and Swany, M. and Yavanamanda, A. and Kissel, E.},
 doi = {10.1109/INM.2015.7140283},
 booktitle = {Proceedings of the 2015 IFIP/IEEE International Symposium on Integrated Network Management, IM 2015}
}

© 2015 IEEE. Network resource measurement is a key functionality for large scale network management. Intelligent, network-aware applications may benefit from access to detailed representations of network resources, including multi-layer topologies and real-time traffic measurement, and shared resources may obtain better overall utilization by identifying performance bottlenecks. In this study, we describe a network measurement framework, which includes a network topology analysis mechanism as well as agent tools for running active probes and collecting data from end hosts. The system includes a centralized coordinator, which abstracts network elements into annotated network graphs and applies scheduling algorithms to calculate conflict free measurement probes over shared links. Our evaluation integrated perfSONAR services into our framework and included deployment scenarios on research and education networks such as Internet2 and ESnet. The data presented in this study offers compelling evidence that supports a method by which to measure the performance of real world networks.

@inproceedings{
 title = {Trust threads: minimal provenance and data publication and reuse},
 type = {inproceedings},
 year = {2015},
 publisher = {Colorado State University. Libraries},
 id = {472a1100-6cf3-3093-8817-72a6a98cc6ed},
 created = {2019-10-01T17:20:48.811Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:29.668Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Plale2015b},
 source_type = {CONF},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Plale, Beth},
 booktitle = {National Data Integrity Conference-2015}
}

@inproceedings{
 title = {New Approaches to Capture High Frequency Agricultural Dynamics in Africa through Mobile Phones},
 type = {inproceedings},
 year = {2015},
 id = {3e1bfb5c-391e-335d-99c9-deb0d0252696},
 created = {2019-10-01T17:20:49.006Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:28.487Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Evans2015},
 source_type = {CONF},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Evans, T P and Attari, S and Plale, B A and Caylor, K K and Estes, L D and Sheffield, J},
 booktitle = {AGU Fall Meeting Abstracts}
}

@article{
 title = {NIST Special Publication 1500-6 : NIST Big Data Interoperability Framework - Reference Architecture},
 type = {article},
 year = {2015},
 keywords = {Adoption,barriers,implementation,interfaces,market maturity,organizational maturity,project maturity,system modernization.},
 pages = {1-62},
 volume = {6},
 id = {6cf91103-5efb-3107-9788-a1711c4c709a},
 created = {2019-10-01T17:20:55.148Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:33.520Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {NBD-PWG2015},
 private_publication = {false},
 abstract = {A341Big Data is a term used to describe the large amount of data in the networked, digitized, sensor-laden, information-driven world. While opportunities exist with Big Data, the data can overwhelm traditional technical approaches and the growth of data is outpacing scientific and technological advances in data analytics. To advance progress in Big Data, the NIST Big Data Public Working Group (NBD-PWG) is working to develop consensus on important, fundamental concepts related to Big Data. The results are reported in the NIST Big Data Interoperability Framework series of volumes. This volume, Volume 6, summarizes the work performed by the NBD-PWG to characterize Big Data from an architecture perspective, presents the NIST Big Data Reference Architecture (NBDRA) conceptual model, and discusses the components and fabrics of the NBDRA.},
 bibtype = {article},
 author = {(NBD-PWG), NIST Big Data Public Working Group},
 doi = {10.6028/NIST.SP.1500-6},
 number = {June}
}

A341Big Data is a term used to describe the large amount of data in the networked, digitized, sensor-laden, information-driven world. While opportunities exist with Big Data, the data can overwhelm traditional technical approaches and the growth of data is outpacing scientific and technological advances in data analytics. To advance progress in Big Data, the NIST Big Data Public Working Group (NBD-PWG) is working to develop consensus on important, fundamental concepts related to Big Data. The results are reported in the NIST Big Data Interoperability Framework series of volumes. This volume, Volume 6, summarizes the work performed by the NBD-PWG to characterize Big Data from an architecture perspective, presents the NIST Big Data Reference Architecture (NBDRA) conceptual model, and discusses the components and fabrics of the NBDRA.

@inproceedings{
 title = {Peer comparison of XSEDE and NCAR publication data},
 type = {inproceedings},
 year = {2015},
 volume = {2015-Octob},
 id = {4aa73316-c4c9-3f2f-8196-5a41d40f2c2e},
 created = {2019-10-01T17:20:56.226Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:42.447Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {VonLaszewski2015},
 folder_uuids = {82975498-107c-4bb3-bb76-f87bce3e9f6e},
 private_publication = {false},
 abstract = {© 2015 IEEE. We present a framework that compares the publication impact based on a comprehensive peer analysis of papers produced by scientists using XSEDE and NCAR resources. The analysis is introducing a percentile ranking based approach of citations of the XSEDE and NCAR papers compared to peer publications in the same journal that do not use these resources. This analysis is unique in that it evaluates the impact of the two facilities by comparing the reported publications from them to their peers from within the same journal issue. From this analysis, we can see that papers that utilize XSEDE and NCAR resources are cited statistically significantly more often. Hence we find that reported publications indicate that XSEDE and NCAR resources exert a strong positive impact on scientific research.},
 bibtype = {inproceedings},
 author = {Von Laszewski, G. and Wang, F. and Fox, Geoffrey Charles G.C. and Hart, D.L. and Furlani, T.R. and Deleon, R.L. and Gallo, S.M.},
 doi = {10.1109/CLUSTER.2015.98},
 booktitle = {Proceedings - IEEE International Conference on Cluster Computing, ICCC}
}

© 2015 IEEE. We present a framework that compares the publication impact based on a comprehensive peer analysis of papers produced by scientists using XSEDE and NCAR resources. The analysis is introducing a percentile ranking based approach of citations of the XSEDE and NCAR papers compared to peer publications in the same journal that do not use these resources. This analysis is unique in that it evaluates the impact of the two facilities by comparing the reported publications from them to their peers from within the same journal issue. From this analysis, we can see that papers that utilize XSEDE and NCAR resources are cited statistically significantly more often. Hence we find that reported publications indicate that XSEDE and NCAR resources exert a strong positive impact on scientific research.

@techreport{
 title = {Scalability Analysis of the Multi-Look Time Domain Processor on XSEDE Compute Resources},
 type = {techreport},
 year = {2015},
 id = {e2c9651a-baff-30d6-9b2f-bef22b47b81d},
 created = {2019-10-01T17:20:56.555Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:40.398Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {VonLaszewski2015a},
 source_type = {JOUR},
 folder_uuids = {82975498-107c-4bb3-bb76-f87bce3e9f6e},
 private_publication = {false},
 bibtype = {techreport},
 author = {von Laszewski, Gregor and Wang, Fugang and Fox, Geoffrey Charles and Li, Jilu and Paden, John}
}

@article{
 title = {Physiological effects of heat stress on Hawaiian picture-wing Drosophila: Genome-wide expression patterns and stress-related traits},
 type = {article},
 year = {2015},
 keywords = {Gene expression,Hawaiian Drosophila,Local adaptation,Microarray,Sperm mobility,Temperature tolerance},
 volume = {3},
 publisher = {Oxford University Press},
 id = {a41fafc4-a0da-3938-bc32-7063510785bb},
 created = {2019-10-01T17:20:57.105Z},
 accessed = {2019-08-29},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:32.564Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Uy2015},
 private_publication = {false},
 abstract = {Two Hawaiian picture-wing Drosophila differ in their temperature tolerances with  the ecologically rare species, D. silvestris, showing reduced survival, reduced sperm mobility and greater gene expression changes at high temperatures compared to the common D. sproati. Thus the rare species may have reduced capacity to adapt to future climate changes.},
 bibtype = {article},
 author = {Uy, Karen L. and LeDuc, R. and Ganote, C. and Price, Donald K.},
 doi = {10.1093/conphys/cou062},
 journal = {Conservation Physiology},
 number = {1}
}

Two Hawaiian picture-wing Drosophila differ in their temperature tolerances with the ecologically rare species, D. silvestris, showing reduced survival, reduced sperm mobility and greater gene expression changes at high temperatures compared to the common D. sproati. Thus the rare species may have reduced capacity to adapt to future climate changes.

@inproceedings{
 title = {TAS view of XSEDE users and usage},
 type = {inproceedings},
 year = {2015},
 volume = {2015-July},
 id = {79b9fcad-c38a-38f7-b13e-0c546b66b422},
 created = {2019-10-01T17:20:58.640Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:31.682Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {DeLeon2015},
 folder_uuids = {82975498-107c-4bb3-bb76-f87bce3e9f6e},
 private_publication = {false},
 abstract = {© 2015 ACM. The Technology Audit Service has developed, XDMoD, a resource management tool. This paper utilizes XDMoD and the XDMoD data warehouse that it draws from to provide a broad overview of several aspects of XSEDE users and their usage. Some important trends include: 1) in spite of a large yearly turnover, there is a core of users persisting over many years, 2) user job submission has changed from primarily faculty members to students and postdocs, 3) increases in usage in Molecular Biosciences and Materials Research has outstripped that of other fields of science, 4) the distribution of user external funding is bimodal with one group having a large ratio of external funding to internal XSEDE funding (ie, CPU cycles) and a second group having a small ratio of external to internal (CPU cycle) funding, 5) user job efficiency is also bimodal with a group of presumably new users running mainly small inefficient jobs and another group of users running larger more efficient jobs, 6) finally, based on an analysis of citations of published papers, the scientific impact of XSEDE coupled with the service providers is demonstrated in the statistically significant advantage it provides to the research of its users.},
 bibtype = {inproceedings},
 author = {DeLeon, R.L. and Furlani, T.R. and Gallo, S.M. and White, J.P. and Jones, M.D. and Patra, A. and Innus, M. and Yearke, T. and Palmer, J.T. and Sperhac, J.M. and Rathsam, R. and Simakov, N. and Von Laszewski, G. and Wang, F.},
 doi = {10.1145/2792745.2792766},
 booktitle = {ACM International Conference Proceeding Series}
}

© 2015 ACM. The Technology Audit Service has developed, XDMoD, a resource management tool. This paper utilizes XDMoD and the XDMoD data warehouse that it draws from to provide a broad overview of several aspects of XSEDE users and their usage. Some important trends include: 1) in spite of a large yearly turnover, there is a core of users persisting over many years, 2) user job submission has changed from primarily faculty members to students and postdocs, 3) increases in usage in Molecular Biosciences and Materials Research has outstripped that of other fields of science, 4) the distribution of user external funding is bimodal with one group having a large ratio of external funding to internal XSEDE funding (ie, CPU cycles) and a second group having a small ratio of external to internal (CPU cycle) funding, 5) user job efficiency is also bimodal with a group of presumably new users running mainly small inefficient jobs and another group of users running larger more efficient jobs, 6) finally, based on an analysis of citations of published papers, the scientific impact of XSEDE coupled with the service providers is demonstrated in the statistically significant advantage it provides to the research of its users.

@inproceedings{
 title = {Towards sustainable curation and preservation: The SEAD project's data services approach},
 type = {inproceedings},
 year = {2015},
 id = {b8e0bd52-a4ad-37ea-a3ee-e8c09a35332e},
 created = {2019-10-01T17:20:58.686Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:59.977Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Myers2015},
 folder_uuids = {2aba6c14-9027-4f47-8627-0902e1e2342b,73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. When the effort to curate and preserve data is made at the end of a project, there is little opportunity to leverage ongoing research work to reduce curation costs or conversely, to leverage curation efforts to improve research productivity. In the Sustainable Environment Actionable Data (SEAD) project, we have envisioned a more active approach to data curation and preservation in which these processes occur in parallel with research and generate sufficient short and long-term return on researcher investments for self-interest to drive their adoption. In this paper, we describe the conceptual framework motivating the SEAD project and the suite of data services we have developed and deployed as an initial implementation of this approach. Use cases in which these services can reduce curation effort and aid ongoing research are highlighted and, based on our experience to date, we identify some key architectural features of our approach as well as open challenges to fully realizing the value of this approach in the broad ecosystem of cyberinfrastructure.},
 bibtype = {inproceedings},
 author = {Myers, J. and Hedstrom, M. and Akmon, D. and Payette, S. and Plale, B.A. and Kouper, I. and McCaulay, S. and McDonald, R. and Suriarachchi, I. and Varadharaju, A. and Kumar, P. and Elag, M. and Lee, J. and Kooper, R. and Marini, L.},
 doi = {10.1109/eScience.2015.56},
 booktitle = {Proceedings - 11th IEEE International Conference on eScience, eScience 2015}
}

© 2015 IEEE. When the effort to curate and preserve data is made at the end of a project, there is little opportunity to leverage ongoing research work to reduce curation costs or conversely, to leverage curation efforts to improve research productivity. In the Sustainable Environment Actionable Data (SEAD) project, we have envisioned a more active approach to data curation and preservation in which these processes occur in parallel with research and generate sufficient short and long-term return on researcher investments for self-interest to drive their adoption. In this paper, we describe the conceptual framework motivating the SEAD project and the suite of data services we have developed and deployed as an initial implementation of this approach. Use cases in which these services can reduce curation effort and aid ongoing research are highlighted and, based on our experience to date, we identify some key architectural features of our approach as well as open challenges to fully realizing the value of this approach in the broad ecosystem of cyberinfrastructure.

@article{
 title = {Komadu: A capture and visualization system for scientific data provenance},
 type = {article},
 year = {2015},
 volume = {3},
 publisher = {Ubiquity Press},
 id = {2e11c809-0cc0-34ec-ab8d-b7df34ff1fa5},
 created = {2019-10-01T17:20:59.209Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:56.497Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Suriarachchi2015},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {article},
 author = {Suriarachchi, Isuru and Zhou, Quan and Plale, Beth},
 journal = {Journal of Open Research Software},
 number = {1}
}

@article{
 title = {Research challenges in future multi-domain network performance measurement and monitoring},
 type = {article},
 year = {2015},
 volume = {45},
 id = {d33c3346-6a2d-3e47-801c-ce5a51cfe2c1},
 created = {2019-10-01T17:20:59.401Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:05.457Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Calyam2015},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {The perfSONAR-based Multi-domain Network Performance Measurement and Monitoring Workshop was held on Febru- Ary 20-21, 2014 in Arlington, VA. The goal of the workshop was to review the state of the perfSONAR effort and cat- Alyze future directions by cross-fertilizing ideas, and distill- ing common themes among the diverse perfSONAR stake- holders that include: network operators and managers, end- users and network researchers. The timing and organiza- Tion for the second workshop is significant because there are an increasing number of groups within NSF supported data-intensive computing and networking programs that are dealing with measurement, monitoring and troubleshoot- ing of multi-domain issues. These groups are forming ex- plicit measurement federations using perfSONAR to address a wide range of issues. In addition, the emergence and wide-adoption of new paradigms such as software-defined networking are taking shape to aid in traffic management needs of scientific communities and network operators. Con- sequently, there are new challenges that need to be addressed for extensible and programmable instrumentation, measure-ment data analysis, visualization and middleware security features in perfSONAR. This report summarizes the work- shop efforts to bring together diverse groups for delivering targeted short/long talks, sharing latest advances, and iden- Tifying gaps that exist in the community for solving end-to- end performance problems in an effctive, scalable fashion.},
 bibtype = {article},
 author = {Calyam, P. and Swany, M.},
 journal = {Computer Communication Review},
 number = {3}
}

The perfSONAR-based Multi-domain Network Performance Measurement and Monitoring Workshop was held on Febru- Ary 20-21, 2014 in Arlington, VA. The goal of the workshop was to review the state of the perfSONAR effort and cat- Alyze future directions by cross-fertilizing ideas, and distill- ing common themes among the diverse perfSONAR stake- holders that include: network operators and managers, end- users and network researchers. The timing and organiza- Tion for the second workshop is significant because there are an increasing number of groups within NSF supported data-intensive computing and networking programs that are dealing with measurement, monitoring and troubleshoot- ing of multi-domain issues. These groups are forming ex- plicit measurement federations using perfSONAR to address a wide range of issues. In addition, the emergence and wide-adoption of new paradigms such as software-defined networking are taking shape to aid in traffic management needs of scientific communities and network operators. Con- sequently, there are new challenges that need to be addressed for extensible and programmable instrumentation, measure-ment data analysis, visualization and middleware security features in perfSONAR. This report summarizes the work- shop efforts to bring together diverse groups for delivering targeted short/long talks, sharing latest advances, and iden- Tifying gaps that exist in the community for solving end-to- end performance problems in an effctive, scalable fashion.

@inproceedings{
 title = {ProvErr: System level statistical fault diagnosis using dependency model},
 type = {inproceedings},
 year = {2015},
 id = {d97bfbf5-2463-3752-9dd8-376e89c033c6},
 created = {2019-10-01T17:20:59.446Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:04.831Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Chen2015a},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {© 2015 IEEE. Large-scale distributed systems are difficult to debug in the event of failure. Yet rapid fault diagnosis that pinpoints failures to the component level is critical to fast recovery. We introduce a statistical approach to fault diagnosis that utilizes a dependency graph of execution to automatically discover the most probable fault cause(s) at a component level (either software or hardware resource). This approach leverages engineers' high level understanding of the system and requires a very small amount of information compared to existing methods. It also utilizes dependency information to eliminate redundant causes while retaining co-causes. Experiments using Apache Pig show that our approach has good, robust performance for diagnosing software bugs and resource shortages, and scales nearly linearly as system size increases.},
 bibtype = {inproceedings},
 author = {Chen, P. and Plale, B.A.},
 doi = {10.1109/CCGrid.2015.86},
 booktitle = {Proceedings - 2015 IEEE/ACM 15th International Symposium on Cluster, Cloud, and Grid Computing, CCGrid 2015}
}

© 2015 IEEE. Large-scale distributed systems are difficult to debug in the event of failure. Yet rapid fault diagnosis that pinpoints failures to the component level is critical to fast recovery. We introduce a statistical approach to fault diagnosis that utilizes a dependency graph of execution to automatically discover the most probable fault cause(s) at a component level (either software or hardware resource). This approach leverages engineers' high level understanding of the system and requires a very small amount of information compared to existing methods. It also utilizes dependency information to eliminate redundant causes while retaining co-causes. Experiments using Apache Pig show that our approach has good, robust performance for diagnosing software bugs and resource shortages, and scales nearly linearly as system size increases.

@techreport{
 title = {Software in Science: a Report of Outcomes of the 2014 National Science Foundation Software Infrastructure for Sustained Innovation (SI2) Meeting},
 type = {techreport},
 year = {2015},
 id = {3d878a60-fe83-3c31-8882-a50761daae93},
 created = {2019-10-01T17:20:59.669Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:24:04.230Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Plale2015a},
 source_type = {RPRT},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {techreport},
 author = {Plale, Beth and Jones, Matt and Thain, Douglas}
}

@techreport{
 title = {The Data Capsule for Non-Consumptive Research},
 type = {techreport},
 year = {2015},
 id = {6cf95121-bdb1-3e88-9032-99bae60f8051},
 created = {2019-10-01T17:21:00.425Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:43.266Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Plale2015},
 source_type = {RPRT},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {techreport},
 author = {Plale, Beth and Prakash, Atul and McDonald, Robert}
}

@article{
 title = {Advantages to Geoscience and disaster response from QuakeSim implementation of interferometric radar maps in a GIS database system},
 type = {article},
 year = {2015},
 pages = {2295-2304},
 volume = {172},
 publisher = {Springer Basel},
 id = {9f0c473a-1ff6-3022-91a5-16aab02d0d44},
 created = {2019-10-01T17:21:05.824Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:28:20.058Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {parker2015advantages},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {article},
 author = {Parker, Jay and Donnellan, Andrea and Glasscoe, Margaret and Fox, Geoffrey Charles and Wang, Jun and Pierce, Marlon and Ma, Yu and Wang, Jun and Ma, Yu},
 journal = {Pure and Applied Geophysics},
 number = {8}
}

@inproceedings{
 title = {Disaster Response Tools for Decision Support and Data Discovery-E-DECIDER and GeoGateway},
 type = {inproceedings},
 year = {2015},
 id = {e9696472-e03f-380f-989e-7a7bb917b348},
 created = {2019-10-01T17:21:07.604Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:36.749Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {glasscoe2015disaster},
 source_type = {inproceedings},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Glasscoe, M T and Donnellan, A and Parker, J W and Granat, R A and Lyzenga, G A and Pierce, M E and Wang, J and Grant Ludwig, L and Eguchi, R T and Huyck, C K and others, undefined},
 booktitle = {AGU Fall Meeting Abstracts}
}

@inproceedings{
 title = {An Overview of the XSEDE Extended Collaborative Support Program},
 type = {inproceedings},
 year = {2015},
 keywords = {Collaborative support,Computer science,Computers,Cyber infrastructures,Sci,Supercomputers},
 pages = {3-13},
 volume = {595},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84964076487&doi=10.1007%2F978-3-319-32243-8_1&partnerID=40&md5=4c26cab3d25ddd6fb9bd0205e1f593e6},
 publisher = {Springer Verlag},
 institution = {Springer International Publishing},
 id = {70f8f545-0c4c-3dfe-b84c-2980677e6083},
 created = {2019-10-01T17:21:07.730Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:35.569Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wilkins-Diehr20163},
 source_type = {article},
 notes = {<b>From Duplicate 2 (<i>An overview of the XSEDE extended collaborative support program</i> - Wilkins-Diehr, N; Sanielevici, S; Alameda, J; Cazes, J; Crosby, L; Pierce, M; Roskies, R)<br/></b><br/>cited By 4; Conference of 6th International Conference on High Performance Computer Applications, ISUM 2015 ; Conference Date: 9 March 2015 Through 13 March 2015; Conference Code:173549},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95,9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 abstract = {The Extreme Science and Engineering Discovery Environment (XSEDE) is a flagship cyberinfrastructure project funded by the US National Science Foundation (NSF). XSEDE’s Extended Collaborative Support Services (ECSS) program is a significant component of the XSEDE effort, dedicated to extended engagements with our user community which transform their research. We describe the organization, operation and some highlights of the program in this submission. © Springer International Publishing Switzerland 2016.},
 bibtype = {inproceedings},
 author = {Wilkins-Diehr, Nancy and Sanielevici, Sergiu and Alameda, Jay and Cazes, John and Crosby, Lonnie and Pierce, Marlon and Roskies, Ralph},
 editor = {Gitler I., Klapp J},
 doi = {10.1007/978-3-319-32243-8_1},
 booktitle = {International Conference on Supercomputing}
}

The Extreme Science and Engineering Discovery Environment (XSEDE) is a flagship cyberinfrastructure project funded by the US National Science Foundation (NSF). XSEDE’s Extended Collaborative Support Services (ECSS) program is a significant component of the XSEDE effort, dedicated to extended engagements with our user community which transform their research. We describe the organization, operation and some highlights of the program in this submission. © Springer International Publishing Switzerland 2016.

@article{
 title = {E-decider: Using earth science data and modeling tools to develop decision support for earthquake disaster response},
 type = {article},
 year = {2015},
 pages = {2305-2324},
 volume = {172},
 publisher = {Springer Basel},
 id = {e874fad9-9c1c-3f67-847e-1c8b927f25cb},
 created = {2019-10-01T17:21:09.287Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:27:31.521Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {glasscoe2015decider},
 source_type = {article},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8},
 private_publication = {false},
 bibtype = {article},
 author = {Glasscoe, Margaret T and Wang, Jun and Pierce, Marlon E and Yoder, Mark R and Parker, Jay W and Burl, Michael C and Stough, Timothy M and Granat, Robert A and Donnellan, Andrea and Rundle, John B and others, undefined},
 journal = {Pure and Applied Geophysics},
 number = {8}
}

@techreport{
 title = {XSIM Final Report: Modelling the Past and Future of Identity Management for Scientific Collaborations},
 type = {techreport},
 year = {2015},
 websites = {http://hdl.handle.net/2022/20399},
 month = {9},
 city = {Bloomington, IN},
 institution = {Indiana University},
 id = {ad6de5b3-31d5-3e3b-8b30-ebd7a4c64907},
 created = {2019-10-01T17:21:11.838Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:43.082Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cowles2015},
 source_type = {techreport},
 notes = {From Duplicate 2 (XSIM Final Report: Modelling the Past and Future of Identity Management for Scientific Collaborations - Cowles, R; Jackson, C; Welch, V)<br/><br/>US Dept of Energy Next-Generation Networks for Science (NGNS) program Grant No. DE-FG02-12ER26111},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {techreport},
 author = {Cowles, R and Jackson, C and Welch, V}
}

@article{
 title = {Risk Management in Data Protection},
 type = {article},
 year = {2015},
 pages = {95-98},
 volume = {5},
 id = {7b9ee1ea-4fd2-3c6f-baca-de59bf6676c0},
 created = {2019-10-01T17:21:12.819Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:38.699Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2015b},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Millard, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla},
 doi = {https://doi.org/10.1093/idpl/ipv005},
 journal = {International Data Privacy Law},
 number = {2}
}

@techreport{
 title = {Center for Applied Cybersecurity Research: 2015 Annual Report and Strategic Plan (2015-2020)},
 type = {techreport},
 year = {2015},
 id = {029ef4b3-168a-3d5b-9a2c-558f1815e9c9},
 created = {2019-10-01T17:21:13.256Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:52.848Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Delaney2015},
 source_type = {RPRT},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {techreport},
 author = {Delaney, David G and Welch, Von and Starzynski Coddens, Amy}
}

@inproceedings{
 title = {Facilitating Scientific Collaborations by Delegating Identity Management},
 type = {inproceedings},
 year = {2015},
 keywords = {Access control,Cyber security,Delegation,Identity,Identity ma,Risk management},
 pages = {15-19},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84979747691&doi=10.1145%2F2752499.2752501&partnerID=40&md5=83a416d313602ca3c75ba6528f549833,http://dl.acm.org/citation.cfm?doid=2752499.2752501},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {c7adbf2e-4b03-3679-9ace-6fca3a6af5e2},
 created = {2019-10-01T17:21:14.967Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:30.179Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cowles201515},
 source_type = {conference},
 notes = {cited By 0; Conference of 2nd Workshop on Changing Landscapes in HPC Security, CLHS 2015 ; Conference Date: 16 June 2015; Conference Code:122511},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {DOE Labs are often presented with conflicting requirements for providing services to scientific collaboratories. An identity management model involving transitive trust is increasingly common. We show how existing policies allow for increased delegation of identity management within an acceptable risk management framework. Specific topics addressed include deemed exports, DOE orders, Inertia and Risk, Traceability, and Technology Limitations. Real life examples of an incremental approach to implementing transitive trust are presented.},
 bibtype = {inproceedings},
 author = {Cowles, Robert and Jackson, Craig and Welch, Von},
 doi = {10.1145/2752499.2752501},
 booktitle = {Proceedings of the 2015 Workshop on Changing Landscapes in HPC Security - CLHS '15}
}

DOE Labs are often presented with conflicting requirements for providing services to scientific collaboratories. An identity management model involving transitive trust is increasingly common. We show how existing policies allow for increased delegation of identity management within an acceptable risk management framework. Specific topics addressed include deemed exports, DOE orders, Inertia and Risk, Traceability, and Technology Limitations. Real life examples of an incremental approach to implementing transitive trust are presented.

@article{
 title = {Protecting Privacy in Big Data},
 type = {article},
 year = {2015},
 pages = {7-20},
 volume = {8},
 id = {01a877c7-197e-3756-8b4d-56d3b2825a72},
 created = {2019-10-01T17:21:16.335Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:01.154Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2015c},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H},
 journal = {Journal of Law and Economic Regulation},
 number = {1}
}

@article{
 title = {The Data Protection Credibility Crisis},
 type = {article},
 year = {2015},
 pages = {3},
 volume = {5},
 id = {942523fb-1030-336d-a071-f6b0ee27335f},
 created = {2019-10-01T17:21:16.471Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:22:59.763Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2015a},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Millard, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla},
 journal = {International Data Privacy Law}
}

@article{
 title = {Internet Balkanization Gathers Pace: Is Privacy the Real Driver?},
 type = {article},
 year = {2015},
 pages = {1-2},
 volume = {5},
 id = {eb5bcd5e-1f7a-3165-8551-abaceefd1664},
 created = {2019-10-01T17:21:17.296Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:04.936Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cate2015},
 source_type = {JOUR},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b},
 private_publication = {false},
 bibtype = {article},
 author = {Cate, Fred H and Kuner, Christopher and Millard, Christopher and Svantesson, Dan Jerker B and Lynskey, Orla},
 doi = {https://doi.org/10.1093/idpl/ipu032},
 journal = {International Data Privacy Law},
 number = {1}
}

@article{
 title = {Apache Airavata: design and directions of a science gateway framework},
 type = {article},
 year = {2015},
 keywords = {cyberinfrastructure,distributed computing infrastructure,science gateways},
 pages = {4282-4291},
 volume = {27},
 websites = {http://doi.wiley.com/10.1002/cpe.3534},
 month = {11},
 publisher = {John Wiley and Sons Ltd},
 day = {1},
 id = {e818ceea-1fe4-3ab7-ac18-b8751a839685},
 created = {2019-10-01T17:21:23.785Z},
 accessed = {2019-09-19},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:27.558Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pierce2015},
 private_publication = {false},
 abstract = {This paper provides an overview of the Apache Airavata software system for science gateways. Gateways use Airavata to manage application and workflow executions on a range of backend resources (grids, computing clouds, and local clusters). Airavata's design goal is to provide component abstractions for major tasks required to provide gateway application management. Components are not directly accessed but are instead exposed through a client Application Programming Interface. This design allows gateway developers to take full advantage of Airavata's capabilities, and Airavata developers (including those interested in middleware research) to modify Airavata's implementations and behavior. This is particularly important as Airavata evolves to become a scalable, elastic "platform as a service" for science gateways. We illustrate the capabilities of Airavata through the discussion of usage vignettes. As an Apache Software Foundation project, Airavata's open community governance model is as important as its software base. We discuss how this works within Airavata and how it may be applicable to other distributed computing infrastructure and cyber infrastructure efforts. © 2014 IEEE.},
 bibtype = {article},
 author = {Pierce, Marlon E. and Marru, Suresh and Gunathilake, Lahiru and Wijeratne, Don Kushan and Singh, Raminder and Wimalasena, Chathuri and Ratnayaka, Shameera and Pamidighantam, Sudhakar},
 doi = {10.1002/cpe.3534},
 journal = {Concurrency and Computation: Practice and Experience},
 number = {16}
}

This paper provides an overview of the Apache Airavata software system for science gateways. Gateways use Airavata to manage application and workflow executions on a range of backend resources (grids, computing clouds, and local clusters). Airavata's design goal is to provide component abstractions for major tasks required to provide gateway application management. Components are not directly accessed but are instead exposed through a client Application Programming Interface. This design allows gateway developers to take full advantage of Airavata's capabilities, and Airavata developers (including those interested in middleware research) to modify Airavata's implementations and behavior. This is particularly important as Airavata evolves to become a scalable, elastic "platform as a service" for science gateways. We illustrate the capabilities of Airavata through the discussion of usage vignettes. As an Apache Software Foundation project, Airavata's open community governance model is as important as its software base. We discuss how this works within Airavata and how it may be applicable to other distributed computing infrastructure and cyber infrastructure efforts. © 2014 IEEE.

@inproceedings{
 title = {Programmable Immersive Peripheral Environmental System (PIPES): A prototype control system for environmental feedback devices},
 type = {inproceedings},
 year = {2015},
 keywords = {Computer programming,Control systems,Electronic systems,End to end latencies,Environ,Sensory fee,Virtual reality},
 volume = {9392},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84928473397&doi=10.1117%2F12.2083410&partnerID=40&md5=2ae847efbe40d8d9e0a6ac4d2ed3f04e},
 publisher = {SPIE},
 id = {2a81e0fd-0fdc-3cb6-b6cd-fb5ea7a87c80},
 created = {2019-10-01T17:21:27.468Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:35.828Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Frend2015},
 source_type = {conference},
 notes = {cited By 0; Conference of Engineering Reality of Virtual Reality 2015 ; Conference Date: 9 February 2015 Through 10 February 2015; Conference Code:111961},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {This paper describes an environmental feedback device (EFD) control system aimed at simplifying the VR development cycle. Programmable Immersive Peripheral Environmental System (PIPES) affords VR developers a custom approach to programming and controlling EFD behaviors while relaxing the required knowledge and expertise of electronic systems. PIPES has been implemented for the Unity engine and features EFD control using the Arduino integrated development environment. PIPES was installed and tested on two VR systems, a large format CAVE system and an Oculus Rift HMD system. A photocell based end-to-end latency experiment was conducted to measure latency within the system. This work extends previously unpublished prototypes of a similar design. Development and experiments described in this paper are part of the VR community goal to understand and apply environment effects to VEs that ultimately add to users' perceived presence. © 2015 SPIE-IS&T.},
 bibtype = {inproceedings},
 author = {Frend, C and Boyles, M},
 editor = {Dolinsky M., McDowall I E},
 doi = {10.1117/12.2083410},
 booktitle = {Proceedings of SPIE - The International Society for Optical Engineering}
}

This paper describes an environmental feedback device (EFD) control system aimed at simplifying the VR development cycle. Programmable Immersive Peripheral Environmental System (PIPES) affords VR developers a custom approach to programming and controlling EFD behaviors while relaxing the required knowledge and expertise of electronic systems. PIPES has been implemented for the Unity engine and features EFD control using the Arduino integrated development environment. PIPES was installed and tested on two VR systems, a large format CAVE system and an Oculus Rift HMD system. A photocell based end-to-end latency experiment was conducted to measure latency within the system. This work extends previously unpublished prototypes of a similar design. Development and experiments described in this paper are part of the VR community goal to understand and apply environment effects to VEs that ultimately add to users' perceived presence. © 2015 SPIE-IS&T.

@article{
 title = {SPEC ACCEL: A standard application suite for measuring hardware accelerator performance},
 type = {article},
 year = {2015},
 keywords = {Acceleration,Benchmarking,Electric power measurement,Hardware,Hardware accelerators,OpenCL,Openacc,Performan},
 pages = {46-67},
 volume = {8966},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84942519551&doi=10.1007%2F978-3-319-17248-4_3&partnerID=40&md5=45e24ea1a1e933c5be9ffb4b9c7a137b},
 publisher = {Springer Verlag},
 id = {dc794ce9-6f4f-3dab-9024-b9038304b913},
 created = {2019-10-01T17:21:28.276Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:23.100Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Juckeland201546},
 source_type = {article},
 notes = {cited By 8; Conference of 5th International Workshop on Performance Modeling, Benchmarking, and Simulation of High Performance Computing Systems, PMBS 2014 ; Conference Date: 16 November 2014 Through 16 November 2014; Conference Code:142329},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Hybrid nodes with hardware accelerators are becoming very common in systems today. Users often find it difficult to characterize and understand the performance advantage of such accelerators for their applications. The SPEC High Performance Group (HPG) has developed a set of performance metrics to evaluate the performance and power consumption of accelerators for various science applications. The new benchmark comprises two suites of applications written in OpenCL and OpenACC and measures the performance of accelerators with respect to a reference platform. The first set of published results demonstrate the viability and relevance of the new metrics in comparing accelerator performance. This paper discusses the benchmark suites and selected published results in great detail. © Springer International Publishing Switzerland 2015.},
 bibtype = {article},
 author = {Juckeland, G and Brantley, W and Chandrasekaran, S and Chapman, B and Che, S and Colgrove, M and Feng, H and Grund, A and Henschel, R and Hwu, W.-M.W. and Li, H and Müller, M S and Nagel, W E and Perminov, M and Shelepugin, P and Skadron, K and Stratton, J and Titov, A and Wang, K and Van Waveren, M and Whitney, B and Wienke, S and Xu, R and Kumaran, K},
 editor = {Hammond S.D. Jarvis S.A., Wright S A},
 doi = {10.1007/978-3-319-17248-4_3},
 journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}
}

Hybrid nodes with hardware accelerators are becoming very common in systems today. Users often find it difficult to characterize and understand the performance advantage of such accelerators for their applications. The SPEC High Performance Group (HPG) has developed a set of performance metrics to evaluate the performance and power consumption of accelerators for various science applications. The new benchmark comprises two suites of applications written in OpenCL and OpenACC and measures the performance of accelerators with respect to a reference platform. The first set of published results demonstrate the viability and relevance of the new metrics in comparing accelerator performance. This paper discusses the benchmark suites and selected published results in great detail. © Springer International Publishing Switzerland 2015.

@inproceedings{
 title = {Cyberinfrastructure Begins at Home},
 type = {inproceedings},
 year = {2015},
 websites = {http://hdl.handle.net/2022/20139},
 id = {3d8918a5-2fc4-3154-8d2d-219ad2c73d31},
 created = {2019-10-01T17:21:28.526Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.904Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hancock2015},
 source_type = {inproceedings},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {This presentation at the SPXXL winter workshop looks at cyberinfrastructure at Indiana University. It begins with an extensive look at IU's background and discusses where we are now, how we got here, and where we think we are going. It also covers some generalizations as well as looking at specific services and projects at IU including participation in XSEDE and the Jetstream system.},
 bibtype = {inproceedings},
 author = {Hancock, D Y and Henschel, R and Kallback-Rose, K and Stewart, C A},
 booktitle = {SPXXL Winter Workshop}
}

This presentation at the SPXXL winter workshop looks at cyberinfrastructure at Indiana University. It begins with an extensive look at IU's background and discusses where we are now, how we got here, and where we think we are going. It also covers some generalizations as well as looking at specific services and projects at IU including participation in XSEDE and the Jetstream system.

@techreport{
 title = {Use of IU parallel computing resources and high performance file systems-July 2013 to Dec 2014},
 type = {techreport},
 year = {2015},
 id = {115eb13e-d7b5-3907-a619-f7eee8275918},
 created = {2019-10-01T17:21:29.111Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:28.677Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Link2015},
 source_type = {RPRT},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Link, Matthew R and Simms, Stephen C and Stewart, Craig A and Henschel, Robert and Fulton, Benjamin}
}

@techreport{
 title = {IU PTI/UITS Research Technologies Annual Report: FY 2014},
 type = {techreport},
 year = {2015},
 id = {9c8a8997-4544-3f92-a5a1-ba222508a416},
 created = {2019-10-01T17:21:29.428Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:26:25.623Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015e},
 source_type = {RPRT},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A and Miller, Therese}
}

@inproceedings{
 title = {Authentication and Authorization Considerations for a Multi-tenant Service},
 type = {inproceedings},
 year = {2015},
 pages = {29-35},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84962377238&doi=10.1145%2F2753524.2753534&partnerID=40&md5=06c617703f402a57e3922f8a290ed55d,http://dl.acm.org/citation.cfm?doid=2753524.2753534},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {b25b955b-6670-308f-b1bf-4d5aa029fdd3},
 created = {2019-10-01T18:06:09.871Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:06:33.599Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Heiland201529},
 source_type = {conference},
 notes = {cited By 3; Conference of 1st Workshop on the Science of Cyberinfrastructure: Research, Experience, Applications and Models, SCREAM 2015 ; Conference Date: 16 June 2015; Conference Code:116136},
 folder_uuids = {f285719c-254b-42e8-a6fb-527e7c80488b,9559c0d7-009c-4a6d-b131-a3e3d83d98b8,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Distributed cyberinfrastructure requires users (and machines) to perform some sort of authentication and authorization (together simply known as auth). In the early days of computing, authentication was performed with just a username and password combination, and this is still prevalent today. But during the past several years, we have seen an evolution of approaches and protocols for auth: Kerberos, SSH keys, X.509, OpenID, API keys, OAuth, and more. Not surprisingly, there are trade-offs, both technical and social, for each approach. The NSF Science Gateway communities have had to deal with a variety of auth issues. However, most of the early gateways were rather restrictive in their model of access and development. The practice of using community credentials (certificates), a well-intentioned idea to alleviate restrictive access, still posed a barrier to researchers and challenges for security and auditing. And while the web portal-based gateway clients offered users easy access from a browser, both the interface and the back-end functionality were constrained in the flexibility and extensibility they could provide. Designing a well-defined application programming interface (API) to fine-grained, generic gateway services (on secure, hosted cyberinfrastructure), together with an auth approach that has a lower barrier to entry, will hopefully present a more welcoming environment for both users and developers. This paper provides a review and some thoughts on these topics, with a focus on the role of auth between a Science Gateway and a service provider.},
 bibtype = {inproceedings},
 author = {Heiland, Randy and Koranda, Scott and Marru, Suresh and Pierce, Marlon and Welch, Von},
 doi = {10.1145/2753524.2753534},
 booktitle = {Proceedings of the 1st Workshop on The Science of Cyberinfrastructure Research, Experience, Applications and Models - SCREAM '15}
}

Distributed cyberinfrastructure requires users (and machines) to perform some sort of authentication and authorization (together simply known as auth). In the early days of computing, authentication was performed with just a username and password combination, and this is still prevalent today. But during the past several years, we have seen an evolution of approaches and protocols for auth: Kerberos, SSH keys, X.509, OpenID, API keys, OAuth, and more. Not surprisingly, there are trade-offs, both technical and social, for each approach. The NSF Science Gateway communities have had to deal with a variety of auth issues. However, most of the early gateways were rather restrictive in their model of access and development. The practice of using community credentials (certificates), a well-intentioned idea to alleviate restrictive access, still posed a barrier to researchers and challenges for security and auditing. And while the web portal-based gateway clients offered users easy access from a browser, both the interface and the back-end functionality were constrained in the flexibility and extensibility they could provide. Designing a well-defined application programming interface (API) to fine-grained, generic gateway services (on secure, hosted cyberinfrastructure), together with an auth approach that has a lower barrier to entry, will hopefully present a more welcoming environment for both users and developers. This paper provides a review and some thoughts on these topics, with a focus on the role of auth between a Science Gateway and a service provider.

@article{
 title = {Science gateways today and tomorrow: Positive perspectives of nearly 5000 members of the research community},
 type = {article},
 year = {2015},
 keywords = {Web interfaces,cyberinfrastructure,portals,science/engineering gateways,software development,survey},
 pages = {4252-4268},
 volume = {27},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84944911705&doi=10.1002%2Fcpe.3526&partnerID=40&md5=4dc5238c285d05e8713efa05ff42d452},
 publisher = {John Wiley and Sons Ltd},
 id = {5ae33fdd-f126-37c6-b91d-5f5ec6c5fdae},
 created = {2019-10-01T18:06:10.789Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T18:06:36.572Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lawrence20154252},
 source_type = {article},
 notes = {cited By 11},
 folder_uuids = {9559c0d7-009c-4a6d-b131-a3e3d83d98b8,22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {Science gateways are digital interfaces to advanced technologies that\nsupport science/engineering research/education. Frequently implemented\nas Web and mobile applications, they provide access to community\nresources such as software, data, collaboration tools, instrumentation,\nand high-performance computing. We anticipate opportunities for growth\nwithin a fragmented community. Through a large-scale survey, we measured\nthe extent and characteristics of the gateway community (reliance on\ngateways and nature of existing resources) to understand useful services\nand support for builders and users. We administered an online survey to\nnearly 29,000 principal investigators, senior administrators, and people\nwith gateway affiliations. Nearly 5000 respondents represented diverse\nexpertise and geography. The majority of researchers/educators indicated\nthat specialized online resources were important to their work. They\nchoose technologies by asking colleagues and looking for documentation,\ndemonstrated reliability, and technical support; adaptability via\ncustomizing or open-source standards was another priority. Research\ngroups commonly provide their own resources, but public/academic\ninstitutions and commercial services also provide substantial offerings.\nApplication creators and administrators welcome external services\nproviding guidance such as technology selection, sustainability\nplanning, evaluation, and specialized expertise (e.g., quality assurance\nand design). Technologies are diverse, so flexibility and ongoing\ncommunity input are essential, as is offering specific, easy-to-access\ntraining, community support, and professional development. Copyright (c)\n2015 John Wiley & Sons, Ltd.},
 bibtype = {article},
 author = {Lawrence, Katherine A. and Zentner, Michael and Wilkins-Diehr, Nancy and Wernert, Julie A. and Pierce, Marlon and Marru, Suresh and Michael, Scott},
 doi = {10.1002/cpe.3526},
 journal = {Concurrency Computation },
 number = {16}
}

Science gateways are digital interfaces to advanced technologies that\nsupport science/engineering research/education. Frequently implemented\nas Web and mobile applications, they provide access to community\nresources such as software, data, collaboration tools, instrumentation,\nand high-performance computing. We anticipate opportunities for growth\nwithin a fragmented community. Through a large-scale survey, we measured\nthe extent and characteristics of the gateway community (reliance on\ngateways and nature of existing resources) to understand useful services\nand support for builders and users. We administered an online survey to\nnearly 29,000 principal investigators, senior administrators, and people\nwith gateway affiliations. Nearly 5000 respondents represented diverse\nexpertise and geography. The majority of researchers/educators indicated\nthat specialized online resources were important to their work. They\nchoose technologies by asking colleagues and looking for documentation,\ndemonstrated reliability, and technical support; adaptability via\ncustomizing or open-source standards was another priority. Research\ngroups commonly provide their own resources, but public/academic\ninstitutions and commercial services also provide substantial offerings.\nApplication creators and administrators welcome external services\nproviding guidance such as technology selection, sustainability\nplanning, evaluation, and specialized expertise (e.g., quality assurance\nand design). Technologies are diverse, so flexibility and ongoing\ncommunity input are essential, as is offering specific, easy-to-access\ntraining, community support, and professional development. Copyright (c)\n2015 John Wiley & Sons, Ltd.

@inproceedings{
 title = {XCBC and XNIT - Tools for Cluster Implementation and Management in Research and Training},
 type = {inproceedings},
 year = {2015},
 pages = {857-864},
 volume = {2015-Octob},
 websites = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7307692},
 month = {9},
 publisher = {IEEE},
 id = {b7eaad5c-e325-3bcc-9fce-f79afbec2e17},
 created = {2019-10-01T18:06:11.076Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-04-29T16:40:54.779Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fischer2015},
 folder_uuids = {ec6ad3c6-db7d-494d-863c-ef38d23f1f7e,22c3b665-9e84-4884-8172-710aa9082eaf,b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 abstract = {© 2015 IEEE. The Extreme Science and Engineering Discovery Environment has created a suite of software collectively known as the XSEDE-compatible basic cluster (XCBC). It has been distributed as a Rocks Roll for some time. The same scientific and supporting packages are available as individual RPM packages as the XSEDE National Integration Toolkit (XNIT), so they can be downloaded and installed in portions as appropriate on existing clusters. In this paper, we examine using the LittleFe design created by the Earlham College Cluster Computing Group as a teaching tool to show the deployment of XCBC from Rocks. In addition, the demonstration of the commercial Limulus HPC200 Deskside Cluster solution is shown as a viable, off-the-shelf cluster that can be adapted to become an XSEDE-like cluster through the use of the XNIT repository. The goal is to demonstrate building practical XCBCs while showing that an XCBC need not be an expensive resource to be useful.},
 bibtype = {inproceedings},
 author = {Fischer, Jeremy and Coulter, Eric and Knepper, Richard and Peck, Charles and Stewart, Craig A.},
 doi = {10.1109/CLUSTER.2015.143},
 booktitle = {2015 IEEE International Conference on Cluster Computing}
}

© 2015 IEEE. The Extreme Science and Engineering Discovery Environment has created a suite of software collectively known as the XSEDE-compatible basic cluster (XCBC). It has been distributed as a Rocks Roll for some time. The same scientific and supporting packages are available as individual RPM packages as the XSEDE National Integration Toolkit (XNIT), so they can be downloaded and installed in portions as appropriate on existing clusters. In this paper, we examine using the LittleFe design created by the Earlham College Cluster Computing Group as a teaching tool to show the deployment of XCBC from Rocks. In addition, the demonstration of the commercial Limulus HPC200 Deskside Cluster solution is shown as a viable, off-the-shelf cluster that can be adapted to become an XSEDE-like cluster through the use of the XNIT repository. The goal is to demonstrate building practical XCBCs while showing that an XCBC need not be an expensive resource to be useful.

@inproceedings{
 title = {Building a Chemical-Protein Interactome on the Open Science Grid},
 type = {inproceedings},
 year = {2015},
 pages = {15-20},
 websites = {https://scitech.isi.edu/wordpress/wp-content/papercite-data/pdf/osg-splinter-2015.pdf},
 id = {73df60a8-0d64-3500-98ff-c5ae602c9e2e},
 created = {2020-04-23T05:29:20.343Z},
 accessed = {2020-04-22},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-04-29T16:40:54.775Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Quick2015},
 private_publication = {false},
 abstract = {The Structural Protein-Ligand Interactome (SPLINTER) project predicts the interaction of thousands of small molecules with thousands of proteins. These interactions are predicted using the three-dimensional structure of the bound complex between each pair of protein and compound that is predicted by molecular docking. These docking runs consist of millions of individual short jobs each lasting only minutes. However, computing resources to execute these jobs (which cumulatively take tens of millions of CPU hours) are not readily or easily available in a cost effective manner. By looking to National Cyberinfrastructure resources, and specifically the Open Science Grid (OSG), we have been able to harness CPU power for researchers at the Indiana University School of Medicine to provide a quick and efficient solution to their unmet computing needs. Using the job submission infrastructure provided by the OSG, the docking data and simulation executable was sent to more than 100 universities and research centers worldwide. These op-portunistic resources provided millions of CPU hours in a matter of days, greatly reducing time docking simulation time for the research group. The overall impact of this approach allows researchers to identify small molecule candidates for individual proteins, or new protein targets for existing FDA-approved drugs and biologically active compounds.},
 bibtype = {inproceedings},
 author = {Quick, Rob and Hayashi, Soichi and Meroueh, Samy and Rynge, Mats and Teige, Scott and Wang, Bo and Xu, David and Sinica, Academia and Taipei, Taiwan},
 doi = {https://doi.org/10.22323/1.239.0024},
 booktitle = {International Symposium on Grids and Clouds (ISGC)}
}

The Structural Protein-Ligand Interactome (SPLINTER) project predicts the interaction of thousands of small molecules with thousands of proteins. These interactions are predicted using the three-dimensional structure of the bound complex between each pair of protein and compound that is predicted by molecular docking. These docking runs consist of millions of individual short jobs each lasting only minutes. However, computing resources to execute these jobs (which cumulatively take tens of millions of CPU hours) are not readily or easily available in a cost effective manner. By looking to National Cyberinfrastructure resources, and specifically the Open Science Grid (OSG), we have been able to harness CPU power for researchers at the Indiana University School of Medicine to provide a quick and efficient solution to their unmet computing needs. Using the job submission infrastructure provided by the OSG, the docking data and simulation executable was sent to more than 100 universities and research centers worldwide. These op-portunistic resources provided millions of CPU hours in a matter of days, greatly reducing time docking simulation time for the research group. The overall impact of this approach allows researchers to identify small molecule candidates for individual proteins, or new protein targets for existing FDA-approved drugs and biologically active compounds.

@inproceedings{
 title = {Jetstream: A Distributed Cloud Infrastructure for Under Resourced Higher Education Communities},
 type = {inproceedings},
 year = {2015},
 keywords = {Atmosphere,Cloud,Cyberinfrastructure,Digital,EOT,EXtreme,Education,Globus,Jetstream,Outreach,Research,Training,XD,XSEDE},
 pages = {53-61},
 websites = {http://dl.acm.org/citation.cfm?doid=2753524.2753530},
 month = {6},
 publisher = {ACM Press},
 day = {16},
 city = {New York, New York, USA},
 id = {4804d94b-75b2-3afd-8ec8-bc3b5c5bba56},
 created = {2020-09-09T19:33:17.821Z},
 accessed = {2019-09-12},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:01.162Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fischer2015a},
 private_publication = {false},
 abstract = {Copyright © 2015 ACM. The US National Science Foundation (NSF) in 2015 awarded funding for a first-of-a-kind distributed cyberinfrastructure (DCI) system called Jetstream. Jetstream will be the NSF's first production cloud for general-purpose science and engineering research and education. Jetstream, scheduled for production in January 2016, will be based on the OpenStack cloud environment software with a menu-driven interface to make it easy for users to select a pre-composed Virtual Machine (VM) to perform a particular discipline-specific analysis. Jetstream will use the Atmosphere user interface developed as part of iPlant, providing a low barrier to use by practicing scientists, engineers, educators, and students, and Globus services from the University of Chicago for seamless integration into the national cyberinfrastructure fabric. The team implementing Jetstream has as their primary mission extending the reach of the NSF's eXtreme Digital (XD) program to researchers, educators, and research students who have not previously used NSF XD program resources, including those in communities and at institutions that traditionally lack significant cyberinfrastructure resources. We will, for example, use virtual Linux Desktops to deliver DCI capabilities supporting research and research education at small colleges and universities, including Historically Black Colleges and Universities (HBCUs), Minority Serving Institutions (MSIs), Tribal colleges, and higher education institutions in states designated by the NSF as eligible for funding via the Experimental Program to Stimulate Competitive Research (EPSCoR). Jetstream will be a novel distributed cyberinfrastructure, with production components in Indiana and Texas. In particular, Jetstream will deliver virtual Linux desktops to tablet devices and PDAs with reasonable responsiveness running over cellular networks. This paper will discuss design and application plans for Jetstream as a novel Distributed CyberInfrastructure system for research education.},
 bibtype = {inproceedings},
 author = {Fischer, Jeremy and Tuecke, Steven and Foster, Ian and Stewart, Craig A.},
 doi = {10.1145/2753524.2753530},
 booktitle = {Proceedings of the 1st Workshop on The Science of Cyberinfrastructure Research, Experience, Applications and Models - SCREAM '15}
}

Copyright © 2015 ACM. The US National Science Foundation (NSF) in 2015 awarded funding for a first-of-a-kind distributed cyberinfrastructure (DCI) system called Jetstream. Jetstream will be the NSF's first production cloud for general-purpose science and engineering research and education. Jetstream, scheduled for production in January 2016, will be based on the OpenStack cloud environment software with a menu-driven interface to make it easy for users to select a pre-composed Virtual Machine (VM) to perform a particular discipline-specific analysis. Jetstream will use the Atmosphere user interface developed as part of iPlant, providing a low barrier to use by practicing scientists, engineers, educators, and students, and Globus services from the University of Chicago for seamless integration into the national cyberinfrastructure fabric. The team implementing Jetstream has as their primary mission extending the reach of the NSF's eXtreme Digital (XD) program to researchers, educators, and research students who have not previously used NSF XD program resources, including those in communities and at institutions that traditionally lack significant cyberinfrastructure resources. We will, for example, use virtual Linux Desktops to deliver DCI capabilities supporting research and research education at small colleges and universities, including Historically Black Colleges and Universities (HBCUs), Minority Serving Institutions (MSIs), Tribal colleges, and higher education institutions in states designated by the NSF as eligible for funding via the Experimental Program to Stimulate Competitive Research (EPSCoR). Jetstream will be a novel distributed cyberinfrastructure, with production components in Indiana and Texas. In particular, Jetstream will deliver virtual Linux desktops to tablet devices and PDAs with reasonable responsiveness running over cellular networks. This paper will discuss design and application plans for Jetstream as a novel Distributed CyberInfrastructure system for research education.

@techreport{
 title = {Pervasive Technology Institute annual report: Research innovations and advanced cyberinfrastructure services in support of IU strategic goals during FY 2015},
 type = {techreport},
 year = {2015},
 keywords = {CACR,D2I,DSC,NCGAS,PTI,RT,Technical Report,advanced cyberinfrastructure,engagement,outreach,research,storage,students},
 websites = {http://hdl.handle.net/2022/20566},
 id = {90795fbd-2659-312d-a37e-41799152543e},
 created = {2020-09-10T14:22:29.844Z},
 accessed = {2020-09-10},
 file_attached = {true},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:00.305Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015d},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A.; and Plale, Beth; and Welch, Von; and Link, Matthew R.; and Miller, Therese; and Wernert, Eric A.; and Boyles, Michael J.; and Fulton, Ben; and Hancock, David Y.; Henschel, Robert; and Michael, Scott A.; and Pierce, Marlon; and Ping, Robert J.; and Gniady, Tassie; and Fox, Geoffrey C.; and Miksik, Gary;}
}

@techreport{
 title = {Indiana University’s advanced cyberinfrastructure in service of IU strategic goals: Activities of the Research Technologies Division of UITS and National Center for Genome Analysis Support – two Pervasive Technology Institute cyberinfrastructure and servi},
 type = {techreport},
 year = {2015},
 keywords = {ABITC,Clinical Affairs Schools,IUSM,NCGAS,PTI,advanced cyberinfrastructure,digital collections,engagement,health sciences,research,storage,students},
 websites = {https://scholarworks.iu.edu/dspace/handle/2022/19805},
 publisher = {Indiana University},
 id = {f42a904d-830e-364b-ba3d-f71cc0219692},
 created = {2020-09-10T16:50:22.920Z},
 accessed = {2020-09-10},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-15T22:44:00.201Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stewart2015g},
 folder_uuids = {3b35931e-fb6d-48f9-8e01-87ee16ef0331},
 private_publication = {false},
 bibtype = {techreport},
 author = {Stewart, Craig A.; and Plale, Beth; and Welch, Von; and Fox, Geoffrey C.; and Link, Matthew R.; and Miller, Therese; and Wernert, Eric A.; and Boyles, Michael J.; and Fulton, Ben; and Hancock, David Y.; Henschel, Robert; and Michael, Scott A.; and Pierce, Marlon; and Ping, Robert J.; and Miksik, Gary; and Gniady, Tassie;}
}

@article{
 title = {Enabling large scale next-generation sequence assembly with Blacklight},
 type = {article},
 year = {2014},
 id = {ccca790e-4f59-366e-adfe-c8ff279796a3},
 created = {2017-11-07T19:20:51.616Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:37.061Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {cpspspkmb14},
 source_type = {article},
 folder_uuids = {b4e18ac7-7050-4c86-a2a4-a9c35bdc74d4},
 private_publication = {false},
 bibtype = {article},
 author = {Couger, Brian M and Pipes, Lenore and Squina, Fabio and Prade, Rolf and Siepel, Adam and Palermo, Robert and Katze, Michael G and Mason, Christopher E and Blood, Philip D},
 doi = {10.1002/cpe.3231},
 journal = {CONCURRENCY AND COMPUTATION: PRACTICE AND EXPERIENCE}
}

@inproceedings{
 title = {Distributed control: priority scheduling for single source shortest paths without synchronization},
 type = {inproceedings},
 year = {2014},
 pages = {17-24},
 publisher = {IEEE Press},
 id = {5ae12bfc-3489-32bb-9b47-ddc2edc089ae},
 created = {2017-11-22T21:02:01.541Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:31.362Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Zalewski2014},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Zalewski, Marcin and Kanewala, Thejaka Amila and Firoz, Jesun Sahariar and Lumsdaine, Andrew},
 booktitle = {Proceedings of the 4th Workshop on Irregular Applications: Architectures and Algorithms}
}

@article{
 title = {Region-based memory management for GPU programming languages: enabling rich data structures on a spartan host},
 type = {article},
 year = {2014},
 pages = {141-155},
 volume = {49},
 publisher = {ACM},
 id = {56a8688d-3c66-3f58-9561-911a5cdf3b5f},
 created = {2017-11-22T21:02:01.664Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:30.930Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Holk2014},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {article},
 author = {Holk, Eric and Newton, Ryan and Siek, Jeremy and Lumsdaine, Andrew},
 journal = {ACM SIGPLAN Notices},
 number = {10}
}

@article{
 title = {Towards Automated Coding of Program Comprehension Gaze Data},
 type = {article},
 year = {2014},
 pages = {9},
 id = {a7e0d950-b76a-3a8e-9070-2fb14b554ad7},
 created = {2017-11-22T21:02:48.412Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:31.188Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Hansen2014},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 abstract = {Gaze data collected during program comprehension provides insight into programmers’ thought processes. Manual coding of this data, however, can be tedious and subjective. We de- fine and demonstrate an automated coding scheme for most categories in this workshop’s coding scheme. We discuss potential sources of error when abstracting from fixations to areas of interest and patterns, and consider alternative definitions for some codes. For the high-level Strategy cat- egory, we inform coding decisions with metrics computed over a rolling time window.},
 bibtype = {article},
 author = {Hansen, Michael and Goldstone, Robert L and Lumsdaine, Andrew},
 journal = {Eye Movements in Programming Education}
}

Gaze data collected during program comprehension provides insight into programmers’ thought processes. Manual coding of this data, however, can be tedious and subjective. We de- fine and demonstrate an automated coding scheme for most categories in this workshop’s coding scheme. We discuss potential sources of error when abstracting from fixations to areas of interest and patterns, and consider alternative definitions for some codes. For the high-level Strategy cat- egory, we inform coding decisions with metrics computed over a rolling time window.

@inproceedings{
 title = {Scoping Rules on a Platter},
 type = {inproceedings},
 year = {2014},
 pages = {59-70},
 id = {279670b8-e8ca-3c6a-a2d1-627ca93434ed},
 created = {2017-11-22T21:02:48.820Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:30.701Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Voufo2014a},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Voufo, Larisse and Zalewski, Marcin and Lumsdaine, Andrew},
 doi = {10.1145/2633628.2633633},
 booktitle = {Proceedings of the 10th ACM SIGPLAN workshop on Generic programming}
}

@article{
 title = {A dynamic execution model applied to distributed collision detection},
 type = {article},
 year = {2014},
 keywords = {Collision detection,Collision detection algorithm,Computer aided design,Concurrent threads,Distributed Memory,Dynamic management,Execution strategies,Global address spaces,Programming methodology,Scalability},
 pages = {470-477},
 volume = {8488 LNCS},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84903716715&doi=10.1007%2F978-3-319-07518-1_32&partnerID=40&md5=7d9007c6ad13f01ea4e2774ea551e1b3,https://www.scopus.com/inward/record.uri?eid=2-s2.0-84903795511&doi=10.1007%2F978-3-319-07518-1-32&partner},
 publisher = {Springer Verlag},
 city = {Leipzig},
 id = {f3193439-e9ec-31bd-b1f3-5e4d5a19ebc2},
 created = {2017-11-27T19:30:57.367Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:31.978Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Anderson2014470},
 source_type = {article},
 notes = {cited By 1; Conference of 29th International Supercomputing Conference, ISC 2014 ; Conference Date: 22 June 2014 Through 26 June 2014; Conference Code:105925},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 abstract = {The end of Dennard scaling and the looming Exascale challenges of efficiency, reliability, and scalability are driving a shift in programming methodologies away from conventional practices towards dynamic runtimes and asynchronous, data driven execution. Since Exascale machines are not yet available, however, experimental runtime systems and application co-design can expose application-specific overhead and scalability concerns at extreme scale, while also investigating the execution model defined by the runtime system itself. Such results may also contribute to the development of effective Exascale hardware. This work presents a case study evaluating a dynamic, Exascale-inspired execution model and its associated experimental runtime system consisting of lightweight concurrent threads with dynamic management in the context of a global address space examining the problem of mesh collision detection. This type of problem constitutes an essential component of many CAD systems and industrial crash applications. The core of the algorithm depends upon determining if two triangles intersect in three dimensions for large meshes. The resulting collision detection algorithm exploits distributed memory to enable extremely large mesh simulation and is shown to be scalable thereby lending support to the execution strategy. © 2014 Springer International Publishing.},
 bibtype = {article},
 author = {Anderson, M and Brodowicz, M and Dalessandro, L and DeBuhr, J and Sterling, T},
 doi = {10.1007/978-3-319-07518-1-32},
 journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}
}

The end of Dennard scaling and the looming Exascale challenges of efficiency, reliability, and scalability are driving a shift in programming methodologies away from conventional practices towards dynamic runtimes and asynchronous, data driven execution. Since Exascale machines are not yet available, however, experimental runtime systems and application co-design can expose application-specific overhead and scalability concerns at extreme scale, while also investigating the execution model defined by the runtime system itself. Such results may also contribute to the development of effective Exascale hardware. This work presents a case study evaluating a dynamic, Exascale-inspired execution model and its associated experimental runtime system consisting of lightweight concurrent threads with dynamic management in the context of a global address space examining the problem of mesh collision detection. This type of problem constitutes an essential component of many CAD systems and industrial crash applications. The core of the algorithm depends upon determining if two triangles intersect in three dimensions for large meshes. The resulting collision detection algorithm exploits distributed memory to enable extremely large mesh simulation and is shown to be scalable thereby lending support to the execution strategy. © 2014 Springer International Publishing.

@article{
 title = {Offloading MPI Parallel Prefix Scan (MPI_Scan) with the NetFPGA},
 type = {article},
 year = {2014},
 id = {24cff3e5-13f6-3363-bedb-f01ddfcb71fb},
 created = {2017-11-28T16:29:37.335Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:05.788Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Arap2014a},
 source_type = {JOUR},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 bibtype = {article},
 author = {Arap, Omer and Swany, Martin},
 journal = {arXiv preprint arXiv:1408.4939}
}

@article{
 title = {Special issue for emerging computational methods for the life sciences workshop},
 type = {article},
 year = {2014},
 volume = {26},
 id = {871f0b0b-4561-3683-9609-b1879de58773},
 created = {2017-11-28T17:32:48.497Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:29.183Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Qiu2014},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {The Special Issue of Concurrency and Computation: Practice and Experience 2013 deals with the latest trends in parallel and distributed high-performance systems applied to life science problems. Mitchel and co-researchers present parallel implementations of two popular microarray data analysis techniques, exploratory clustering analyses using the random forest classifier and feature selection through identification of differentially expressed genes using the rank product method. Luo and co-researchers describe an enhanced MapReduce-based programming model 'Map-Reduce-GlobalReduce', where the computations are expressed as three functions: Map, Reduce, and GlobalReduce. Jha and co-researchers present a runtime environment, Distributed Application Runtime Environment (DARE), that supports the scalable, flexible, and extensible composition of capabilities exploring the interoperability among heterogeneously distributed computing environments for pleasingly parallel applications.},
 bibtype = {article},
 author = {Qiu, J. and Foster, I. and Taylor, R.},
 doi = {10.1002/cpe.2998},
 journal = {Concurrency Computation Practice and Experience},
 number = {4}
}

The Special Issue of Concurrency and Computation: Practice and Experience 2013 deals with the latest trends in parallel and distributed high-performance systems applied to life science problems. Mitchel and co-researchers present parallel implementations of two popular microarray data analysis techniques, exploratory clustering analyses using the random forest classifier and feature selection through identification of differentially expressed genes using the rank product method. Luo and co-researchers describe an enhanced MapReduce-based programming model 'Map-Reduce-GlobalReduce', where the computations are expressed as three functions: Map, Reduce, and GlobalReduce. Jha and co-researchers present a runtime environment, Distributed Application Runtime Environment (DARE), that supports the scalable, flexible, and extensible composition of capabilities exploring the interoperability among heterogeneously distributed computing environments for pleasingly parallel applications.

@inproceedings{
 title = {Integrating pig with harp to support iterative applications with fast cache and customized communication},
 type = {inproceedings},
 year = {2014},
 id = {50cba507-c12b-38fa-b8d3-1bfbf928e235},
 created = {2017-11-28T17:32:48.518Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:29.410Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Wu2014},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {© 2014 IEEE. Use of high-level scripting languages to solve big data problems has become a mainstream approach for sophisticated machine learning data analysis. Often data must be used in several steps of a computation to complete a full task. Composing default data transformation operators with the standard Hadoop MapReduce runtime is very convenient. However, the current strategy of using high-level languages to support iterative applications with Hadoop MapReduce relies on an external wrapper script in other languages such as Python and Groovy, which causes significant performance loss when restarting mappers and reducers between jobs. In this paper, we reduce the extra job startup overheads by integrating Apache Pig with the high-performance Hadoop plug-in Harp developed at Indiana University. This provides fast data caching and customized communication patterns among iterations for data analysis. The results show performance improvements of factors from 2 to 5.},
 bibtype = {inproceedings},
 author = {Wu, T.-L. and Koppula, A. and Qiu, J.},
 doi = {10.1109/DataCloud.2014.8},
 booktitle = {Proceedings of DataCloud 2014: 5th International Workshop on Data Intensive Computing in the Clouds - Held in Conjunction with SC 2014: The International Conference for High Performance Computing, Networking, Storage and Analysis}
}

© 2014 IEEE. Use of high-level scripting languages to solve big data problems has become a mainstream approach for sophisticated machine learning data analysis. Often data must be used in several steps of a computation to complete a full task. Composing default data transformation operators with the standard Hadoop MapReduce runtime is very convenient. However, the current strategy of using high-level languages to support iterative applications with Hadoop MapReduce relies on an external wrapper script in other languages such as Python and Groovy, which causes significant performance loss when restarting mappers and reducers between jobs. In this paper, we reduce the extra job startup overheads by integrating Apache Pig with the high-performance Hadoop plug-in Harp developed at Indiana University. This provides fast data caching and customized communication patterns among iterations for data analysis. The results show performance improvements of factors from 2 to 5.

@inproceedings{
 title = {Towards an understanding of facets and exemplars of big data applications},
 type = {inproceedings},
 year = {2014},
 pages = {7-16},
 publisher = {ACM},
 id = {33166d5d-4605-37b9-a260-bddb16707ecc},
 created = {2017-11-28T17:33:36.803Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.719Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2014a},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Fox, Geoffrey Charles and Jha, Shantenu and Qiu, Judy and Luckow, Andre},
 booktitle = {Proceedings of the 20 Years of Beowulf Workshop on Honor of Thomas Sterling's 65th Birthday}
}

@inproceedings{
 title = {Integrating Pig with Harp to support iterative applications with fast cache and customized communication},
 type = {inproceedings},
 year = {2014},
 pages = {33-39},
 publisher = {IEEE Press},
 id = {4c760e28-9633-343e-97c0-239e9d9af5e7},
 created = {2017-11-28T17:33:37.351Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.084Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wu2014a},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Wu, Tak-Lon and Koppula, Abhilash and Qiu, Judy},
 booktitle = {Proceedings of the 5th International Workshop on Data-Intensive Computing in the Clouds}
}

@inproceedings{
 title = {Integrating the Apache Big Data Stack with HPC for Big Data},
 type = {inproceedings},
 year = {2014},
 id = {435c9d4e-f70f-3c49-a29c-8c97f0eec633},
 created = {2017-11-28T17:33:37.488Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.318Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fox2014},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Fox, Geoffrey Charles and Qiu, Judy and Jha, Shantenu},
 booktitle = {AGU Fall Meeting Abstracts}
}

@book{
 title = {Cloud Computing for Data-Intensive Applications},
 type = {book},
 year = {2014},
 volume = {1},
 publisher = {Springer},
 id = {86718d46-9a8f-3b0a-848e-54429d9d2c6b},
 created = {2017-11-28T17:33:39.330Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.594Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Li2014},
 source_type = {BOOK},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {book},
 author = {Li, Xiaolin and Qiu, Judy}
}

@article{
 title = {Towards HPC-ABDS: an initial high-performance big data stack},
 type = {article},
 year = {2014},
 pages = {18-21},
 volume = {1},
 id = {e00f3421-5c17-37bc-be6f-87f93bc9fc1c},
 created = {2017-11-28T17:33:39.632Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.161Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Qiu2014c},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {article},
 author = {Qiu, Judy and Jha, Shantenu and Luckow, Andre and Fox, Geoffrey Charles},
 journal = {Building Robust Big Data Ecosystem ISO/IEC JTC}
}

@article{
 title = {Comparative study on structure and correlation among author co-occurrence networks in bibliometrics},
 type = {article},
 year = {2014},
 pages = {1345-1360},
 volume = {101},
 publisher = {Springer},
 id = {428b9473-5f96-3dcc-8159-b6ff54d5adb9},
 created = {2017-11-28T17:33:39.677Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.310Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Qiu2014a},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {article},
 author = {Qiu, Jun-Ping and Dong, Ke and Yu, Hou-Qiang},
 journal = {Scientometrics},
 number = {2}
}

@inbook{
 type = {inbook},
 year = {2014},
 pages = {401-427},
 publisher = {Springer},
 id = {ddd1aa6c-f22e-30cd-8db3-8bf4e7cdc628},
 created = {2017-11-28T17:33:40.242Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:47.092Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gao2014a},
 source_type = {CHAP},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {inbook},
 author = {Gao, Xiaoming and Roth, Evan and McKelvey, Karissa and Davis, Clayton and Younge, Andrew and Ferrara, Emilio and Menczer, Filippo and Qiu, Judy},
 chapter = {Supporting a social media observatory with customizable index structures: architecture and performance},
 title = {Cloud Computing for Data-Intensive Applications}
}

@article{
 title = {Parallelizing Big Data Machine Learning Algorithms with Model Rotation},
 type = {article},
 year = {2014},
 id = {acdf14ff-7479-3c12-ab52-6ffb9b7797ad},
 created = {2017-11-28T17:33:40.573Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:26.551Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Zhang2014},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 bibtype = {article},
 author = {Zhang, Bingjing and Peng, Bo and Qiu, Judy},
 journal = {NIPS}
}

@article{
 title = {SLOWER: A performance model for Exascale computing},
 type = {article},
 year = {2014},
 pages = {42-57},
 volume = {1},
 id = {038a5471-ebea-37c5-96e4-de54be33b39b},
 created = {2017-12-18T17:23:37.585Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:02:56.650Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Sterling2014a},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {article},
 author = {Sterling, Thomas and Kogler, Daniel and Anderson, Matthew and Brodowicz, Maciej},
 journal = {Supercomputing frontiers and innovations},
 number = {2}
}

@inproceedings{
 title = {Towards exascale co-design in a runtime system},
 type = {inproceedings},
 year = {2014},
 pages = {85-99},
 publisher = {Springer},
 id = {7206342c-4618-3f04-8497-c434c07f6b38},
 created = {2017-12-18T17:23:37.762Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:02:57.177Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Sterling2014},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Sterling, Thomas and Anderson, Matthew and Bohan, P Kevin and Brodowicz, Maciej and Kulkarni, Abhishek and Zhang, Bo},
 booktitle = {International Conference on Exascale Applications and Software}
}

@inproceedings{
 title = {Abstract rendering: Out-of-core rendering for information visualization},
 type = {inproceedings},
 year = {2014},
 pages = {90170K},
 volume = {9017},
 publisher = {International Society for Optics and Photonics},
 id = {d7b8acb2-b2bd-3dc8-84e4-a8da6a0410c4},
 created = {2017-12-18T18:01:08.312Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:02:56.840Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 citation_key = {Cottam2014a},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Cottam, Joseph A and Lumsdaine, Andrew and Wang, Peter},
 booktitle = {Visualization and Data Analysis 2014}
}

@article{
 title = {Preserving Data While Rendering},
 type = {article},
 year = {2014},
 volume = {6},
 websites = {http://pjim.newschool.edu/issues/2014/01/pdfs/ParsonsJournalForInformationMapping_Joseph_Cottam.pdf},
 id = {96ca2bcd-ad84-3224-bd9f-4b0959ce7b8f},
 created = {2017-12-18T18:01:21.738Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:15.711Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Cottam2014},
 source_type = {JOUR},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 bibtype = {article},
 author = {Cottam, Joseph A and Wang, Peter},
 journal = {Parsons Journal for Information Mapping},
 number = {1}
}

@article{
 title = {The green bank northern celestial cap pulsar survey. I. Survey description, data analysis, and initial results},
 type = {article},
 year = {2014},
 pages = {67},
 volume = {791},
 publisher = {IOP Publishing},
 id = {d9c5bfa7-c9d7-3d4c-8a7e-5bc90137580b},
 created = {2017-12-18T21:43:39.452Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:28.850Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Stovall2014},
 source_type = {JOUR},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {article},
 author = {Stovall, K and Lynch, R S and Ransom, S M and Archibald, A M and Banaszak, S and Biwer, C M and Boyles, J and Dartez, L P and Day, D and Ford, A J},
 journal = {The Astrophysical Journal},
 number = {1}
}

@article{
 title = {Emerging Computational Methods for the Life Sciences Workshop 2012},
 type = {article},
 year = {2014},
 pages = {1231-1233},
 volume = {26},
 publisher = {Wiley Online Library},
 id = {c88082ec-031f-352b-95e6-36273544b542},
 created = {2017-12-18T21:44:03.911Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:43.824Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Qiu2014b},
 source_type = {JOUR},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {The Special Issue of Concurrency and Computation: Practice and Experience, August 2013, discusses papers presented at the Emerging Computational Methods for the Life Sciences Workshop (ECMLS2012). Weber and colleague note that GPUs and multicore processors are now pervasive in computational sciences and high-performance computing. Their high-arithmetic throughput and memory bandwidth combined with their ever increasing programmability make them suitable for a widening variety of applications. Yang and researchers study a difficulty in building a mechanistic model of biological systems coming from determination of correct parameter values. Their paper proposes a novel parameter estimation method to infer unknown parameters, such as kinetic rates, from noisy experimental observations. Elllingson and colleagues present the current state of high-throughput virtual screening. They describe a case study of using a task-parallel MPI version of Autodock4 to run a virtual high-throughput screen of one million compounds on the Jaguar Cray XK6 Supercomputer.},
 bibtype = {article},
 author = {Qiu, Judy and Foster, Ian and Goble, Carole},
 doi = {10.1002/cpe.3101},
 journal = {Concurrency and Computation: Practice and Experience},
 number = {6}
}

The Special Issue of Concurrency and Computation: Practice and Experience, August 2013, discusses papers presented at the Emerging Computational Methods for the Life Sciences Workshop (ECMLS2012). Weber and colleague note that GPUs and multicore processors are now pervasive in computational sciences and high-performance computing. Their high-arithmetic throughput and memory bandwidth combined with their ever increasing programmability make them suitable for a widening variety of applications. Yang and researchers study a difficulty in building a mechanistic model of biological systems coming from determination of correct parameter values. Their paper proposes a novel parameter estimation method to infer unknown parameters, such as kinetic rates, from noisy experimental observations. Elllingson and colleagues present the current state of high-throughput virtual screening. They describe a case study of using a task-parallel MPI version of Autodock4 to run a virtual high-throughput screen of one million compounds on the Jaguar Cray XK6 Supercomputer.

@inproceedings{
 title = {A tale of two data-intensive paradigms: Applications, abstractions, and architectures},
 type = {inproceedings},
 year = {2014},
 pages = {645-652},
 publisher = {IEEE},
 id = {9f8cd8c0-ef36-33d5-8d01-46b03af1d0e6},
 created = {2017-12-18T21:44:04.703Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.278Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Jha2014},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {© 2014 IEEE. Scientific problems that depend on processing largeamounts of data require overcoming challenges in multiple areas:managing large-scale data distribution, co-placement andscheduling of data with compute resources, and storing and transferringlarge volumes of data. We analyze the ecosystems of thetwo prominent paradigms for data-intensive applications, hereafterreferred to as the high-performance computing and theApache-Hadoop paradigm. We propose a basis, common terminologyand functional factors upon which to analyze the two approachesof both paradigms. We discuss the concept of 'Big DataOgres' and their facets as means of understanding and characterizingthe most common application workloads found acrossthe two paradigms. We then discuss the salient features of thetwo paradigms, and compare and contrast the two approaches.Specifically, we examine common implementation/approaches ofthese paradigms, shed light upon the reasons for their current'architecture' and discuss some typical workloads that utilizethem. In spite of the significant software distinctions, we believethere is architectural similarity. We discuss the potential integrationof different implementations, across the different levelsand components. Our comparison progresses from a fully qualitativeexamination of the two paradigms, to a semi-quantitativemethodology. We use a simple and broadly used Ogre (K-meansclustering), characterize its performance on a range of representativeplatforms, covering several implementations from bothparadigms. Our experiments provide an insight into the relativestrengths of the two paradigms. We propose that the set of Ogreswill serve as a benchmark to evaluate the two paradigms alongdifferent dimensions.},
 bibtype = {inproceedings},
 author = {Jha, Shantenu and Qiu, Judy and Luckow, Andre and Mantha, Pradeep and Fox, Geoffrey C G.C.},
 doi = {10.1109/BigData.Congress.2014.137},
 booktitle = {Big Data (BigData Congress), 2014 IEEE International Congress on}
}

© 2014 IEEE. Scientific problems that depend on processing largeamounts of data require overcoming challenges in multiple areas:managing large-scale data distribution, co-placement andscheduling of data with compute resources, and storing and transferringlarge volumes of data. We analyze the ecosystems of thetwo prominent paradigms for data-intensive applications, hereafterreferred to as the high-performance computing and theApache-Hadoop paradigm. We propose a basis, common terminologyand functional factors upon which to analyze the two approachesof both paradigms. We discuss the concept of 'Big DataOgres' and their facets as means of understanding and characterizingthe most common application workloads found acrossthe two paradigms. We then discuss the salient features of thetwo paradigms, and compare and contrast the two approaches.Specifically, we examine common implementation/approaches ofthese paradigms, shed light upon the reasons for their current'architecture' and discuss some typical workloads that utilizethem. In spite of the significant software distinctions, we believethere is architectural similarity. We discuss the potential integrationof different implementations, across the different levelsand components. Our comparison progresses from a fully qualitativeexamination of the two paradigms, to a semi-quantitativemethodology. We use a simple and broadly used Ogre (K-meansclustering), characterize its performance on a range of representativeplatforms, covering several implementations from bothparadigms. Our experiments provide an insight into the relativestrengths of the two paradigms. We propose that the set of Ogreswill serve as a benchmark to evaluate the two paradigms alongdifferent dimensions.

@techreport{
 title = {IceCube Cybersecurity Improvement Plan},
 type = {techreport},
 year = {2014},
 id = {67af314e-430d-3160-8b8d-5d2dabfcce9c},
 created = {2017-12-18T21:44:05.539Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.042Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Marsteller2014},
 source_type = {RPRT},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {techreport},
 author = {Marsteller, James and Heiland, Randy}
}

@article{
 title = {The cell behavior ontology: describing the intrinsic biological behaviors of real and model cells seen as active agents},
 type = {article},
 year = {2014},
 pages = {2367-2374},
 volume = {30},
 publisher = {Oxford University Press},
 id = {e962240c-e36b-3027-b8ea-fda6ab97ea29},
 created = {2017-12-18T21:44:05.766Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:45.724Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Sluka2014},
 source_type = {JOUR},
 folder_uuids = {089a8687-5c2e-4a40-91e2-0a855ea1eb95},
 private_publication = {false},
 bibtype = {article},
 author = {Sluka, James P and Shirinifard, Abbas and Swat, Maciej and Cosmanescu, Alin and Heiland, Randy W and Glazier, James A},
 journal = {Bioinformatics},
 number = {16}
}

@inproceedings{
 title = {PXFS: A persistent storage model for extreme scale},
 type = {inproceedings},
 year = {2014},
 keywords = {Algorithms; Digital storage; Energy efficiency; Fi,Computational performance; Massively parallels; M,Distributed computer systems},
 pages = {900-906},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84899566905&doi=10.1109%2FICCNC.2014.6785457&partnerID=40&md5=5328204bef92e19ea687fea548b4682d},
 publisher = {IEEE Computer Society},
 city = {Honolulu, HI},
 id = {8d0ef1c7-587a-3590-8c09-b33b2de309af},
 created = {2018-01-08T19:03:41.992Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:16.226Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Yang2014900},
 source_type = {conference},
 notes = {cited By 1; Conference of 2014 International Conference on Computing, Networking and Communications, ICNC 2014 ; Conference Date: 3 February 2014 Through 6 February 2014; Conference Code:104772},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e},
 private_publication = {false},
 abstract = {The continuing technological progress resulted in sustained increase in the number of transistors per chip as well as improved energy efficiency per FLOPS. This spurred a dramatic growth in aggregate computational performance of the largest supercomputing systems, yielding multiple Petascale implementations deployed in various locations over the world. Unfortunately, these advances did not translate to the required extent into accompanying I/O systems, which primarily saw the improvement in cumulative storage sizes required to match the ever expanding volume of scientific data sets, but little more in terms of architecture or effective access latency. Moreover, while new models of computations are formulated to handle the burden of efficiently structuring the parallel computations in anticipation of the arrival of Exascale systems, a meager progress is observed in the area of storage subsystems. New classes of algorithms developed for massively parallel applications, that gracefully handle the challenges of asynchrony, heavily multithreaded distributed codes, and message-driven computation, must be matched by similar advances in I/O methods and algorithms to produce a well performing and balanced supercomputing system. This paper discusses PXFS, a file system model for persistent objects inspired by the ParalleX model of execution that addresses many of these challenges. An early implementation of PXFS utilizing a well known Orange parallel file system as its back-end via asynchronous I/O layer is also described along with the preliminary performance data. The results show perfect scalability and 3x to 20x times speedup of I/O throughput performance comparing to OrangeFS user interface. Also the PXFS module on OrangeFS with 24 clients sees a 5x to 10x times more throughput than NFS. © 2014 IEEE.},
 bibtype = {inproceedings},
 author = {Yang, S and Brodowicz, M and Ligon III, W B and Kaiser, H},
 doi = {10.1109/ICCNC.2014.6785457},
 booktitle = {2014 International Conference on Computing, Networking and Communications, ICNC 2014}
}

The continuing technological progress resulted in sustained increase in the number of transistors per chip as well as improved energy efficiency per FLOPS. This spurred a dramatic growth in aggregate computational performance of the largest supercomputing systems, yielding multiple Petascale implementations deployed in various locations over the world. Unfortunately, these advances did not translate to the required extent into accompanying I/O systems, which primarily saw the improvement in cumulative storage sizes required to match the ever expanding volume of scientific data sets, but little more in terms of architecture or effective access latency. Moreover, while new models of computations are formulated to handle the burden of efficiently structuring the parallel computations in anticipation of the arrival of Exascale systems, a meager progress is observed in the area of storage subsystems. New classes of algorithms developed for massively parallel applications, that gracefully handle the challenges of asynchrony, heavily multithreaded distributed codes, and message-driven computation, must be matched by similar advances in I/O methods and algorithms to produce a well performing and balanced supercomputing system. This paper discusses PXFS, a file system model for persistent objects inspired by the ParalleX model of execution that addresses many of these challenges. An early implementation of PXFS utilizing a well known Orange parallel file system as its back-end via asynchronous I/O layer is also described along with the preliminary performance data. The results show perfect scalability and 3x to 20x times speedup of I/O throughput performance comparing to OrangeFS user interface. Also the PXFS module on OrangeFS with 24 clients sees a 5x to 10x times more throughput than NFS. © 2014 IEEE.

@inproceedings{
 title = {Supporting queries and analyses of large-scale social media data with customizable and scalable indexing techniques over NoSQL databases},
 type = {inproceedings},
 year = {2014},
 pages = {587-590},
 publisher = {IEEE},
 id = {ffbe3f99-df76-3e38-92d6-b9ea3d6f7ac7},
 created = {2018-01-09T20:30:41.927Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:44.469Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gao2014},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5},
 private_publication = {false},
 abstract = {Social media data analysis demonstrates two special characteristics in Big Data processing. First, most analyses focus on data subsets related to specific social events or activities instead of the whole dataset. Second, analysis workflows consist of multiple stages, and algorithms applied in each stage may use different computation and communication patterns depending on processing frameworks. This paper presents our efforts in supporting the data storage and processing requirements for such characteristics. To achieve efficient queries about target data subsets, we propose a general customizable and scalable indexing framework that can be built over distributed NoSQL databases. This framework allows users to define suitable customized index structures for their query patterns against social media data, and supports scalable indexing of both historical and streaming data. We implement this framework on HBase, and name it Indexed HBase. Starting from Indexed HBase, we build a distributed analysis stack based on YARN to support analysis algorithms using different processing frameworks, such as Hadoop MapReduce, Harp, and Giraph. This analysis stack is used to host the Truthy social media data observatory, and we have applied the customized index structures in supporting both query evaluation and sophisticated analysis algorithms. Performance tests show that our solutions outperform implementations using both direct raw data scans and current indexing mechanisms in existing NoSQL databases. © 2014 IEEE.},
 bibtype = {inproceedings},
 author = {Gao, Xiaoming and Qiu, Judy},
 doi = {10.1109/CCGrid.2014.57},
 booktitle = {Cluster, Cloud and Grid Computing (CCGrid), 2014 14th IEEE/ACM International Symposium on}
}

Social media data analysis demonstrates two special characteristics in Big Data processing. First, most analyses focus on data subsets related to specific social events or activities instead of the whole dataset. Second, analysis workflows consist of multiple stages, and algorithms applied in each stage may use different computation and communication patterns depending on processing frameworks. This paper presents our efforts in supporting the data storage and processing requirements for such characteristics. To achieve efficient queries about target data subsets, we propose a general customizable and scalable indexing framework that can be built over distributed NoSQL databases. This framework allows users to define suitable customized index structures for their query patterns against social media data, and supports scalable indexing of both historical and streaming data. We implement this framework on HBase, and name it Indexed HBase. Starting from Indexed HBase, we build a distributed analysis stack based on YARN to support analysis algorithms using different processing frameworks, such as Hadoop MapReduce, Harp, and Giraph. This analysis stack is used to host the Truthy social media data observatory, and we have applied the customized index structures in supporting both query evaluation and sophisticated analysis algorithms. Performance tests show that our solutions outperform implementations using both direct raw data scans and current indexing mechanisms in existing NoSQL databases. © 2014 IEEE.

@article{
 title = {Approaches to understanding the impact of technologies for aging in place: a mini-review},
 type = {article},
 year = {2014},
 pages = {282-288},
 volume = {60},
 publisher = {Karger Publishers},
 id = {9b2563c9-beee-3121-b68c-a91663bd565d},
 created = {2018-01-19T07:02:32.257Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:35.427Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Connelly2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {article},
 author = {Connelly, Kay and Mokhtari, Mounir and Falk, Tiago H},
 journal = {Gerontology},
 number = {3}
}

@techreport{
 title = {Routing events using the Grid Event Service},
 type = {techreport},
 year = {2014},
 websites = {dsc.soic.indiana.edu/publications/GESRouting-libre.pdf},
 id = {913ee26b-1a9e-37bf-b702-cf6a572ab1ff},
 created = {2018-01-19T07:02:33.002Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-09-09T20:16:18.811Z},
 read = {false},
 starred = {true},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Pallickara},
 source_type = {JOUR},
 folder_uuids = {44d48ba6-09b0-4f17-965c-b625aaaba978},
 private_publication = {false},
 bibtype = {techreport},
 author = {Pallickara, Shrideep and Fox, Geoffrey Charles}
}

@article{
 title = {Search for invisible decays of a Higgs boson produced in association with a Z boson in ATLAS},
 type = {article},
 year = {2014},
 pages = {201802},
 volume = {112},
 publisher = {APS},
 id = {5b344b39-1e0a-3772-ab37-b4ae9de4a026},
 created = {2018-01-19T07:02:33.605Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:37.976Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Aad2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {article},
 author = {Aad, Georges and Abajyan, T and Abbott, B and Abdallah, J and Khalek, S Abdel and Abdinov, O and Aben, R and Abi, B and Abolins, M and AbouZeid, O S},
 journal = {Physical Review Letters},
 number = {20}
}

@article{
 title = {Property Damage, Purchasing Orders, and Power Outages, Oh My!: Suggestions for Planning Your Next In-The-Wild Deployment},
 type = {article},
 year = {2014},
 pages = {1-14},
 publisher = {Springer},
 id = {a65a0a60-58ea-3fca-81ce-23e4e7768d8a},
 created = {2018-01-19T07:02:34.576Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:37.564Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Hazlewood2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {article},
 author = {Hazlewood, William R and Connelly, Kay and Caine, Kelly E and Schall-Zimmerman, Zachary and Blanton, Greg},
 journal = {Handbook of Smart Homes, Health Care and Well-Being}
}

@inproceedings{
 title = {Designing Aging-In-Place Technologies to Reflect the Lifestyles and Precious Artifacts of Urban and Rural Older Adults},
 type = {inproceedings},
 year = {2014},
 pages = {145-149},
 volume = {58},
 issue = {1},
 publisher = {SAGE Publications Sage CA: Los Angeles, CA},
 id = {35d93de3-6f2d-3eed-8ceb-1f7c3abf40e6},
 created = {2018-01-19T07:02:34.621Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:36.966Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {White2014a},
 source_type = {CONF},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {White, Ginger and Evans, Robyn and Connelly, Kay and Caine, Kelly},
 booktitle = {Proceedings of the Human Factors and Ergonomics Society Annual Meeting}
}

@article{
 title = {PM2: A Partitioning-Mining-Measuring Method for Identifying Progressive Changes in Older Adults’ Sleeping Activity},
 type = {article},
 year = {2014},
 pages = {205-228},
 volume = {5},
 publisher = {Hindawi Publishing Corporation},
 id = {98ee092d-171b-3620-a337-e4ab3e20345c},
 created = {2018-01-19T07:02:34.691Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:36.521Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Lin2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {article},
 author = {Lin, Qiang and Zhang, Daqing and Connelly, Kay and Zhou, Xingshe and Ni, Hongbo},
 journal = {Journal of healthcare engineering},
 number = {2}
}

@article{
 title = {Designing consumer health technologies for the treatment of patients with depression: A health practitioner's perspective},
 type = {article},
 year = {2014},
 volume = {3},
 publisher = {JMIR Publications Inc.},
 id = {c7f6230a-873c-3df8-a239-62acb3bbd9d8},
 created = {2018-01-19T07:02:34.862Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:34.926Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {White2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {article},
 author = {White, Ginger and Caine, Kelly and Connelly, Kay and Selove, Rebecca and Doub, Tom},
 journal = {Interactive journal of medical research},
 number = {1}
}

@article{
 title = {Privacy concerns in assisted living technologies},
 type = {article},
 year = {2014},
 pages = {75-88},
 volume = {69},
 publisher = {Springer},
 id = {3f3eb17d-8385-3c5b-9d2a-58bce0fb0238},
 created = {2018-01-19T07:02:34.916Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:37.177Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Garg2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {article},
 author = {Garg, Vaibhav and Camp, L Jean and Lorenzen-Huber, Lesa and Shankar, Kalpana and Connelly, Kay},
 journal = {annals of telecommunications-annales des télécommunications},
 number = {1-2}
}

@techreport{
 title = {David J. Crandall},
 type = {techreport},
 year = {2014},
 source = {EMBO Reports},
 pages = {131-133},
 volume = {15},
 id = {328bbe38-aac6-3c2b-98cd-81248db87720},
 created = {2018-01-19T07:02:35.432Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:35.179Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ahmed2014},
 source_type = {JOUR},
 folder_uuids = {230122f3-d462-4245-85c6-2d367f7eb43f},
 private_publication = {false},
 bibtype = {techreport},
 author = {Ahmed, Tousif and Hoyle, Roberto and Shaffer, Patrick and Connelly, Kay}
}

@inproceedings{
 title = {Scoping rules on a platter: A framework for understanding and specifying name binding},
 type = {inproceedings},
 year = {2014},
 pages = {59-70},
 publisher = {ACM},
 id = {a5bfa718-c4ef-3796-95f4-d847e4059947},
 created = {2018-01-23T20:26:56.241Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:20.973Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Voufo2014},
 source_type = {CONF},
 folder_uuids = {a0f5ac31-a393-4a7b-b7db-64a126a80f6e,2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 bibtype = {inproceedings},
 author = {Voufo, Larisse and Zalewski, Marcin and Lumsdaine, Andrew},
 booktitle = {Proceedings of the 10th ACM SIGPLAN workshop on Generic programming}
}

@inproceedings{
 title = {Towards a collective layer in the big data stack},
 type = {inproceedings},
 year = {2014},
 pages = {236-245},
 publisher = {IEEE},
 id = {ce8bebec-b646-343e-af5b-f23c37135ee2},
 created = {2018-01-23T20:26:56.597Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2020-05-11T14:43:40.054Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Gunarathne2014},
 source_type = {CONF},
 folder_uuids = {36d8ccf4-7085-47fa-8ab9-897283d082c5,023ec473-8727-4d72-856c-6687cf1caa00},
 private_publication = {false},
 abstract = {We generalize MapReduce, Iterative MapReduce and data intensive MPI runtime as a layered Map-Collective architecture with Map-All Gather, Map-All Reduce, MapReduce Merge Broadcast and Map-Reduce Scatter patterns as the initial focus. Map-collectives improve the performance and efficiency of the computations while at the same time facilitating ease of use for the users. These collective primitives can be applied to multiple runtimes and we propose building high performance robust implementations that cross cluster and cloud systems. Here we present results for two collectives shared between Hadoop (where we term our extension H-Collectives) on clusters and the Twister4Azure Iterative MapReduce for the Azure Cloud. Our prototype implementations of Map-All Gather and Map-All Reduce primitives achieved up to 33% performance improvement for K-means Clustering and up to 50% improvement for Multi-Dimensional Scaling, while also improving the user friendliness. In some cases, use of Map-collectives virtually eliminated almost all the overheads of the computations. © 2014 IEEE.},
 bibtype = {inproceedings},
 author = {Gunarathne, Thilina and Qiu, Judy and Gannon, Dennis},
 doi = {10.1109/CCGrid.2014.123},
 booktitle = {Cluster, Cloud and Grid Computing (CCGrid), 2014 14th IEEE/ACM International Symposium on}
}

We generalize MapReduce, Iterative MapReduce and data intensive MPI runtime as a layered Map-Collective architecture with Map-All Gather, Map-All Reduce, MapReduce Merge Broadcast and Map-Reduce Scatter patterns as the initial focus. Map-collectives improve the performance and efficiency of the computations while at the same time facilitating ease of use for the users. These collective primitives can be applied to multiple runtimes and we propose building high performance robust implementations that cross cluster and cloud systems. Here we present results for two collectives shared between Hadoop (where we term our extension H-Collectives) on clusters and the Twister4Azure Iterative MapReduce for the Azure Cloud. Our prototype implementations of Map-All Gather and Map-All Reduce primitives achieved up to 33% performance improvement for K-means Clustering and up to 50% improvement for Multi-Dimensional Scaling, while also improving the user friendliness. In some cases, use of Map-collectives virtually eliminated almost all the overheads of the computations. © 2014 IEEE.