Electronic poster: Performance studies of a molecular dynamics code. William, T., Berry, D., K., & Henschel, R. In SC'11 - Proceedings of the 2011 High Performance Computing Networking, Storage and Analysis Companion, Co-located with SC'11, pages 105-106, 2011.
Electronic poster: Performance studies of a molecular dynamics code [link]Website  doi  abstract   bibtex   
A molecular dynamics code simulating the diffusion in dense nuclear matter in white dwarf stars is analyzed in this collaboration between PTI (Indiana University) and ZIH (Technische Universität Dresden). The code is highly configurable allowing MPI, OpenMP, or hybrid runs and additional fine-tuning with a range of parameters. The first step in the code analysis is to identify the best performing parameter set of the serial version. This configuration represents the most promising candidate for further studies. Aim of the parallel analysis is then to measure the scalability limits of the different parallel code implementations and to detect bottlenecks possibly preventing higher parallel efficiency. This work has been done with the parallel analysis framework Vampir.
@inproceedings{
 title = {Electronic poster: Performance studies of a molecular dynamics code},
 type = {inproceedings},
 year = {2011},
 keywords = {Application programming interfaces (API),Code analysis,Configurable,Dresdens,Electronic,Scalability,Stars},
 pages = {105-106},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84859019048&doi=10.1145%2F2148600.2148654&partnerID=40&md5=43c9fa0c4131be323c148c2a14186a2d},
 city = {Seattle, WA},
 id = {297f4d22-cbd6-3ebf-a137-d3cd31083c60},
 created = {2019-10-01T17:20:35.949Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:25:35.381Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {William2011105},
 source_type = {conference},
 notes = {cited By 0; Conference of 2011 High Performance Computing Networking, Storage and Analysis, SC'11, Co-located with SC'11 ; Conference Date: 12 November 2011 Through 18 November 2011; Conference Code:89149},
 folder_uuids = {22c3b665-9e84-4884-8172-710aa9082eaf},
 private_publication = {false},
 abstract = {A molecular dynamics code simulating the diffusion in dense nuclear matter in white dwarf stars is analyzed in this collaboration between PTI (Indiana University) and ZIH (Technische Universität Dresden). The code is highly configurable allowing MPI, OpenMP, or hybrid runs and additional fine-tuning with a range of parameters. The first step in the code analysis is to identify the best performing parameter set of the serial version. This configuration represents the most promising candidate for further studies. Aim of the parallel analysis is then to measure the scalability limits of the different parallel code implementations and to detect bottlenecks possibly preventing higher parallel efficiency. This work has been done with the parallel analysis framework Vampir.},
 bibtype = {inproceedings},
 author = {William, T and Berry, D K and Henschel, R},
 doi = {10.1145/2148600.2148654},
 booktitle = {SC'11 - Proceedings of the 2011 High Performance Computing Networking, Storage and Analysis Companion, Co-located with SC'11}
}

Downloads: 0