Implementation and analysis of nonblocking collective operations on SCI networks. Kaiser, C., Hoefler, T., Bierbaum, B., & Bemmerl, T. In IPDPS 2009 - Proceedings of the 2009 IEEE International Parallel and Distributed Processing Symposium, 2009.
Implementation and analysis of nonblocking collective operations on SCI networks [link]Website  doi  abstract   bibtex   
Nonblocking collective communication operations are currently being considered for inclusion into the MPI stan-dard and are an area of active research. The benefits of such operations are documented by several recent publications, but so far, research concentrates on InfiniBand clusters. This paper describes an implementation of nonblocking collectives for clusters with the Scalable Coherent Interface (SCI) interconnect. We use synthetic and application kernel benchmarks to show that with nonblocking functions for collective communication performance enhancements can be achieved on SCI systems. Our results indicate that for the implementation of these nonblocking collectives data trans-fer methods other than those usually used for the blocking version should be considered to realize such improvements. © 2009 IEEE.
@inproceedings{
 title = {Implementation and analysis of nonblocking collective operations on SCI networks},
 type = {inproceedings},
 year = {2009},
 keywords = {Benchmarking,Collective communication operations; Collective co,Distributed parameter networks; Large scale syste},
 websites = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-70450092016&doi=10.1109%2FIPDPS.2009.5160892&partnerID=40&md5=8797ed83a7b38142aaffa6d7d5da8c07},
 city = {Rome},
 id = {6842b6b8-c0e5-39ee-bb24-51681273823a},
 created = {2018-01-09T20:30:38.296Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2018-03-12T19:03:18.857Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Kaiser2009},
 source_type = {conference},
 notes = {cited By 1; Conference of 23rd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2009 ; Conference Date: 23 May 2009 Through 29 May 2009; Conference Code:78504},
 folder_uuids = {2aba6c14-9027-4f47-8627-0902e1e2342b},
 private_publication = {false},
 abstract = {Nonblocking collective communication operations are currently being considered for inclusion into the MPI stan-dard and are an area of active research. The benefits of such operations are documented by several recent publications, but so far, research concentrates on InfiniBand clusters. This paper describes an implementation of nonblocking collectives for clusters with the Scalable Coherent Interface (SCI) interconnect. We use synthetic and application kernel benchmarks to show that with nonblocking functions for collective communication performance enhancements can be achieved on SCI systems. Our results indicate that for the implementation of these nonblocking collectives data trans-fer methods other than those usually used for the blocking version should be considered to realize such improvements. © 2009 IEEE.},
 bibtype = {inproceedings},
 author = {Kaiser, C and Hoefler, T and Bierbaum, B and Bemmerl, T},
 doi = {10.1109/IPDPS.2009.5160892},
 booktitle = {IPDPS 2009 - Proceedings of the 2009 IEEE International Parallel and Distributed Processing Symposium}
}

Downloads: 0