Loading…

369 Tflop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer

We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten for the heterogeneous Roadrunner supercomputer. Each Roadrunner compute node consists of two AMD Opteron dualcore microprocessors and four PowerXCell 8i enhanced Cell mic...

Full description

Saved in:
Bibliographic Details
Main Authors: Swaminarayan, S., Germann, T.C., Kadau, K., Fossum, G.C.
Format: Conference Proceeding
Language:English
Subjects:
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 10
container_issue
container_start_page 1
container_title
container_volume
creator Swaminarayan, S.
Germann, T.C.
Kadau, K.
Fossum, G.C.
description We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten for the heterogeneous Roadrunner supercomputer. Each Roadrunner compute node consists of two AMD Opteron dualcore microprocessors and four PowerXCell 8i enhanced Cell microprocessors, so that there are four MPI ranks per node, each with one Opteron and one Cell. The interatomic forces are computed on the Cells (each with one PPU and eight SPU cores), while the Opterons are used to direct inter-rank communication and perform I/O-heavy periodic analysis, visualization, and checkpointing tasks. The performance measured for our initial implementation of a standard Lennard-Jones pair potential benchmark reached a peak of 369 Tflop/s double-precision floating-point performance on the full Roadrunner system (27.7% of peak), corresponding to 124 MFlop/Watt/s at a price of approximately 3.69 MFlops/dollar. We demonstrate an initial target application, the jetting and ejection of material from a shocked surface.
doi_str_mv 10.1109/SC.2008.5214713
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_5214713</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>5214713</ieee_id><sourcerecordid>5214713</sourcerecordid><originalsourceid>FETCH-LOGICAL-i1313-f18319a1b99cd88c754a20e1cc57fc4e41a296b166e1681a8f26557c3549cdab3</originalsourceid><addsrcrecordid>eNo9kEtvwjAQhN0HUoFy7qEX_4GA128fK9SXhFSppWdkzKakSuLIJgf-fYNKO4cd6RvNHJaQO2BzAOYWH8s5Z8zOFQdpQFyQCUguJbdCwSUZc9CmkEKYKzJzxv5lkl__Z9yNyOS04ZhhoG7ILOdvNkgqIZgZk73Qjq7LOnaLTJtYY-hrn-ju2PqmCpnmqhnAoYptprGlhz3S9-h3qW9bTPQLh-vroutTFzPSPR4wxRON_dDtO0whNl0_0FsyKn2dcXb2Kfl8elwvX4rV2_Pr8mFVVCBAFCVYAc7D1rmwszYYJT1nCCEoUwaJEjx3egtaI2gL3pZcK2WCUHIo-K2Ykvvf3QoRN12qGp-Om_MHxQ-6k17i</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>369 Tflop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer</title><source>IEEE Xplore All Conference Series</source><creator>Swaminarayan, S. ; Germann, T.C. ; Kadau, K. ; Fossum, G.C.</creator><creatorcontrib>Swaminarayan, S. ; Germann, T.C. ; Kadau, K. ; Fossum, G.C.</creatorcontrib><description>We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten for the heterogeneous Roadrunner supercomputer. Each Roadrunner compute node consists of two AMD Opteron dualcore microprocessors and four PowerXCell 8i enhanced Cell microprocessors, so that there are four MPI ranks per node, each with one Opteron and one Cell. The interatomic forces are computed on the Cells (each with one PPU and eight SPU cores), while the Opterons are used to direct inter-rank communication and perform I/O-heavy periodic analysis, visualization, and checkpointing tasks. The performance measured for our initial implementation of a standard Lennard-Jones pair potential benchmark reached a peak of 369 Tflop/s double-precision floating-point performance on the full Roadrunner system (27.7% of peak), corresponding to 124 MFlop/Watt/s at a price of approximately 3.69 MFlops/dollar. We demonstrate an initial target application, the jetting and ejection of material from a shocked surface.</description><identifier>ISSN: 2167-4329</identifier><identifier>ISBN: 9781424428342</identifier><identifier>ISBN: 1424428343</identifier><identifier>EISSN: 2167-4337</identifier><identifier>EISBN: 1424428351</identifier><identifier>EISBN: 9781424428359</identifier><identifier>DOI: 10.1109/SC.2008.5214713</identifier><identifier>LCCN: 2008907015</identifier><language>eng</language><publisher>IEEE</publisher><subject>Clustering algorithms ; Collaboration ; Computer architecture ; Coprocessors ; Laboratories ; Memory management ; Microprocessors ; Permission ; Postal services ; Supercomputers</subject><ispartof>2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, 2008, p.1-10</ispartof><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/5214713$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54555,54920,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/5214713$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Swaminarayan, S.</creatorcontrib><creatorcontrib>Germann, T.C.</creatorcontrib><creatorcontrib>Kadau, K.</creatorcontrib><creatorcontrib>Fossum, G.C.</creatorcontrib><title>369 Tflop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer</title><title>2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis</title><addtitle>SC</addtitle><description>We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten for the heterogeneous Roadrunner supercomputer. Each Roadrunner compute node consists of two AMD Opteron dualcore microprocessors and four PowerXCell 8i enhanced Cell microprocessors, so that there are four MPI ranks per node, each with one Opteron and one Cell. The interatomic forces are computed on the Cells (each with one PPU and eight SPU cores), while the Opterons are used to direct inter-rank communication and perform I/O-heavy periodic analysis, visualization, and checkpointing tasks. The performance measured for our initial implementation of a standard Lennard-Jones pair potential benchmark reached a peak of 369 Tflop/s double-precision floating-point performance on the full Roadrunner system (27.7% of peak), corresponding to 124 MFlop/Watt/s at a price of approximately 3.69 MFlops/dollar. We demonstrate an initial target application, the jetting and ejection of material from a shocked surface.</description><subject>Clustering algorithms</subject><subject>Collaboration</subject><subject>Computer architecture</subject><subject>Coprocessors</subject><subject>Laboratories</subject><subject>Memory management</subject><subject>Microprocessors</subject><subject>Permission</subject><subject>Postal services</subject><subject>Supercomputers</subject><issn>2167-4329</issn><issn>2167-4337</issn><isbn>9781424428342</isbn><isbn>1424428343</isbn><isbn>1424428351</isbn><isbn>9781424428359</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2008</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNo9kEtvwjAQhN0HUoFy7qEX_4GA128fK9SXhFSppWdkzKakSuLIJgf-fYNKO4cd6RvNHJaQO2BzAOYWH8s5Z8zOFQdpQFyQCUguJbdCwSUZc9CmkEKYKzJzxv5lkl__Z9yNyOS04ZhhoG7ILOdvNkgqIZgZk73Qjq7LOnaLTJtYY-hrn-ju2PqmCpnmqhnAoYptprGlhz3S9-h3qW9bTPQLh-vroutTFzPSPR4wxRON_dDtO0whNl0_0FsyKn2dcXb2Kfl8elwvX4rV2_Pr8mFVVCBAFCVYAc7D1rmwszYYJT1nCCEoUwaJEjx3egtaI2gL3pZcK2WCUHIo-K2Ykvvf3QoRN12qGp-Om_MHxQ-6k17i</recordid><startdate>200811</startdate><enddate>200811</enddate><creator>Swaminarayan, S.</creator><creator>Germann, T.C.</creator><creator>Kadau, K.</creator><creator>Fossum, G.C.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>200811</creationdate><title>369 Tflop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer</title><author>Swaminarayan, S. ; Germann, T.C. ; Kadau, K. ; Fossum, G.C.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i1313-f18319a1b99cd88c754a20e1cc57fc4e41a296b166e1681a8f26557c3549cdab3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2008</creationdate><topic>Clustering algorithms</topic><topic>Collaboration</topic><topic>Computer architecture</topic><topic>Coprocessors</topic><topic>Laboratories</topic><topic>Memory management</topic><topic>Microprocessors</topic><topic>Permission</topic><topic>Postal services</topic><topic>Supercomputers</topic><toplevel>online_resources</toplevel><creatorcontrib>Swaminarayan, S.</creatorcontrib><creatorcontrib>Germann, T.C.</creatorcontrib><creatorcontrib>Kadau, K.</creatorcontrib><creatorcontrib>Fossum, G.C.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Swaminarayan, S.</au><au>Germann, T.C.</au><au>Kadau, K.</au><au>Fossum, G.C.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>369 Tflop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer</atitle><btitle>2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis</btitle><stitle>SC</stitle><date>2008-11</date><risdate>2008</risdate><spage>1</spage><epage>10</epage><pages>1-10</pages><issn>2167-4329</issn><eissn>2167-4337</eissn><isbn>9781424428342</isbn><isbn>1424428343</isbn><eisbn>1424428351</eisbn><eisbn>9781424428359</eisbn><abstract>We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten for the heterogeneous Roadrunner supercomputer. Each Roadrunner compute node consists of two AMD Opteron dualcore microprocessors and four PowerXCell 8i enhanced Cell microprocessors, so that there are four MPI ranks per node, each with one Opteron and one Cell. The interatomic forces are computed on the Cells (each with one PPU and eight SPU cores), while the Opterons are used to direct inter-rank communication and perform I/O-heavy periodic analysis, visualization, and checkpointing tasks. The performance measured for our initial implementation of a standard Lennard-Jones pair potential benchmark reached a peak of 369 Tflop/s double-precision floating-point performance on the full Roadrunner system (27.7% of peak), corresponding to 124 MFlop/Watt/s at a price of approximately 3.69 MFlops/dollar. We demonstrate an initial target application, the jetting and ejection of material from a shocked surface.</abstract><pub>IEEE</pub><doi>10.1109/SC.2008.5214713</doi><tpages>10</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 2167-4329
ispartof 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, 2008, p.1-10
issn 2167-4329
2167-4337
language eng
recordid cdi_ieee_primary_5214713
source IEEE Xplore All Conference Series
subjects Clustering algorithms
Collaboration
Computer architecture
Coprocessors
Laboratories
Memory management
Microprocessors
Permission
Postal services
Supercomputers
title 369 Tflop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T15%3A46%3A08IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=369%20Tflop/s%20molecular%20dynamics%20simulations%20on%20the%20Roadrunner%20general-purpose%20heterogeneous%20supercomputer&rft.btitle=2008%20SC%20-%20International%20Conference%20for%20High%20Performance%20Computing,%20Networking,%20Storage%20and%20Analysis&rft.au=Swaminarayan,%20S.&rft.date=2008-11&rft.spage=1&rft.epage=10&rft.pages=1-10&rft.issn=2167-4329&rft.eissn=2167-4337&rft.isbn=9781424428342&rft.isbn_list=1424428343&rft_id=info:doi/10.1109/SC.2008.5214713&rft.eisbn=1424428351&rft.eisbn_list=9781424428359&rft_dat=%3Cieee_CHZPO%3E5214713%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i1313-f18319a1b99cd88c754a20e1cc57fc4e41a296b166e1681a8f26557c3549cdab3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=5214713&rfr_iscdi=true