000809217 001__ 809217
000809217 005__ 20250314084114.0
000809217 020__ $$a0927-5452
000809217 0247_ $$2doi$$a10.3233/978-1-61499-621-7-63
000809217 0247_ $$2Handle$$a2128/11187
000809217 0247_ $$2WOS$$aWOS:000578348400007
000809217 037__ $$aFZJ-2016-02510
000809217 1001_ $$0P:(DE-HGF)0$$aKitayama, Itaru$$b0
000809217 1112_ $$aInternational Conference on Parallel Computing 2015$$cEdinburgh$$d2015-09-01 - 2015-09-04$$gParCo 2015$$wScotland
000809217 245__ $$aExecution Performance Analysis of the ABySS Genome Sequence Assembler using Scalasca on the K Computer
000809217 260__ $$bIOS Press$$c2016
000809217 29510 $$aParallel Computing: On the Road to Exascale
000809217 300__ $$a63-72
000809217 3367_ $$2ORCID$$aCONFERENCE_PAPER
000809217 3367_ $$033$$2EndNote$$aConference Paper
000809217 3367_ $$2BibTeX$$aINPROCEEDINGS
000809217 3367_ $$2DRIVER$$aconferenceObject
000809217 3367_ $$2DataCite$$aOutput Types/Conference Paper
000809217 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1463467200_27925
000809217 4900_ $$aAdvances in Parallel Computing$$v27
000809217 520__ $$aPerformance analysis of the ABySS genome sequence assembler (ABYSS-P) executing on the K computer with up to 8192 compute nodes is described which identified issues that limited scalability to less than 1024 compute nodes and required prohibitive message buffer memory with 16384 or more compute nodes. The open-source Scalasca toolset was employed to analyse executions, revealing the impact of massive amounts of MPI point-to-point communication used particularly for master/worker process coordination, and inefficient parallel file operations that manifest as waiting time at later MPI collective synchronisations and communications. Initial remediation via use of collective communication operations and alternate strategies for parallel file handling show large performance and scalability improvements, with partial executions validated on the full 82,944 compute nodes of the K computer.
000809217 536__ $$0G:(DE-HGF)POF3-511$$a511 - Computational Science and Mathematical Methods (POF3-511)$$cPOF3-511$$fPOF III$$x0
000809217 536__ $$0G:(DE-Juel-1)ATMLPP$$aATMLPP - ATML Parallel Performance (ATMLPP)$$cATMLPP$$x1
000809217 7001_ $$0P:(DE-Juel1)132302$$aWylie, Brian J. N.$$b1$$eCorresponding author$$ufzj
000809217 7001_ $$0P:(DE-HGF)0$$aMaeda, Toshiyuki$$b2
000809217 773__ $$a10.3233/978-1-61499-621-7-63
000809217 8564_ $$uhttps://juser.fz-juelich.de/record/809217/files/parco15.pdf$$yOpenAccess
000809217 8564_ $$uhttps://juser.fz-juelich.de/record/809217/files/parco15.gif?subformat=icon$$xicon$$yOpenAccess
000809217 8564_ $$uhttps://juser.fz-juelich.de/record/809217/files/parco15.jpg?subformat=icon-1440$$xicon-1440$$yOpenAccess
000809217 8564_ $$uhttps://juser.fz-juelich.de/record/809217/files/parco15.jpg?subformat=icon-180$$xicon-180$$yOpenAccess
000809217 8564_ $$uhttps://juser.fz-juelich.de/record/809217/files/parco15.jpg?subformat=icon-640$$xicon-640$$yOpenAccess
000809217 8564_ $$uhttps://juser.fz-juelich.de/record/809217/files/parco15.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000809217 909CO $$ooai:juser.fz-juelich.de:809217$$pdnbdelivery$$pVDB$$pdriver$$popen_access$$popenaire
000809217 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132302$$aForschungszentrum Jülich$$b1$$kFZJ
000809217 9131_ $$0G:(DE-HGF)POF3-511$$1G:(DE-HGF)POF3-510$$2G:(DE-HGF)POF3-500$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bKey Technologies$$lSupercomputing & Big Data$$vComputational Science and Mathematical Methods$$x0
000809217 9141_ $$y2016
000809217 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000809217 920__ $$lyes
000809217 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
000809217 980__ $$acontrib
000809217 980__ $$aVDB
000809217 980__ $$aUNRESTRICTED
000809217 980__ $$aI:(DE-Juel1)JSC-20090406
000809217 9801_ $$aFullTexts