Afficher la notice abrégée

Optimizing MPI Communication within large Multicore nodes with Kernel assistance

hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorMOREAUD, Stéphanie
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorGOGLIN, Brice
hal.structure.identifierMathematics and Computer Science Division [ANL] [MCS]
dc.contributor.authorGOODELL, David
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorNAMYST, Raymond
dc.contributor.editorIEEE
dc.date.accessioned2024-04-15T09:49:41Z
dc.date.available2024-04-15T09:49:41Z
dc.date.issued2010-04
dc.date.conference2010-04-19
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/198279
dc.description.abstractEnAs the number of cores per node increases in modern clusters, intra-node communication efficiency becomes critical to application performance. We present a study of the traditional double-copy model in MPICH2 and a kernel-assisted single-copy strategy with KNEM on different shared-memory hosts with up to 96 cores. We show that KNEM suffers less from process placement on these complex architectures. It improves throughput up to a factor of 2 for large messages for both point-to-point and collective operations, and significantly improves NPB execution time. We detail when to switch from one strategy to the other depending on the communication pattern and we show that \ioat copy offload only appears to be an interesting solution for older architectures.
dc.language.isoen
dc.title.enOptimizing MPI Communication within large Multicore nodes with Kernel assistance
dc.typeCommunication dans un congrès
dc.identifier.doi10.1109/IPDPSW.2010.5470849
dc.subject.halInformatique [cs]/Système d'exploitation [cs.OS]
bordeaux.page7 p.
bordeaux.hal.laboratoriesLaboratoire Bordelais de Recherche en Informatique (LaBRI) - UMR 5800*
bordeaux.institutionUniversité de Bordeaux
bordeaux.institutionBordeaux INP
bordeaux.institutionCNRS
bordeaux.conference.titleWorkshop on Communication Architecture for Clusters, held in conjunction with IPDPS 2010
bordeaux.countryUS
bordeaux.conference.cityAtlanta
bordeaux.peerReviewedoui
hal.identifierinria-00451471
hal.version1
hal.invitednon
hal.proceedingsoui
hal.conference.end2010-04-23
hal.popularnon
hal.audienceInternationale
hal.origin.linkhttps://hal.archives-ouvertes.fr//inria-00451471v1
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.date=2010-04&rft.spage=7%20p.&rft.epage=7%20p.&rft.au=MOREAUD,%20St%C3%A9phanie&GOGLIN,%20Brice&GOODELL,%20David&NAMYST,%20Raymond&rft.genre=unknown


Fichier(s) constituant ce document

FichiersTailleFormatVue

Il n'y a pas de fichiers associés à ce document.

Ce document figure dans la(les) collection(s) suivante(s)

Afficher la notice abrégée