Mostrar el registro sencillo del ítem

StarPU-MPI: Task Programming over Clusters of Machines Enhanced with Accelerators

hal.structure.identifierNVIDIA [NVIDIA]
dc.contributor.authorAUGONNET, Cédric
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorAUMAGE, Olivier
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorFURMENTO, Nathalie
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorNAMYST, Raymond
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
hal.structure.identifierEfficient runtime systems for parallel architectures [RUNTIME]
dc.contributor.authorTHIBAULT, Samuel
dc.contributor.editorJesper Larsson Träff, Siegfried Benkner and Jack Dongarra
dc.date.accessioned2024-04-15T09:54:12Z
dc.date.available2024-04-15T09:54:12Z
dc.date.issued2012
dc.date.conference2012-09-23
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/198654
dc.description.abstractEnGPUs clusters are becoming widespread HPC platforms. Ex- ploiting them is however challenging, as this requires two separate paradigms (MPI and CUDA or OpenCL) and careful load balancing due to node heterogeneity. Current paradigms usually either limit themselves to of- fload part of the computation and leave CPUs idle, or require static CPU/GPU work partitioning. We thus have previously proposed StarPU, a runtime system able to dynamically scheduling tasks within a single heterogeneous node. We show how we extended the task paradigm of StarPU with MPI to easily map the task graph on MPI clusters and automatically benefit from optimized execution.
dc.language.isoen
dc.publisherSpringer
dc.rights.urihttp://creativecommons.org/licenses/by/
dc.subject.enAccelerators
dc.subject.enGPUs
dc.subject.enMPI
dc.subject.enTask-based model
dc.subject.enAccelerators
dc.title.enStarPU-MPI: Task Programming over Clusters of Machines Enhanced with Accelerators
dc.typeCommunication dans un congrès
dc.subject.halInformatique [cs]/Calcul parallèle, distribué et partagé [cs.DC]
bordeaux.volume7490
bordeaux.hal.laboratoriesLaboratoire Bordelais de Recherche en Informatique (LaBRI) - UMR 5800*
bordeaux.institutionUniversité de Bordeaux
bordeaux.institutionBordeaux INP
bordeaux.institutionCNRS
bordeaux.conference.titleEuroMPI 2012 - The 19th European MPI Users' Group Meeting
bordeaux.countryAT
bordeaux.conference.cityVienna
bordeaux.peerReviewedoui
hal.identifierhal-00725477
hal.version1
hal.invitednon
hal.proceedingsoui
hal.conference.end2012-09-26
hal.popularnon
hal.audienceInternationale
hal.origin.linkhttps://hal.archives-ouvertes.fr//hal-00725477v1
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.date=2012&rft.volume=7490&rft.au=AUGONNET,%20C%C3%A9dric&AUMAGE,%20Olivier&FURMENTO,%20Nathalie&NAMYST,%20Raymond&THIBAULT,%20Samuel&rft.genre=unknown


Archivos en el ítem

ArchivosTamañoFormatoVer

No hay archivos asociados a este ítem.

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem