Mostrar el registro sencillo del ítem

hal.structure.identifierReformulations based algorithms for Combinatorial Optimization [Realopt]
hal.structure.identifierLaboratoire Bordelais de Recherche en Informatique [LaBRI]
dc.contributor.authorEYRAUD-DUBOIS, Lionel
hal.structure.identifierReformulations based algorithms for Combinatorial Optimization [Realopt]
hal.structure.identifierSchool of Computer Science [Manchester]
dc.contributor.authorLAMBERT, Thomas
dc.date.accessioned2024-04-04T03:05:53Z
dc.date.available2024-04-04T03:05:53Z
dc.date.conference2018-08-13
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/193293
dc.description.abstractEnWe consider the problem of data allocation when performing matrix multiplication on a heterogeneous node, with multicores and GPUs. Classical (cyclic) allocations designed for homogeneous settings are not appropriate, but the advent of task-based runtime systems makes it possible to use more general allocations. Previous theoretical work has proposed square and cube partitioning algorithms aimed at minimizing data movement for matrix multiplication. We propose techniques to adapt these continuous square partitionings to allocating discrete tiles of a matrix, and strategies to adapt the static allocation at run-time. We use these techniques in an implementation of Matrix Multiplication based on the StarPU runtime system, and we show through extensive experiments that this implementation allows to consistently obtain a lower communication volume while improving slightly the execution time, compared to standard state-of-the-art dynamic strategies.
dc.description.sponsorshipSolveurs pour architectures hétérogènes utilisant des supports d'exécution - ANR-13-MONU-0007
dc.language.isoen
dc.subject.enMathematics of computing → Solvers
dc.subject.enComputing methodologies → Linear algebra algorithms
dc.subject.enParallel algorithms
dc.subject.enGeneral and reference → Performance
dc.title.enUsing Static Allocation Algorithms for Matrix Matrix Multiplication on Multicores and GPUs
dc.typeCommunication dans un congrès
dc.identifier.doi10.1145/3225058.3225066
dc.subject.halInformatique [cs]/Calcul parallèle, distribué et partagé [cs.DC]
bordeaux.hal.laboratoriesInstitut de Mathématiques de Bordeaux (IMB) - UMR 5251*
bordeaux.institutionUniversité de Bordeaux
bordeaux.institutionBordeaux INP
bordeaux.institutionCNRS
bordeaux.conference.titleICPP 2018 - 47th International Conference on Parallel Processing
bordeaux.countryUS
bordeaux.conference.cityEugene, OR
bordeaux.peerReviewedoui
hal.identifierhal-01670678
hal.version1
hal.invitednon
hal.proceedingsoui
hal.conference.end2018-08-16
hal.popularnon
hal.audienceInternationale
hal.origin.linkhttps://hal.archives-ouvertes.fr//hal-01670678v1
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.au=EYRAUD-DUBOIS,%20Lionel&LAMBERT,%20Thomas&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