Vers des supports d'exécution capables d'exploiter les machines multicœurs hétérogènes
hal.structure.identifier | Efficient runtime systems for parallel architectures [RUNTIME] | |
hal.structure.identifier | Laboratoire Bordelais de Recherche en Informatique [LaBRI] | |
dc.contributor.author | AUGONNET, Cédric | |
dc.date.accessioned | 2024-04-15T09:55:32Z | |
dc.date.available | 2024-04-15T09:55:32Z | |
dc.date.issued | 2008 | |
dc.identifier.uri | https://oskar-bordeaux.fr/handle/20.500.12278/198764 | |
dc.description.abstractEn | Approaching the theoretical performance of heterogeneous multicore architectures, equipped with specialized accelerators, is a challenging issue. Unlike regular CPUs that can transparently access the whole global memory address range, accelerators usually embed local memory on which they perform all their computations using a specific instruction set. While many research efforts have been devoted to offloading parts of a program over such coprocessors, the real challenge is to find a programming model providing a unified view of all available computing units. In this document, we present an original runtime system providing a high-level, unified execution model allowing seamless execution of tasks over the underlying heterogeneous hardware. The runtime is based on a hierarchical memory management facility and on a codelet scheduler. We demonstrate the efficiency of our solution with a LU decomposition for both homogeneous (3.8 speedup on 4 cores) and heterogeneous machines (95% efficiency). We also show that a "granularity aware" scheduling can improve execution time by 35%. | |
dc.language.iso | fr | |
dc.title | Vers des supports d'exécution capables d'exploiter les machines multicœurs hétérogènes | |
dc.type | Rapport | |
dc.subject.hal | Informatique [cs]/Système d'exploitation [cs.OS] | |
dc.subject.hal | Informatique [cs]/Calcul parallèle, distribué et partagé [cs.DC] | |
bordeaux.page | 48 | |
bordeaux.hal.laboratories | Laboratoire Bordelais de Recherche en Informatique (LaBRI) - UMR 5800 | * |
bordeaux.institution | Université de Bordeaux | |
bordeaux.institution | Bordeaux INP | |
bordeaux.institution | CNRS | |
bordeaux.type.report | tu | |
hal.identifier | inria-00289361 | |
hal.version | 1 | |
hal.audience | Non spécifiée | |
hal.origin.link | https://hal.archives-ouvertes.fr//inria-00289361v1 | |
bordeaux.COinS | ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.title=Vers%20des%20supports%20d'ex%C3%A9cution%20capables%20d'exploiter%20les%20machines%20multic%C5%93urs%20h%C3%A9t%C3%A9rog%C3%A8nes&rft.atitle=Vers%20des%20supports%20d'ex%C3%A9cution%20capables%20d'exploiter%20les%20machines%20multic%C5%93urs%20h%C3%A9t%C3%A9rog%C3%A8nes&rft.date=2008&rft.spage=48&rft.epage=48&rft.au=AUGONNET,%20C%C3%A9dric&rft.genre=unknown |
Fichier(s) constituant ce document
Fichiers | Taille | Format | Vue |
---|---|---|---|
Il n'y a pas de fichiers associés à ce document. |