THIBAULT, Samuel; BROQUEDIS, François; GOGLIN, Brice; NAMYST, Raymond; WACRENIER, Pierre-André

doi:10.1007/978-3-540-69303-1_19

Métadonnées

Licence d’utilisation du document

THIBAULT, Samuel
Efficient runtime systems for parallel architectures [RUNTIME]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

BROQUEDIS, François
Efficient runtime systems for parallel architectures [RUNTIME]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

GOGLIN, Brice
Efficient runtime systems for parallel architectures [RUNTIME]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

Langue

Communication dans un congrès

Ce document a été publié dans

International Workshop on OpenMP (IWOMP), 2007-06-04, Beijing. 2007p. 148--159

Résumé en anglais

Exploiting the full computational power of always deeper hierarchical multiprocessor machines requires a very careful distribution of threads and data among the underlying non-uniform architecture. The emergence of multi-core chips and NUMA machines makes it important to minimize the number of remote memory accesses, to favor cache affinities, and to guarantee fast completion of synchronization steps. By using the BubbleSched platform as a threading backend for the GOMP OpenMP compiler, we are able to easily transpose affinities of thread teams into scheduling hints using abstractions called bubbles. We then propose a scheduling strategy suited to nested OpenMP parallelism. The resulting preliminary performance evaluations show an important improvement of the speedup on a typical NAS OpenMP benchmark application.< Réduire

Mots clés en anglais

SMP

OpenMP

Nested Parallelism

Hierarchical Thread Scheduling

Bubbles

Multi-Core

NUMA

SMP.

Métadonnées

Partager cette publication !

Licence d’utilisation du document

An Efficient OpenMP Runtime System for Hierarchical Architectures

Langue

Ce document a été publié dans

Résumé en anglais

Mots clés en anglais

URI

DOI

Origine

Unités de recherche