XKaapi: A Runtime System for Data-Flow Task Programming on Heterogeneous Architectures

Abstract : Most recent HPC platforms have heterogeneous nodes composed of multi-core CPUs and accelerators, like GPUs. Programming such nodes is typically based on a combination of OpenMP and CUDA/OpenCL codes; scheduling relies on a static partitioning and cost model. We present the XKaapi runtime system for data-flow task programming on multi-CPU and multi-GPU architectures, which supports a data-flow task model and a locality-aware work stealing scheduler. XKaapi enables task multi-implementation on CPU or GPU and multi-level parallelism with different grain sizes. We show performance results on two dense linear algebra kernels, matrix product (GEMM) and Cholesky factorization (POTRF), to evaluate XKaapi on a heterogeneous architecture composed of two hexa-core CPUs and eight NVIDIA Fermi GPUs. Our conclusion is two-fold. First, fine grained parallelism and online scheduling achieve performance results as good as static strategies, and in most cases outperform them. This is due to an improved work stealing strategy that includes locality information; a very light implementation of the tasks in XKaapi; and an optimized search for ready tasks. Next, the multi-level parallelism on multiple CPUs and GPUs enabled by XKaapi led to a highly efficient Cholesky factorization. Using eight NVIDIA Fermi GPUs and four CPUs, we measure up to 2.43 TFlop/s on double precision matrix product and 1.79 TFlop/s on Cholesky factorization; and respectively 5.09 TFlop/s and 3.92 TFlop/s in single precision.
Type de document :
Communication dans un congrès
27th IEEE International Parallel & Distributed Processing Symposium (IPDPS), May 2013, Boston, Massachusetts, United States. 2013
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00799904
Contributeur : Joao Vicente Ferreira Lima <>
Soumis le : mardi 12 mars 2013 - 18:41:34
Dernière modification le : jeudi 9 août 2018 - 14:42:02
Document(s) archivé(s) le : dimanche 2 avril 2017 - 11:48:16

Fichier

ipdps2013.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00799904, version 1

Collections

Citation

Thierry Gautier, Joao Vicente Ferreira Lima, Nicolas Maillard, Bruno Raffin. XKaapi: A Runtime System for Data-Flow Task Programming on Heterogeneous Architectures. 27th IEEE International Parallel & Distributed Processing Symposium (IPDPS), May 2013, Boston, Massachusetts, United States. 2013. 〈hal-00799904〉

Partager

Métriques

Consultations de la notice

2842

Téléchargements de fichiers

848