Impact of Blocking Strategies for Sparse Direct Solvers on Top of Generic Runtimes

Abstract : Among the preprocessing steps of a sparse direct solver, reordering and block symbolic factorization are two major steps to reach a suitable granularity for BLAS kernels efficiency and runtime management. In this talk, we present a reordering strategy to increase off-diagonal block sizes. It enhances BLAS kernels and allows to handle larger tasks, reducing runtime overhead. Finally, we will comment the resulting gain in the PaStiX solver implemented over StarPU and PaRSEC.
Type de document :
Communication dans un congrès
SIAM Conference on Computation Science and Engineering (CSE'17), Feb 2017, Atlanta, United States. 〈http://www.siam.org/meetings/cse17/〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01421384
Contributeur : Pierre Ramet <>
Soumis le : jeudi 22 décembre 2016 - 10:39:34
Dernière modification le : jeudi 11 janvier 2018 - 06:22:35

Identifiants

  • HAL Id : hal-01421384, version 1

Citation

Grégoire Pichon, Mathieu Faverge, Pierre Ramet, Jean Roman. Impact of Blocking Strategies for Sparse Direct Solvers on Top of Generic Runtimes. SIAM Conference on Computation Science and Engineering (CSE'17), Feb 2017, Atlanta, United States. 〈http://www.siam.org/meetings/cse17/〉. 〈hal-01421384〉

Partager

Métriques

Consultations de la notice

191