Blocking strategy optimizations for sparse direct linear solver on heterogeneous architectures

Abstract : In the context of solving sparse linear systems, an ordering process partitions the matrix graph to minimize both fill-in and computational cost. We found that the ordering strategy used within supernodes might be enhanced to reduce the number of off-diagonal blocks, and then increases block sizes and kernel performance. This turns to be into the same complexity as the factorization algorithm, but allows for more efficient BLAS kernels. On the other side, supernodes that are too large need to be split to create more parallelism. The regular splitting strategy when applied locally impacts significantly the number of off-diagonal blocks and might have negative effect on the efficiency. In this talk, we present both a new strategy to improve supernodes ordering and splitting strategy that both enlarge the off-diagonal block sizes without changing the computational cost of the factorization. Performance improvement gains on the supernodal solver PaStiX are shown on multi-cores and heterogeneous architectures.
Type de document :
Communication dans un congrès
Sparse Days, Jun 2015, Saint Girons, France
Liste complète des métadonnées

https://hal.inria.fr/hal-01187881
Contributeur : Pierre Ramet <>
Soumis le : jeudi 27 août 2015 - 20:40:32
Dernière modification le : mercredi 11 juillet 2018 - 14:44:01

Identifiants

  • HAL Id : hal-01187881, version 1

Citation

Mathieu Faverge, Grégoire Pichon, Pierre Ramet, Jean Roman. Blocking strategy optimizations for sparse direct linear solver on heterogeneous architectures. Sparse Days, Jun 2015, Saint Girons, France. 〈hal-01187881〉

Partager

Métriques

Consultations de la notice

281