Towards an efficient tile matrix inversion of symmetric positive definite matrices on multicore architectures

Abstract : The algorithms in the current sequential numerical linear algebra libraries (e.g. LAPACK) do not parallelize well on multicore architectures. A new family of algorithms, the tile algorithms, has recently been introduced. Previous research has shown that it is possible to write efficient and scalable tile algorithms for performing a Cholesky factorization, a (pseudo) LU factorization, a QR factorization, and computing the inverse of a symmetric positive denite matrix. In this extended abstract, we revisit the computation of the inverse of a symmetric positive denite matrix. We observe that, using a dynamic task scheduler, it is relatively painless to translate existing LAPACK code to obtain a ready-to-be-executed tile algorithm. However we demonstrate that, for some variants, non trivial compiler techniques (array renaming, loop reversal and pipelining) need then to be applied to further increase the parallelism of the application. We present preliminary experimental results.
Type de document :
Communication dans un congrès
9th International Meeting on High Performance Computing for Computational Science (VecPar'10), Jun 2010, Berkeley, United States. 2010
Liste complète des métadonnées


https://hal.inria.fr/inria-00548906
Contributeur : Emmanuel Agullo <>
Soumis le : lundi 20 décembre 2010 - 19:33:59
Dernière modification le : jeudi 10 septembre 2015 - 01:08:30
Document(s) archivé(s) le : lundi 5 novembre 2012 - 14:41:36

Fichier

vecpar10.pdf
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

  • HAL Id : inria-00548906, version 1

Collections

Citation

Emmanuel Agullo, Henricus Bouwmeester, Jack Dongarra, Jakub Kurzak, Julien Langou, et al.. Towards an efficient tile matrix inversion of symmetric positive definite matrices on multicore architectures. 9th International Meeting on High Performance Computing for Computational Science (VecPar'10), Jun 2010, Berkeley, United States. 2010. <inria-00548906>

Partager

Métriques

Consultations de
la notice

301

Téléchargements du document

204