Tile QR Factorization with Parallel Panel Processing for Multicore Architectures

Abstract : To exploit the potential of multicore architectures, recent dense linear algebra libraries have used tile algorithms, which consist in scheduling a Directed Acyclic Graph (DAG) of tasks of fine granularity where nodes represent tasks, either panel factorization or update of a block-column, and edges represent dependencies among them. Although past approaches already achieve high performance on moderate and large square matrices, their way of processing a panel in sequence leads to limited performance when factorizing tall and skinny matrices or small square matrices. We present a new fully asynchronous method for computing a QR factorization on shared-memory multicore architectures that overcomes this bottleneck. Our contribution is to adapt an existing algorithm that performs a panel factorization in parallel (named Communication-Avoiding QR and initially designed for distributed-memory machines), to the context of tile algorithms using asynchronous computations. An experimental study shows significant improvement (up to almost 10 times faster) compared to state-of-the-art approaches. We aim to eventually incorporate this work into the Parallel Linear Algebra for Scalable Multi-core Architectures (PLASMA) library.
Type de document :
Communication dans un congrès
24th IEEE International Parallel and Distributed Processing Symposium (IPDPS 2010), Apr 2010, Atlanta, United States. 2010
Liste complète des métadonnées


https://hal.inria.fr/inria-00548899
Contributeur : Emmanuel Agullo <>
Soumis le : lundi 20 décembre 2010 - 18:46:47
Dernière modification le : jeudi 10 septembre 2015 - 01:08:27
Document(s) archivé(s) le : lundi 21 mars 2011 - 03:42:29

Fichier

tile-caqr-ipdps.pdf
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

  • HAL Id : inria-00548899, version 1

Collections

Citation

Bilel Hadri, Hatem Ltaief, Emmanuel Agullo, Jack Dongarra. Tile QR Factorization with Parallel Panel Processing for Multicore Architectures. 24th IEEE International Parallel and Distributed Processing Symposium (IPDPS 2010), Apr 2010, Atlanta, United States. 2010. <inria-00548899>

Partager

Métriques

Consultations de
la notice

380

Téléchargements du document

332