Performance and Scalability of the Block Low-Rank Multifrontal Factorization on Multicore Architectures

Abstract : Matrices coming from elliptic Partial Differential Equations have been shown to have a low-rank property which can be efficiently exploited in multifrontal solvers to provide a substantial reduction of their complexity. Among the possible low-rank formats, the Block Low-Rank format (BLR) is easy to use in a general purpose multifrontal solver and its potential compared to standard (full-rank) solvers has been demonstrated. Recently, new variants have been introduced and it was proved that they can further reduce the complexity but their performance has never been analyzed. In this paper, we present a multithreaded BLR factorization, and analyze its efficiency and scalability in shared-memory multicore environments. We identify the challenges posed by the use of BLR approximations in multifrontal solvers and put forward several algorithmic variants of the BLR factorization that overcome these challenges by improving its efficiency and scalability. We illustrate the performance analysis of the BLR multifrontal factorization with numerical experiments on a large set of problems coming from a variety of real-life applications.
Type de document :
Rapport
[Research Report] INPT-IRIT; CNRS-IRIT; INRIA-LIP; UPS-IRIT. 2017
Liste complète des métadonnées

Littérature citée [39 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01505070
Contributeur : Theo Mary <>
Soumis le : mardi 11 avril 2017 - 00:56:16
Dernière modification le : jeudi 15 juin 2017 - 09:09:18
Document(s) archivé(s) le : mercredi 12 juillet 2017 - 12:34:01

Fichier

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

Identifiants

  • HAL Id : hal-01505070, version 1

Collections

Citation

Patrick Amestoy, Alfredo Buttari, Jean-Yves L 'Excellent, Théo Mary. Performance and Scalability of the Block Low-Rank Multifrontal Factorization on Multicore Architectures. [Research Report] INPT-IRIT; CNRS-IRIT; INRIA-LIP; UPS-IRIT. 2017. 〈hal-01505070〉

Partager

Métriques

Consultations de
la notice

324

Téléchargements du document

126