Implementation study of the Parallel Sparse LU Factorization - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2000

Implementation study of the Parallel Sparse LU Factorization

Résumé

This paper describes an efficient parallel implementation of our previously published LU factorization method for sparse matrices. The main achievement is the usage of a message-passing paradigm instead of the already implemented shared-memory solver. A necessary presentation of the theoretical context is included first, then we briefly show the symbolic factorization steps (build LU elimination forest, postorder traversal, supernode identification) and the numerical factorization steps (Factor and Update steps and algorithm). The experiments were run on an SGI Origin2000 multiprocessor with 64 nodes.

Domaines

Autre [cs.OH]
Fichier non déposé

Dates et versions

inria-00099237 , version 1 (26-09-2006)

Identifiants

  • HAL Id : inria-00099237 , version 1

Citer

Michel Cosnard, Laura Grigori. Implementation study of the Parallel Sparse LU Factorization. International Conference on Parallel & Distributed Computing and Systems - PDCS'2000, Nov 2000, none, 6 p. ⟨inria-00099237⟩
55 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More