| HAL-Inria | Publications, software ... of Inria's scientists |
Reports
Using Postordering and Static Symbolic Factorization for Parallel Sparse LU
Abstract : In this report we present several improvements of widely used parallel LU factorization methods on sparse matrices. First we characterize the L, U factors in terms of their corresponding LU elimination forest. This characterization can be used as a compact storage scheme of the matrix as well as of the task dependence graph. To improve the use of BLAS in the numerical factorization, we perform a postorder traversal of the LU eforest thus obtaining larger supernodes. To expose more task parallelism for a sparse matrix, we build a more accurate task dependence graph that includes only the least necessary dependencies. Experiments compared favorably our methods against methods implemented in the S* environment on the SGI's Origin2000 multiprocessor.
https://hal.inria.fr/inria-00069935
Contributor : Rapport de Recherche Inria <>
Submitted on : Friday, May 19, 2006 - 6:37:20 PM
Last modification on : Friday, February 26, 2021 - 3:28:07 PM
Long-term archiving on: : Saturday, April 3, 2010 - 9:42:53 PM
Contributor : Rapport de Recherche Inria <>
Submitted on : Friday, May 19, 2006 - 6:37:20 PM
Last modification on : Friday, February 26, 2021 - 3:28:07 PM
Long-term archiving on: : Saturday, April 3, 2010 - 9:42:53 PM