An MPI implementation of the SPAI preconditioner on the T3E
Abstract
The authors describe and test spai_1.1, a parallel MPIimplementation of the sparse approximate inverse (SPAI) preconditioner.They show that SPAI can be very effective for solving a set of very largeand difficult problems on a Cray T3E. The results clearly show the valueof SPAI (and approximate inverse methods in general) as the viablealternative to ILU-type methods when facing very large and difficultproblems. The authorsstrengthen this conclusion by showing that spai_1.1also has very good scaling behavior.
- Authors:
- Publication Date:
- Research Org.:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Org.:
- USOER Director. Office of Computational and TechnologyResearch. Division of Mathematical Information and ComputationalSciences, Laboratory Directed Research and Development
- OSTI Identifier:
- 899790
- Report Number(s):
- LBNL-40794
- DOE Contract Number:
- DE-AC02-05CH11231
- Resource Type:
- Journal Article
- Journal Name:
- International Journal of High Performance ComputingApplications
- Additional Journal Information:
- Journal Volume: 13; Journal Issue: 2; Related Information: Journal Publication Date: 1999
- Country of Publication:
- United States
- Language:
- English
Citation Formats
Barnard, Stephen T, Bernardo, Luis M, and Simon, Horst D. An MPI implementation of the SPAI preconditioner on the T3E. United States: N. p., 1997.
Web.
Barnard, Stephen T, Bernardo, Luis M, & Simon, Horst D. An MPI implementation of the SPAI preconditioner on the T3E. United States.
Barnard, Stephen T, Bernardo, Luis M, and Simon, Horst D. 1997.
"An MPI implementation of the SPAI preconditioner on the T3E". United States. https://www.osti.gov/servlets/purl/899790.
@article{osti_899790,
title = {An MPI implementation of the SPAI preconditioner on the T3E},
author = {Barnard, Stephen T and Bernardo, Luis M and Simon, Horst D},
abstractNote = {The authors describe and test spai_1.1, a parallel MPIimplementation of the sparse approximate inverse (SPAI) preconditioner.They show that SPAI can be very effective for solving a set of very largeand difficult problems on a Cray T3E. The results clearly show the valueof SPAI (and approximate inverse methods in general) as the viablealternative to ILU-type methods when facing very large and difficultproblems. The authorsstrengthen this conclusion by showing that spai_1.1also has very good scaling behavior.},
doi = {},
url = {https://www.osti.gov/biblio/899790},
journal = {International Journal of High Performance ComputingApplications},
number = 2,
volume = 13,
place = {United States},
year = {Mon Sep 08 00:00:00 EDT 1997},
month = {Mon Sep 08 00:00:00 EDT 1997}
}
Other availability
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.