Abstract
A number of techniques are described for solving sparse linear systems on parallel platforms. The general approach used is a domai n-decomposition type method in which a processor is assigned a certain number of rows of the linear system to be solved. Strategies that are discussed include non-standard graph partitioners, and a forced loadbalance technique for the local iterations. A common practice when partitioning a graph is to seek to minimize the number of cut-edges and to have an equal number of equations per processor. It is shown that partitioners that take into account the values of the matrix entries may be more effective.
| Original language | English (US) |
|---|---|
| Title of host publication | Parallel Computation - 4th International ACPC Conference Including Special Tracks on Parallel Numerics (ParNum 1999) and Parallel Computing in Image Processing, Video Processing, and Multimedia, Proceedings |
| Editors | Peter Zinterhof, Marian Vajteršic, Andreas Uhl |
| Publisher | Springer Verlag |
| Pages | 13-27 |
| Number of pages | 15 |
| ISBN (Print) | 3540656413, 9783540656418 |
| DOIs | |
| State | Published - 1999 |
| Event | 4th International ACPC Conference on Parallel Computation, ACPC 1999 - Salzburg, Austria Duration: Feb 16 1999 → Feb 18 1999 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 1557 |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Other
| Other | 4th International ACPC Conference on Parallel Computation, ACPC 1999 |
|---|---|
| Country/Territory | Austria |
| City | Salzburg |
| Period | 2/16/99 → 2/18/99 |
Bibliographical note
Publisher Copyright:© Springer-Verlag Berlin Heidelberg 1999.