Parallel pivoting combined with parallel reduction and fill-in control (Q1124267)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4111893
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Parallel pivoting combined with parallel reduction and fill-in control |
scientific article; zbMATH DE number 4111893 |
Statements
Parallel pivoting combined with parallel reduction and fill-in control (English)
0 references
1989
0 references
The paper deals with parallel algorithms for LU decomposition of large sparse unsymmetric matrices on multiprocessor systems. A computational model considered is the shared-memory MIMD where the synchronization allows multiple updates. The pivoting strategy is incorporated in the triangularization algorithms whereby the numerical stability can be checked in the parallel process. Actual performance results of the decomposition on the HEP (heterogeneous element processor) are presented.
0 references
parallel reduction
0 references
fill-in control
0 references
parallel algorithms
0 references
LU decomposition
0 references
large sparse unsymmetric matrices
0 references
multiprocessor systems
0 references
shared-memory MIMD
0 references
pivoting strategy
0 references
triangularization algorithms
0 references
numerical stability
0 references
heterogeneous element processor
0 references
0.871954619884491
0 references
0.8682011365890503
0 references
0.8485767841339111
0 references
0.8423476219177246
0 references
