A class of Lanczos-like algorithms implemented on parallel computers (Q1179222): Difference between revisions
From MaRDI portal
Changed an Item |
Set profile property. |
||
| Property / MaRDI profile type | |||
| Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Latest revision as of 23:34, 4 March 2024
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | A class of Lanczos-like algorithms implemented on parallel computers |
scientific article |
Statements
A class of Lanczos-like algorithms implemented on parallel computers (English)
0 references
26 June 1992
0 references
At first the standard Lanczos algorithm for computing extreme eigenvalues of symmetric matrices and then its parallel implementation (without reorthogonalization) on a hypercube computer and a CRAY-2 are discussed. The inner products in the standard algorithm cannot be performed in parallel. The Lanczos algorithm is reconstructed to gain better performance. Finally an s-step method is developed which is a new version of the Lanczos method. Some numerical examples are given.
0 references
sparse matrix
0 references
extreme eigenvalues
0 references
symmetric matrices
0 references
parallel implementation
0 references
hypercube computer
0 references
CRAY-2
0 references
performance
0 references
s-step method
0 references
numerical examples
0 references
Lanczos algorithm
0 references
