Iterative methods for solution of linear system equations: From past to future (Q2747638)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Iterative methods for solution of linear system equations: From past to future |
scientific article; zbMATH DE number 1658094
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Iterative methods for solution of linear system equations: From past to future |
scientific article; zbMATH DE number 1658094 |
Statements
24 June 2003
0 references
large systems
0 references
iterative solver
0 references
steepest descent method
0 references
conjugate gradient method
0 references
Krylov method
0 references
GMRES
0 references
GMERR
0 references
survey paper
0 references
parallel computation
0 references
matrix-vector product
0 references
0 references
0.93886954
0 references
0.93402344
0 references
0 references
0.9241341
0 references
Iterative methods for solution of linear system equations: From past to future (English)
0 references
The authors give an overview on iterative methods for solving large systems of linear algebraic equations. At first, classical methods for systems with a symmetric, positive definite matrix (steepest descent method and conjugate gradient method) are presented. Then, it is explained how modern iterative methods can be constructed by using the basic ideas of the classical methods. The Krylov methods GMRES and GMERR are discussed more in detail. Furthermore, the parallelization of the iterative solvers is studied. Especially, the parallel computation of the matrix-vector product, one of the basic operations in all iterative methods, is explained in detail.
0 references
