A linear system solver based on a modified Krylov subspace method for breakdown recovery (Q1921331): Difference between revisions
From MaRDI portal
Changed an Item |
Changed an Item |
||
Property / describes a project that uses | |||
Property / describes a project that uses: Harwell-Boeing sparse matrix collection / rank | |||
Normal rank |
Revision as of 07:48, 29 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A linear system solver based on a modified Krylov subspace method for breakdown recovery |
scientific article |
Statements
A linear system solver based on a modified Krylov subspace method for breakdown recovery (English)
0 references
11 August 1996
0 references
The authors present a scheme based on a modified Krylov subspace approach to Lanczos' biorthogonalization method of the solution of nonsymmetric linear systems. When the scheme breaks down it is replaced by a modified dual Krylov subspace which is the sum of the original subspace and a new Krylov subspace which depends on a new start vector and the original matrix. Based on this strategy, they develop a practical algorithm for solving linear systems (the MLAN/QM algorithm) which also incorporates the residual quasi-minimization proposed by \textit{R. W. Freund} and \textit{N. M. Nachtigal} [Numer. Math. 60, No. 3, 315-339 (1991; Zbl 0754.65034)]. The algorithm is presented in some computational detail. Convergence bounds for the method are presented as well as numerical results to show its effectiveness.
0 references
Krylov subspace method
0 references
breakdown recovery
0 references
convergence
0 references
Lanczos' biorthogonalization method
0 references
nonsymmetric linear systems
0 references
MLAN/QM algorithm
0 references
residual quasi-minimization
0 references
algorithm
0 references
numerical results
0 references