학술논문

Iterative solution of linear systems.
Document Type
Proceedings Paper
Author
Freund, Roland W. (1-NASA9-AD) AMS Author Profile; Golub, Gene H. (1-STF-C) AMS Author Profile; Nachtigal, Noël M. (1-NASA9-AD) AMS Author Profile
Source
Acta numerica, 1992 (19920101), 57-100.
Subject
65 Numerical analysis -- 65F Numerical linear algebra
  65F10 Iterative methods for linear systems
Language
English
Abstract
In this paper, the authors give an excellent account of recent advances in the solution by iterative methods of large non-Hermitian linear systems. The main focus is on developments based on Krylov subspace (polynomial) methods. Emphasized in this survey are the generalized minimum residual algorithm and Lanczos-based Krylov methods related to the biconjugate gradient method and the OMR algorithm. Also considered are parameter dependent Krylov subspace algorithms that require information on the spectrum of the coefficient matrix. The use of preconditioning procedures is briefly discussed.MR

Online Access