Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-6572
Main Title: Computable convergence bounds for GMRES
Author(s): Liesen, Jörg
Type: Article
Language Code: en
Abstract: The purpose of this paper is to derive new computable convergence bounds for GMRES. The new bounds depend on the initial guess and are thus conceptually different from standard "worst-case" bounds. Most importantly, approximations to the new bounds can be computed from information generated during the run of a certain GMRES implementation. The approximations allow predictions of how the algorithm will perform. Heuristics for such predictions are given. Numerical experiments illustrate the behavior of the new bounds as well as the use of the heuristics.
URI: https://depositonce.tu-berlin.de//handle/11303/7299
http://dx.doi.org/10.14279/depositonce-6572
Issue Date: 2006
Date Available: 20-Dec-2017
DDC Class: DDC::500 Naturwissenschaften und Mathematik::510 Mathematik::518 Numerische Analysis
Subject(s): linear systems
convergence analysis
GMRES method
Krylov subspace methods
iterative methods
Usage rights: Terms of German Copyright Law
Journal Title: SIAM Journal on Matrix Analysis and Applications
Publisher: Society for Industrial and Applied Mathematics
Publisher Place: Philadelphia, Pa
Volume: 21
Issue: 3
Publisher DOI: 10.1137/S0895479898341669
Page Start: 882
Page End: 903
Appears in Collections:Fachgebiet Numerische Lineare Algebra » Publications

Files in This Item:
File Description SizeFormat 
2006_Liesen_et-al.pdf415.8 kBAdobe PDFThumbnail
View/Open


Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.