Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A technique for accelerating the convergence of restarted GMRES

Journal Article · · SIAM Journal on Matrix Analysis and Applications
OSTI ID:15020368
We have observed that the residual vectors at the end of each restart cycle of restarted GMRES often alternate direction in a cyclic fashion, thereby slowing convergence. We present a new technique for accelerating the convergence of restarted GMRES by disrupting this alternating pattern. The new algorithm resembles a full conjugate gradient method with polynomial preconditioning, and its implementation requires minimal changes to the standard restarted GMRES algorithm.
Research Organization:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA
Sponsoring Organization:
USDOE
DOE Contract Number:
W-7405-ENG-48
OSTI ID:
15020368
Report Number(s):
UCRL-JRNL-202883
Journal Information:
SIAM Journal on Matrix Analysis and Applications, Journal Name: SIAM Journal on Matrix Analysis and Applications Journal Issue: 4 Vol. 25
Country of Publication:
United States
Language:
English

Similar Records

Minimal residual method stronger than polynomial preconditioning
Conference · Sat Dec 31 04:00:00 UTC 1994 · OSTI ID:223856

Some comparison of restarted GMRES and QMR for linear and nonlinear problems
Conference · Sat Dec 31 04:00:00 UTC 1994 · OSTI ID:219562

Toward efficient polynomial preconditioning for GMRES
Journal Article · Thu Dec 30 00:00:00 UTC 2021 · Numerical Linear Algebra with Applications · OSTI ID:1838187