论文信息 - Dynamic block GMRES: an iterative method for block linear systems

Dynamic block GMRES: an iterative method for block linear systems

We present variants of the block-GMRES($m$) algorithms due to Vital and the block-LGMRES($m$,$k$) by Baker, Dennis and Jessup, obtained with replacing the standard QR factorization by a rank-revealing QR factorization in the Arnoldi process. The resulting algorithm allows for dynamic block deflation whenever there is a linear dependency between the Krylov vectors or the convergence of a right-hand-side occurs. $\textsc{Fortran 90}$ implementations of the algorithms were tested on a number of test matrices and the results show that in some cases a substantial reduction of the execution time is obtained. Also a parallel implementation of our variant of the block-GMRES($m$) algorithm, using $\textsc{Fortran 90}$ and $\textsc{MPI}$ was tested on $\textsc{SunFire 15K}$ parallel computer, showing good parallel efficiency.

Rudnei Dias da Cunha | Dulcenéia Becker

[1] H. Walker. Implementation of the GMRES method using householder transformations , 1988 .

[2] Elizabeth R. Jessup,et al. A Technique for Accelerating the Convergence of Restarted GMRES , 2005, SIAM J. Matrix Anal. Appl..

[3] Rudnei Dias da Cunha,et al. New Parallel (Rank-Revealing) QR Factorization Algorithms , 2002, Euro-Par.

[4] P. Sonneveld. CGS, A Fast Lanczos-Type Solver for Nonsymmetric Linear systems , 1989 .

[5] B. Vital. Etude de quelques methodes de resolution de problemes lineaires de grande taille sur multiprocesseur , 1990 .

[6] T. Chan. Rank revealing QR factorizations , 1987 .

[7] E. Jessup,et al. AN EFFICIENT BLOCK VARIANT OF GMRES , 2022 .

[8] L. Foster. Rank and null space calculations using matrix decomposition without column interchanges , 1986 .

[9] Tim Hopkins,et al. The Parallel Iterative Methods (PIM) package for the solution of systems of linear equations on para , 1995 .

[10] Yousef Saad,et al. Iterative methods for sparse linear systems , 2003 .

[11] Gene H. Golub,et al. Matrix computations , 1983 .

[12] Guangye Li,et al. A Block Variant of the GMRES Method on Massively Parallel Processors , 1997, Parallel Comput..

[13] Christian H. Bischof,et al. Computing rank-revealing QR factorizations of dense matrices , 1998, TOMS.