On short recurrence Krylov type methods for linear systems with many right-hand sides
From MaRDI portal
Publication:5964591
DOI10.1016/J.CAM.2015.11.040zbMATH Open1382.65092arXiv1504.04395OpenAlexW2963059781MaRDI QIDQ5964591FDOQ5964591
Authors: S. Rashedi, Ghodrat Ebadi, Sebastian Birk, Andreas Frommer 
Publication date: 29 February 2016
Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)
Abstract: Block and global Krylov subspace methods have been proposed as methods adapted to the situation where one iteratively solves systems with the same matrix and several right hand sides. These methods are advantageous, since they allow to cast the major part of the arithmetic in terms of matrix-block vector products, and since, in the block case, they take their iterates from a potentially richer subspace. In this paper we consider the most established Krylov subspace methods which rely on short recurrencies, i.e. BiCG, QMR and BiCGStab. We propose modifications of their block variants which increase numerical stability, thus at least partly curing a problem previously observed by several authors. Moreover, we develop modifications of the "global" variants which almost halve the number of matrix-vector multiplications. We present a discussion as well as numerical evidence which both indicate that the additional work present in the block methods can be substantial, and that the new "economic" versions of the "global" BiCG and QMR method can be considered as good alternatives to the BiCGStab variants.
Full work available at URL: https://arxiv.org/abs/1504.04395
Recommendations
- Matrix Krylov subspace methods for linear systems with multiple right-hand sides
- Convergence properties of some block Krylov subspace methods for multiple linear systems
- A hybrid block GMRES method for nonsymmetric systems with multiple right-hand sides
- Algebraic properties of the block GMRES and block Arnoldi methods
- A block QMR algorithm for non-Hermitian linear systems with multiple right-hand sides
Computational methods for sparse matrices (65F50) Iterative numerical methods for linear systems (65F10) Numerical computation of solutions to systems of equations (65H10)
Cites Work
- The University of Florida sparse matrix collection
- QMR: A quasi-minimal residual method for non-Hermitian linear systems
- Bi-CGSTAB: A Fast and Smoothly Converging Variant of Bi-CG for the Solution of Nonsymmetric Linear Systems
- BiCGstab(\(l\)) for linear equations involving unsymmetric matrices with complex spectrum
- Title not available (Why is that?)
- A set of level 3 basic linear algebra subprograms
- A new family of global methods for linear systems with multiple right-hand sides
- The block conjugate gradient algorithm and related methods
- A block version of BiCGSTAB for linear systems with multiple right-hand sides
- Global FOM and GMRES algorithms for matrix equations
- Convergence properties of block GMRES and matrix polynomials
- The global Hessenberg and CMRH methods for linear systems with multiple right-hand sides
- A block QMR algorithm for non-Hermitian linear systems with multiple right-hand sides
- Oblique projection methods for linear systems with multiple right-hand sides
- Generalized global conjugate gradient squared algorithm
- An Iterative Method for Nonsymmetric Systems with Multiple Right-Hand Sides
- Krylov-subspace methods for the Sylvester equation
- The block least squares method for solving nonsymmetric linear systems with multiple right-hand sides
- Adaptive aggregation-based domain decomposition multigrid for the lattice Wilson-Dirac operator
- A Lanczos-type method for multiple starting vectors
- Two-Sided Arnoldi and Nonsymmetric Lanczos Algorithms
- Block BiCGGR: A new block Krylov subspace method for computing high accuracy solutions
- A deflated conjugate gradient method for multiple right hand sides and multiple shifts
- Retooling the method of block conjugate gradients
- A Stabilized QMR Version of Block BiCG
- Global SCD algorithm for real positive definite linear systems with multiple right-hand sides
- A hybrid block GMRES method for nonsymmetric systems with multiple right-hand sides
- Increasing the performance of the Jacobi-Davidson method by blocking
Cited In (13)
- Two new variants of the simpler block GMRES method with vector deflation and eigenvalue deflation for multiple linear systems
- Block BiCGGR: A new block Krylov subspace method for computing high accuracy solutions
- Block minimum perturbation algorithm based on block Arnoldi process for nonsymmetric linear systems with multiple right-hand sides
- Title not available (Why is that?)
- Short-term recurrence Krylov subspace methods for nearly Hermitian matrices
- Block Krylov subspace methods for functions of matrices. II: Modified block FOM
- Hardware-oriented Krylov methods for high-performance computing
- Algebraic properties of the block GMRES and block Arnoldi methods
- Optimal size of the block in block GMRES on GPUs: computational model and experiments
- Matrix Krylov subspace methods for linear systems with multiple right-hand sides
- Degree-based function index for graphs with given diameter
- Global symplectic Lanczos method with application to matrix exponential approximation
- The block grade of a block Krylov space
Uses Software
This page was built for publication: On short recurrence Krylov type methods for linear systems with many right-hand sides
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5964591)
