Improved seed methods for symmetric positive definite linear equations with multiple right‐hand sides

SUMMARY We consider symmetric positive definite systems of linear equations with multiple right-hand sides. The seed conjugate gradient (CG) method solves one right-hand side with the CG method and simultaneously projects over the Krylov subspace thus developed for the other right-hand sides. Then the next system is solved and used to seed the remaining ones. Rounding error in the CG method limits how much the seeding can improve convergence. We propose three changes to the seed CG method: only the first right-hand side is used for seeding, this system is solved past convergence, and the roundoff error is controlled with some reorthogonalization. We will show that results are actually better with only one seeding, even in the case of related right-hand sides. Controlling rounding error gives the potential for rapid convergence for the second and subsequent right-hand sides. Polynomial preconditioning can help reduce storage needed for reorthogonalization. The new seed methods are applied to examples including matrices from quantum chromodynamics. Copyright © 2013 John Wiley & Sons, Ltd.

[1]  Andreas Stathopoulos,et al.  Computing and Deflating Eigenvalues While Solving Multiple Right-Hand Side Linear Systems with an Application to Quantum Chromodynamics , 2007, SIAM J. Sci. Comput..

[2]  Tony F. Chan,et al.  Analysis of Projection Methods for Solving Linear Systems with Multiple Right-Hand Sides , 1997, SIAM J. Sci. Comput..

[3]  D. O’Leary.  The block conjugate gradient algorithm and related methods , 1980 .

[4]  A. Frommer,et al.  Exploiting Structure in Krylov Subspace Methods for the Wilson Fermion Matrix , 1997 .

[5]  R. Narayanan,et al.  Alternative to domain wall fermions , 2000, hep-lat/0005004.

[6]  H. V. D. Vorst,et al.  An iterative solution method for solving f ( A ) x = b , using Krylov subspace information obtained for the symmetric positive definite matrix A , 1987 .

[7]  Efstratios Gallopoulos,et al.  An Iterative Method for Nonsymmetric Systems with Multiple Right-Hand Sides , 1995, SIAM J. Sci. Comput..

[8]  R. Mittra,et al.  A conjugate gradient algorithm for the treatment of multiple incident electromagnetic fields , 1989 .

[9]  Simon Heybrock,et al.  Krylov subspace methods and the sign function: multishifts and deflation in the non-Hermitian case , 2009, 0910.2927.

[10]  L. Reichel,et al.  A stable Richardson iteration method for complex linear systems , 1988 .

[11]  G. Meurant.  The Lanczos and Conjugate Gradient Algorithms: From Theory to Finite Precision Computations , 2006 .

[12]  Ronald B. Morgan,et al.  Deflated and Restarted Symmetric Lanczos Methods for Eigenvalues and Linear Equations with Multiple Right-Hand Sides , 2008, SIAM J. Sci. Comput..

[13]  H. Simon.  Analysis of the symmetric Lanczos algorithm with reorthogonalization methods , 1984 .

[14]  R. Freund.  On conjugate gradient type methods and polynomial preconditioners for a class of complex non-hermitian matrices , 1990 .

[15]  S. Ashby.  Polynomial Preconditioning for Conjugate Gradient Methods , 1988 .

[16]  J. Grcar.  Analyses of the lanczos algorithm and of the approximation problem in richardson's method , 1981 .

[17]  Henk A. van der Vorst,et al.  Approximate solutions and eigenvalue bounds from Krylov subspaces , 1995, Numer. Linear Algebra Appl..

[18]  Charles Frederick Smith.  The Performance of Preconditioned Iterative Methods in Computational Electromagnetics , 1987 .

[19]  A. Frommer.  Linear systems solvers — recent developments and implications for lattice computations , 1996, hep-lat/9608074.

[20]  Eric L. Miller,et al.  QMR-Based Projection Techniques for the Solution of Non-Hermitian Systems with Multiple Right-Hand Sides , 2001, SIAM J. Sci. Comput..

[21]  M. Luscher.  Local coherence and deflation of the low quark modes in lattice QCD , 2007, 0706.2298.

[22]  Y. Saad,et al.  Practical Use of Some Krylov Subspace Methods for Solving Indefinite and Nonsymmetric Linear Systems , 1984 .

[23]  Frédéric Guyomarc'h,et al.  An Augmented Conjugate Gradient Method for Solving Consecutive Symmetric Positive Definite Linear Systems , 2000, SIAM J. Matrix Anal. Appl..

[24]  H. V. D. Vorst,et al.  The rate of convergence of Conjugate Gradients , 1986 .

[25]  Y. Saad,et al.  Practical Use of Polynomial Preconditionings for the Conjugate Gradient Method , 1985 .

[26]  Ronald B. Morgan,et al.  Restarting the Nonsymmetric Lanczos Algorithm for Eigenvalues and Linear Equations Including Multiple Right-Hand Sides , 2011, SIAM J. Sci. Comput..

[27]  Andreas Stathopoulos,et al.  Extending the eigCG algorithm to nonsymmetric Lanczos for linear systems with multiple right‐hand sides , 2013, Numer. Linear Algebra Appl..

[28]  G. Meurant,et al.  The Lanczos and conjugate gradient algorithms in finite precision arithmetic , 2006, Acta Numerica.

[29]  M. Saunders,et al.  Solution of Sparse Indefinite Systems of Linear Equations , 1975 .

[30]  Y. Saad,et al.  On the Lánczos method for solving symmetric linear systems with several right-hand sides , 1987 .

[31]  R. Morgan,et al.  Deflated Iterative Methods for Linear Equations with Multiple Right-Hand Sides , 2004, math-ph/0405053.

[32]  Kesheng Wu,et al.  Thick-Restart Lanczos Method for Symmetric Eigenvalue Problems , 1998, IRREGULAR.

[33]  B. Parlett.  A new look at the Lanczos algorithm for solving symmetric systems of linear equations , 1980 .

[34]  Julien Langou,et al.  Solving large linear systems with multiple right-hand sides , 2003 .

[35]  R C Brower,et al.  Adaptive multigrid algorithm for the lattice Wilson-Dirac operator. , 2010, Physical review letters.

[36]  H. Simon.  The Lanczos algorithm with partial reorthogonalization , 1984 .

[37]  R. Freund,et al.  A block QMR algorithm for non-Hermitian linear systems with multiple right-hand sides , 1997 .

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

[39]  Ronald B. Morgan,et al.  GMRES WITH DEFLATED , 2008 .

[40]  P. Forcrand.  Progress on lattice QCD algorithms , 1995, hep-lat/9509082.

[41]  H. V. D. Vorst,et al.  Numerical methods for the QCDd overlap operator. I. Sign-function and error bounds , 2002, hep-lat/0202025.

[42]  Anne Greenbaum,et al.  Predicting the Behavior of Finite Precision Lanczos and Conjugate Gradient Computations , 2015, SIAM J. Matrix Anal. Appl..

[43]  Kesheng Wu,et al.  Thick-Restart Lanczos Method for Large Symmetric Eigenvalue Problems , 2000, SIAM J. Matrix Anal. Appl..

[44]  R. Morgan,et al.  Deflation of eigenvalues for iterative methods in lattice QCD , 2003, hep-lat/0309068.

[45]  Andrea Benaglia,et al.  Transverse-Momentum and Pseudorapidity Distributions of Charged Hadrons in pp Collisions at root s=7 TeV , 2010 .

[46]  Christopher C. Paige,et al.  The computation of eigenvalues and eigenvectors of very large sparse matrices , 1971 .

[47]  B. Parlett,et al.  The Lanczos algorithm with selective orthogonalization , 1979 .

[48]  B. Parlett.  The Symmetric Eigenvalue Problem , 1981 .

[49]  Eric de Sturler,et al.  Recycling Krylov Subspaces for Sequences of Linear Systems , 2006, SIAM J. Sci. Comput..

[50]  M. Gutknecht.  BLOCK KRYLOV SPACE METHODS FOR LINEAR SYSTEMS WITH MULTIPLE RIGHT-HAND SIDES : AN , 2005 .

[51]  R. Morgan.  Restarted block-GMRES with deflation of eigenvalues , 2005 .

[52]  Ronald B. Morgan,et al.  A Restarted GMRES Method Augmented with Eigenvectors , 1995, SIAM J. Matrix Anal. Appl..

[53]  M. Hestenes,et al.  Methods of conjugate gradients for solving linear systems , 1952 .

[54]  U. Heller,et al.  Study of chiral symmetry in quenched QCD using the overlap Dirac operator , 1998, hep-lat/9811030.

[55]  C. Lanczos.  Chebyshev polynomials in the solution of large-scale linear systems , 1952, ACM '52.

[56]  Andreas Stathopoulos,et al.  Extending the eigCG algorithm to non-symmetric linear systems with multiple right-hand sides , 2009, 0911.2285.