Preconditioned Krylov Subspace Methods for the Numerical Solution of Markov Chains

  • Yousef Saad

Abstract

In a general projection technique the original matrix problem of size N is approximated by one of dimension m, typically much smaller than N. A particularly successful class of techniques based on this principle is that of Krylov subspace methods which utilise subspaces of the form span{v, Av,…., Am-1 v}. This general principle can be used to solve linear systems and eigenvalue problems which arise when computing stationary probability distributions of Markov chains. It can also be used to approximate the product of the exponential of a matrix by a vector as occurs when following the solutions of transient models. In this paper we give an overview of these ideas and discuss preconditioning techniques which constitute an essential ingredient in the success of Krylov subspace methods.

Keywords

Petroleum Resid 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    W. E. Arnoldi. The principle of minimized iteration in the solution of the matrix eigenvalue problem. Quart. Appl. Math., 9:17–29, 1951.MathSciNetMATHGoogle Scholar
  2. [2]
    O. Axelsson. Conjugate gradient type-methods for unsymmetric and inconsistent systems of linear equations. Linear Algebra Appl., 29:1–16, 1980.MathSciNetMATHCrossRefGoogle Scholar
  3. [3]
    O. Axelsson. A generalized conjugate gradient, least squares method. Num. Math., 51:209–227, 1987.MathSciNetMATHCrossRefGoogle Scholar
  4. [4]
    R. Chandra. Conjugate Gradient Methods for Partial Differential Equations. PhD thesis, Yale University, Computer Science Dept., New Haven, CT. 06520, 1978.Google Scholar
  5. [5]
    I. S. Duff and G. A. Meurant. The effect of reordering on preconditioned conjugate gradients. BIT, 29:635–657, 1989.MathSciNetMATHCrossRefGoogle Scholar
  6. [6]
    L. C. Dutto. The effect of reordering on the preconditioned GMRES algorithm for solving the compressible Navier-Stokes equations. International Journal for Numerical Methods in Engineering, 36:457–497, 1993.MathSciNetMATHCrossRefGoogle Scholar
  7. [7]
    S. C. Eisenstat, H. C. Elman, and M. H. Schultz. Variational iterative methods for nonsymmetric systems of linear equations. SIAM J Num. Anal., 20:345–357, 1983.MathSciNetMATHCrossRefGoogle Scholar
  8. [8]
    R. Fletcher. Conjugate gradient methods for indefinite systems. In G. A. Watson, editor, Proceedings of the Dundee Biennal Conference on Numerical Analysis 1974, pages 73–89, New York, 1975. University of Dundee, Scotland, Springer Verlag.Google Scholar
  9. [9]
    Roland W. Freund. A Transpose-Free Quasi-Minimal Residual algorithm for non-Hermitian linear systems. SIAM J. Sci. Comp., 14(2):470–482, 1993.MathSciNetMATHCrossRefGoogle Scholar
  10. [10]
    E. Gallopoulos and Y. Saad. Efficient solution of parabolic equations by polynomial approximation methods. SIAM J. Sci. Stat. Comput., 13:1236–1264, 1992.MathSciNetMATHCrossRefGoogle Scholar
  11. [11]
    G. H. Golub and C. Van Loan. Matriz Computations. Academic Press, New York, 1981.Google Scholar
  12. [12]
    K. C. Jea and D. M. Young. Generalized conjugate gradient acceleration of nonsymmetrizable iterative methods. Linear Algebra Appl., 34:159–194, 1980.MathSciNetMATHCrossRefGoogle Scholar
  13. [13]
    C. Lanczos. Solution of systems of linear equations by minimized iterations. J. Res. Nat. Bur. Standards, 49:33–53, 1952.MathSciNetCrossRefGoogle Scholar
  14. [14]
    C. C. Paige and M. A. Saunders. An algorithm for sparse linear equations and sparse least squares. ACM Trans. Math. Software, 8:43–71, 1982.MathSciNetMATHCrossRefGoogle Scholar
  15. [15]
    B. Philippe, Y. Saad, and W. J. Stewart. Numerical methods in Markov chain modeling. Journal of Operations Research, 40(6):1156–1179, 1992.MATHCrossRefGoogle Scholar
  16. [16]
    B. Philippe and R. B. Sidje. Transient solutions of Markov processes by Krylov subspaces. Technical Report 736, IRISA, University of Rennes, Campus de Beaulieu, Rennes, France, 1993.Google Scholar
  17. [17]
    S. J. Polak, C. Den Heijer, W. H. A. Schilders, and P. Markowich. Semiconductor device modelling from the numerical point of view. Internat. J. Numer. Meth. Eng., 24:763–838, 1987.MATHCrossRefGoogle Scholar
  18. [18]
    Y. Saad. Krylov subspace methods for solving large unsymmetric linear systems. Mathematics of Computation, 37:105–126, 1981.MathSciNetMATHCrossRefGoogle Scholar
  19. [19]
    Y. Saad. The Lanczos biorthogonalization algorithm and other oblique projection methods for solving large unsymmetric systems. SIAM J. Nu-mer. Anal.,19:470–484, 1982.MathSciNetCrossRefGoogle Scholar
  20. [20]
    Y. Saad. Practical use of some Krylov subspace methods for solving indefinite and unsymmetric linear systems. SIAM J. Sci. Statist. Comput., 5:203–228, 1984.MathSciNetMATHCrossRefGoogle Scholar
  21. [21]
    Y. Saad. Analysis of some Krylov subspace approximations to the matrix exponential operator. SIAM J. Numer. Anal., 29:209–228, 1992.MathSciNetMATHCrossRefGoogle Scholar
  22. [22]
    Y. Saad. ILUT: a dual threshold incomplete ILU factorization. Technical Report 92–38, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, 1992. to appear.Google Scholar
  23. [23]
    Y. Saad. Numerical Methods for Large Eigenvalue Problems. Halstead Press, New York, 1992.MATHGoogle Scholar
  24. [24]
    Y. Saad. A flexible inner-outer preconditioned GMRES algorithm. SIAM J. Sci. Stat. Comput., 14:461–469, 1993.MathSciNetMATHCrossRefGoogle Scholar
  25. [25]
    Y. Saad. Iterative methods for sparse linear systems. PWS publishing, New York, 1995. to appear.Google Scholar
  26. [26]
    Y. Saad and M. H. Schultz. Conjugate gradient-like algorithms for solving nonsymmetric linear systems. Mathematics of Computation, 44(170):417–424, 1985.MathSciNetMATHCrossRefGoogle Scholar
  27. [27]
    Y. Saad and M. H. Schultz. GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems. SIAM J. Sci. Statist. Comput., 7:856–869, 1986.MathSciNetMATHCrossRefGoogle Scholar
  28. [28]
    Y. Saad and K. Wu. DQGMRES: a quasi-minimal residual algorithm based on incomplete orthogonalization. Technical Report UMSI-93131, Minnesota Supercomputing Institute, Minneapolis,MN, 1993. submitted.Google Scholar
  29. [29]
    P. Sonneveld. CGS, a fast Lanczos-type solver for nonsymmetric linear systems. SIAM J. Scient. Statist. Comput.,10(1):36–52, 1989.MathSciNetMATHCrossRefGoogle Scholar
  30. [30]
    W. J. Stewart. MARCA: Markov Chain Analyzer, A software package for Markov modeling. In W. J. Stewart, editor, Numerical solution of Markov chains, pages 37–62, New-York, 1991. Marcel Dekker Inc.Google Scholar
  31. [31]
    W. J. Stewart. Introduction to the Numerical Solution of Markov Chains. Princeton University Press, New Jersey, 1994.MATHGoogle Scholar
  32. [32]
    H. A. van der Vorst. Bi-CGSTAB: A fast and smoothly converging variant of Bi-CG for the solution of non-symmetric linear systems. SIAM J. Sci. Stat. Comput.,12:631–644, 1992.Google Scholar
  33. [33]
    J. W. Watts-III. A conjugate gradient truncated direct method for the iterative solution of the reservoir simulation pressure equation. Society of Petroleum Engineer Journal, 21:345–353, 1981.Google Scholar

Copyright information

© Springer Science+Business Media New York 1995

Authors and Affiliations

  • Yousef Saad
    • 1
  1. 1.Department of Computer ScienceUniversity of MinnesotaMinneapolisUSA

Personalised recommendations