ANZIAM J. 46(E) ppC394--C408, 2005.

Computing the preconditioner for the Schur complement

K. Moriya

T. Nodera
(Received 27 October 2004; revised 11 March 2005)

Abstract

The Newton scheme is used to construct an approximate inverse preconditioner for the Schur complement. However, this scheme is very expensive because of the computation cost of the matrix-matrix product. In this paper, the computation cost of the Newton scheme is reduced by implementing the preconditioner implicitly using the matrix-vector product. We also show that such an implementation is less expensive than computing the preconditioner explicitly.

Download to your computer

Authors

K. Moriya
Dept. of Information and Integrated Tech., Aoyama Gakuin University, Japan. mailto:moriya@it.aoyama.ac.jp
T. Nodera
Dept. of Math., Keio University, Japan. mailto: nodera@math.keio.ac.jp

Published May 23, 2005. ISSN 1446-8735

References

  1. G. Shulz. Iterative Berechnung der Reziproken Matrix, Z. Angew. Math. Mech., 13: 57--59, 1933.
  2. A. S. Householder. The Theory of Matrices in Numerical Analysis, Dover, New York, 1964.
  3. A. Ben-Israel and D. Cohen. On Iterative Computation of Generalized Inverses and Associated Projections, SIAM J. Numer. Anal., 3: 410--419, 1966.
  4. Y. Saad and M. H. Schultz. GMRES: A Generalized Minimal Residual Algorithm for Solving Nonsymmetric Linear Systems, SIAM J. Sci. Stat. Comput., 7: 856--869, 1986.
  5. W. Schonauer. Scientific Computing on Vector Computers, North Holland, 1987.
  6. V. Pan and R. Schreiber. An Improved Newton Iteration for the Generalized Inverse of a Matrix with Applications, SIAM J. Sci. Stat. Comp., 12(5): 1109--1130, 1991.
  7. Y. Saad. Iterative Methods for Sparse Linear Systems, PWS Publishing Company, Boston, 1996.
  8. T. Huckel. Approximate Sparsity Patterns for the Inverse of a Matrix and Preconditioning, Appl. Numer. Math., 30: 291--303, 1999.
  9. X. C. Cai and D. E. Keyes. Nonlinearly Preconditioned Inexact Newton Algorithms, SIAM J. Sci. Comput., 41(1): 183--200, 2002.
  10. University of Florida Sparse Matrix Collection. [Online] http://www.cise.ufl.edu/research/sparse/matrices.
Copyright Austral. Mathematical Soc.; prepared May 23, 2005, from Mori.tex with anziamje.sty (v2.3c)