Skip to main content

Advertisement

Log in

Reliable updated residuals in hybrid Bi-CG methods

Zuverläßlich berechnete Residuen in hybriden Bi-CG Verfahren

  • Published:
Computing Aims and scope Submit manuscript

    We’re sorry, something doesn't seem to be working properly.

    Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.

Abstract

Many iterative methods for solving linear equationsAx=b aim for accurate approximations tox, and they do so by updating residuals iteratively. In finite precision arithmetic, these computed residuals may be inaccurate, that is, they may differ significantly from the (true) residuals that correspond to the computed approximations. In this paper we will propose variants on Neumaier's strategy, originally proposed for CGS, and explain its success. In particular, we will propose a more restrictive strategy for accumulating groups of updates for updating the residual and the approximation, and we will show that this may improve the accuracy significantly, while maintaining speed of convergence. This approach avoids restarts and allows for more reliable stopping criteria. We will discuss updating conditions and strategies that are efficient, lead to accurate residuals, and are easy to implement. For CGS and Bi-CG these strategies are particularly attractive, but they may also be used to improve Bi-CGSTAB, BiCGstab(l), as well as other methods.

Zusammenfassung

Viele iterative Methoden zur Lösung linearer Gleichungssysteme berechnen die Iterierten über aufdatierte Residuen. In endlicher Arithmetik können diese Residuen sehr ungenau sein, d.h., sie können sich erheblich von den tatsächlichen unterscheiden. In dieser Arbeit stellen wir Varianten der Neumaier Strategie vor, die ursprünglich für das CGS-Verfahren vorgeschlagen wurde, und erklären deren Erfolge. Insbesondere werden wir eine Variante vorschlagen, bei der mehrere Aufdatierungsschritte zusammengefaßt werden. Wir zeigen, daß sich die Genauigkeit der berechneten Residuen dadurch erheblich verbessern läßt, ohne daß die Konvergenzgeschwindigkeit beeinträchtigt wird. Dieser Ansatz vermeidet Neustarts und ermöglicht zuverlässigere Abbruchkriterien. Wir diskutieren Aufdatierungsbedingungen und Strategien, die effizient und leicht zu implementieren sind. Diese Strategien führen zu genaueren Residuen und sind insbesondere für CGS und Bi-CG-aber auch für Bi-CGSTAB, BiCGstab(l) und andere Verfahren-sehr attraktiv.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

References

  1. Bai, Z.: Error analysis of the Lanczos algorithm for the nonsymmetric eigenvalue problem. Math. Comp.62, 209–226 (1994).

    MATH  MathSciNet  Google Scholar 

  2. Fletcher, R.: Conjugate gradient methods for indefinite systems. In: Proc. of the Dundee Biennial Conference on Numerical Analysis (Watson, G., ed.), pp. 73–89. New York: Springer 1975.

    Google Scholar 

  3. Fokkema, D. R., Sleijpen, G. L. G., van der Vorst, H. A.: Generalized conjugate gradient squared. Preprint 851, Dept. Math., University Utrecht (1994).

  4. Freund, R. W.: A transpose-free quasi-minimal residual algorithm for non-Hermitian linear systems, SIAM J. Sci. Comput.14, 470–482 (1993).

    MATH  MathSciNet  Google Scholar 

  5. Freund, R. W., Gutknecht, M. H., Nachtigal, N. M.: An implementation of the look-ahead Lanczos algorithm for non-Hermitian matrices. SIAM J. Sci. Comput.14, 137–158 (1993).

    MathSciNet  Google Scholar 

  6. Freund, R. W., Nachtigal, N. M.: QMR: a quasi-minimal residual method for non-Hermitian linear systems. Numer. Math.60, 315–339 (1991).

    Article  MathSciNet  Google Scholar 

  7. Greenbaum, A.: Behavior of slightly perturbed Lanczos and conjugate gradient recurrences. Linear Algebra Appl.113, 7–63 (1989).

    Article  MATH  MathSciNet  Google Scholar 

  8. Greenbaum, A.: Estimating the attainable accuracy of recursively computed residual methods. Preprint, Courant Institute of Math. Sc., 1995.

  9. Lanczos, C.: Solution of systems of linear equations by minimized iteration. J. Res. Nat. Bur. Stand.49, 33–53 (1952).

    MathSciNet  Google Scholar 

  10. Meier Yang, U.: Preconditioned conjugate gradient-like methods of nonsymmetric linear systems. Preprint, Center for Research and Development, University of Illinois at Urbana-Champaign, 1992.

  11. Neumaier, A.: Oral presentation at the Oberwolfach meeting “Numerical Linear Algebra”. Oberwolfach, April 1994.

  12. Paige, C. C.: Accuracy and effectiveness of the Lanczos algorithm for the symmetric eigenproblem. Linear Algebra Appl.34, 235–258 (1980).

    Article  MATH  MathSciNet  Google Scholar 

  13. Paige, C. C., Parlett, B. N., Van der Vorst, H. A.: Approximate solutions and eigenvalue bounds from Krylov subspaces. Numer. Lin. Agl. Appl.2, 115–134 (1995).

    Google Scholar 

  14. Saad, Y.: A flexible inner-outer preconditioned GMRES algorithm. SIAM J. Sci. Statist. Comput.14, 461–469 (1993).

    Article  MATH  MathSciNet  Google Scholar 

  15. Saad, Y., Schultz, M. H., GMRES: A generalized minimum residual algorithm for solving nonsymmetric linear system. SIAM J. Sci. Stat. Comp.7, 856–869 (1986).

    Article  MathSciNet  Google Scholar 

  16. Sleijpen, G. L. G., Fokkema, D. R.: BiCGstab(l) for linear equations involving matrices with complex spectrum. Elect. Trans. Numer. Anal. (ETNA)1, 11–32 (1993).

    MathSciNet  Google Scholar 

  17. Sleijpen, G. L. G., Van der Vorst, H. A. Maintaining convergence properties of BiCGstab methods in finite precision arithmetic. Preprint Nr. 861, Dept. Math., University Utrecht, 1994. To appear in Numerical Algorithms.

  18. Sleijpen, G. L. G., Van der Vorst, H. A., Fokkema, D. R.: BiCGstab(l) and other hybrid Bi-CG methods. Numer. Alg.7, 75–109 (1994).

    Google Scholar 

  19. Sonneveld, P.: CGS, a fast Lanczos-type solver for nonsymmetric linear systems. SIAM J. Sci. Stat. Comp.10, 36–52 (1989).

    Article  MATH  MathSciNet  Google Scholar 

  20. Van der Vorst, H. A.: Bi-CGSTAB: A fast and smoothly converging variant of Bi-CG for the solution of nonsymmetric linear systems. SIAM J. Sci. Stat. Comput.13, 631–644 (1992).

    MATH  Google Scholar 

  21. Zhou, L., Walker, H. F.: Residual smoothing techniques for iterative methods. SIAM J. Sci. Comput.15, 297–312 (1994).

    Article  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sleijpen, G.L.G., van der Vorst, H.A. Reliable updated residuals in hybrid Bi-CG methods. Computing 56, 141–163 (1996). https://doi.org/10.1007/BF02309342

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02309342

AMS Subject Classification

Key words

Navigation