Numerische Mathematik

, Volume 57, Issue 1, pp 285–312 | Cite as

On conjugate gradient type methods and polynomial preconditioners for a class of complex non-hermitian matrices

  • Roland Freund


We consider conjugate gradient type methods for the solution of large linear systemsA x=b with complex coefficient matrices of the typeA=T+iσI whereT is Hermitian and σ a real scalar. Three different conjugate gradient type approaches with iterates defined by a minimal residual property, a Galerkin type condition, and an Euclidean error minimization, respectively, are investigated. In particular, we propose numerically stable implementations based on the ideas behind Paige and Saunder's SYMMLQ and MINRES for real symmetric matrices and derive error bounds for all three methods. It is shown how the special shift structure ofA can be preserved by using polynomial preconditioning, and results on the optimal choice of the polynomial preconditioner are given. Also, we report on some numerical experiments for matrices arising from finite difference approximations to the complex Helmholtz equation.

Subject Classifications

AMS(MOS): 65F10, 65N20, 41A50 CR: G1.3 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Ashby, S.F., Manteuffel, T.A., Saylor, P.E.: A taxonomy for conjugate gradient methods. Preprint UCRL-98508, Lawrence Livermore National Laboratory, March 1988Google Scholar
  2. 2.
    Barbour, I.M., Behilil, N.-E., Gibbs, P.E., Rafiq, M., Moriarty, K.J.M., Schierholz, G.: Updating fermions with the Lanzos method. J. Comput. Phys.68, 227–236 (1987)Google Scholar
  3. 3.
    Bayliss, A., Goldstein, C.I., Turkel, E.: The numerical solution of the Helmoholtz equation for wave propagation problems in underwater acoustics. Comp. Maths. Appl.11, 655–665 (1985)Google Scholar
  4. 4.
    Carlson, B.C., Todd, J.: Zolotarev's first problem — the best approximation by polynomials of degree ≤n−2 tox n−n σx n−1 in [−1,1]. Aequationes Math.26, 1–33 (1983)Google Scholar
  5. 5.
    Chandra, R.: Conjugate gradient methods for partial differential equations. Ph.D. Thesis, Computer Science Department, Research Report 129, Yale University, January 1978Google Scholar
  6. 6.
    Concus, P., Golub, G.H.: A generalized conjugate gradient method for nonsymmetric systems of linear equations. In: Glowingski, R., Lions, J.L. (eds.) Computing methods in applied sciences and engineering, pp. 56–65. Lecture Notes in Economics and Mathematical Systems, Vol. 134. Berlin Heidelberg New York: Springer 1976Google Scholar
  7. 7.
    Concus, P., Golub, G.H., O'Leary, D.P.: A generalized conjugate gradient method for the numerical solution of elliptic partial differential equations. In: Bunch, J.R., Rose, D.J. (eds.) Sparse matrix computations, pp. 309–332. New York: Academic Press 1976Google Scholar
  8. 8.
    Delfour, M., Fortin, M., Payre, G.: Finite-differences solutions of a non-linear Schrödinger equation. J. Comput. Phys.44, 277–288 (1981)Google Scholar
  9. 9.
    Duffin, R., Schaeffer, A.C.: Some properties of functions of exponential type. Bull. Am. Math. Soc.44, 236–240 (1938)Google Scholar
  10. 10.
    Eiermann, M., Li, X., Varga, R.S.: On hybrid semiiteratives methods. SIAM J. Numer. Anal.26, 152–168 (1989)Google Scholar
  11. 11.
    Eisenstat, S.C., Elman, H.C., Schultz, M.H.: Variationals iterative methods for non-symmetric systems of linear equations SIAM J. Numer. Anal.20, 345–357 (1983)Google Scholar
  12. 12.
    Elmore, W.C., Heald, M.A.: Physics of waves. New York: McGraw-Hill 1969Google Scholar
  13. 13.
    Faber, V., Manteuffel, T.: Necessary and sufficient conditions for the existence of a conjugate gradient method. SIAM J. Numer. Anal.21, 352–362 (1984)Google Scholar
  14. 14.
    Faber, V., Manteuffel, T.: Orthogonal error methods. SIAM J. Numer. Anal.24, 170–187 (1987)Google Scholar
  15. 15.
    Fischer, B., Freund, R.: On the constrained Chebyshev approximation problem on ellipses. J. Approx. Theory, to appear (1990)Google Scholar
  16. 16.
    Fletcher, R.: Conjugate gradient methods for indefinite systems. In: Watson, G.A. (ed.) Numerical Analysis Dundee 1975, pp. 73–89. Lecture Notes in Mathematics, Vol. 506. Berlin Heidelberg New York: Springer 1976Google Scholar
  17. 17.
    Fridman, V.M.: The method of minimum iterations with minimum errors for a system of linear algebraic equations with a symmetrical matrix. USSR Comput. Math. and Math. Phys.2, 362–363 (1963)Google Scholar
  18. 18.
    Freund, R.: Über einige cg-ähnliche Verfahren zur Lösung linearers Gleichungssysteme. Doctoral Thesis, Universität Würzburg, F.R. of Germany May, 1983Google Scholar
  19. 19.
    Freund, R., Ruscheweyh, St.: On a class of Chebyshev approximation problems which arise in connection with a conjugate gradient type method. Numer. Math.48, 525–542 (1986)Google Scholar
  20. 20.
    Goldstein, C.I.: Multigrid preconditioners applied to three-dimensional parabolic equation type models. In: Lee, D., Sternberg, R.L., Schultz, M.H. (eds.) Computational acoustics: wave propagation, pp. 57–74. Amsterdam: North-Holland 1988Google Scholar
  21. 21.
    Golub, G.H., Van Loan, C.F.: Matrix computatiom. Baltimore: The Johns Hopkins University Press 1983Google Scholar
  22. 22.
    Golub, G.H., Varga, R.S.: Chebyshev semi-iterative methods, successive overrelaxation iterative methods, and second order Richardson iterative methods. Numer. Math.3, 147–168 (1961)Google Scholar
  23. 23.
    Hestenes, M.R., Stiefel, E.: Methods of conjugate gradients for solving linear systems. J. Res. Natl. Bur. Stand.49, 409–436 (1952)Google Scholar
  24. 24.
    Householder, A.S.: The theory of matrices in numerical analysis. New York: Blaisdell 1964Google Scholar
  25. 25.
    Johnson, O.G., Micchelli, C.A., Paul, G.: Polynomial preconditioners for conjugate gradient calculations. SIAM J. Numer. Anal.20, 362–376 (1983)Google Scholar
  26. 26.
    Joubert, W.D., Young, D.M.: Necessary and sufficient conditions for the simplification of generalized conjugate-gradient algorithms. Linear Algebra Appl.88/89, 449–485 (1987)Google Scholar
  27. 27.
    Lanczos, C.: An iteration method for the solution of the eigenvalue problem of linear differential and integrals operators. J. Res. Natl. Bur. Stand.45, 255–282 (1950)Google Scholar
  28. 28.
    Marfurt, K.J.: Finite element modeling of elastodynamic and electromagnetic wave propagation for geophysical exploration. In: Glowinski, R., Lions, J.L. (eds.) Computing methods in applied sciences and engineering VII, pp. 517–547. Amsterdam: North Holland 1986Google Scholar
  29. 29.
    Meinardus, G.: Approximation of functions: Theory and numerical methods. Berlin Heidelberg New York: Springer 1967Google Scholar
  30. 30.
    Paige, C.C., Saunders, M.A.: Solution of sparse indefinite systems of linear equations. SIAM J. Numer. Anal.12, 617–629 (1975)Google Scholar
  31. 31.
    Peterson, A.F., Mittra, R.: Method of conjugate gradients for the numerical solution of large-body electromagnetic scattering problems. J. Opt. Soc. Am. A2, 971–977 (1985)Google Scholar
  32. 32.
    Pierce, A.D.: Acoustics: An introduction to its physical principles and applications. New York: McGraw-Hill 1981Google Scholar
  33. 33.
    Rapoport, D.: A nonlinear Lanczos algorithm and the stationary Navier-Stokes equation. Ph.D. Thesis, Department of Mathematics, New York University, October 1978Google Scholar
  34. 34.
    Rutishauser, H.: Theory of gradient methods. In: Refined iterative methods for computation of the solution and the eigenvalues of self-adjoint boundary value problems, pp. 24–49. Mitteilungen aus dem Institut für Angewandte Mathematik an der ETH Zürich, Vol. 8. Basel: Birkhäuser 1959Google Scholar
  35. 35.
    Saad, Y.: Krylov subspace methods on supercomputers. SIAM J. Sci. Stat. Comput.10, 1200–1232 (1989)Google Scholar
  36. 36.
    Saad, Y., Schultz, M.H.: Conjugate gradient-like algorithm for solving nonsymmetric linear systems. Math. Comput.44, 417–424 (1985)Google Scholar
  37. 37.
    Schultz, M.H., Lee, D., Jackson, K.R.: Application of the Yale sparse technique to solve the 3-dimensional parabolic wave equation. In: Scully-Power. P.D., Lee, D. (eds.) Recent progress in the development and application of the parabolic equation. Naval Underwater Systems Center, Technical Document 7145, May 1984Google Scholar
  38. 38.
    Sidi, A.: Extrapolation vs. projection methods for linear systems of equations. J. Comput. Appl. Math.22, 71–88 (1988)Google Scholar
  39. 39.
    Stoer, J.: Solution of large linear systems of equations by conjugate gradient type methods. In: Bachem, A., Grötschel, M., Korte, B. (eds.) Mathematical programming — The state of the art, pp 540–565. Berlin Heidelberg New York Tokyo: Springer 1983Google Scholar
  40. 40.
    Stoer, J., Freund, R.: On the solution of large indefinite systems of linear equations by conjugate gradient algorithms In: Glowinski, R., Lions, J.L. (eds.) Computing methods in applied sciences and engineering V, pp 35–53. Amsterdam: North Holland 1982Google Scholar
  41. 41.
    Szyld, D.B.: A two-level iterative method for large sparse generalized eigen value calculations. Ph.D. Thesis, Department of Mathematics, New York University, October 1983Google Scholar
  42. 42.
    Szyld, D.B., Widlund, O.: Applications of conjugate gradient type methods to eigenvalue calculations. In: Vichnevetsky, R., Stepleman, R.S. (eds.) Advances in computer methods for partial differential equations III, pp. 167–173. New Brunswick: IMACS 1979Google Scholar
  43. 43.
    Trummer, M.: An efficient implementation of a conformal mapping method based on the Szegö kernel. SIAM J. Numer. Anal.23, 853–872 (1986)Google Scholar
  44. 44.
    Widlund, O.: A Lanczos method for a class of nonsymmetric systems of linear equations. SIAM J. Numer. Anal.15, 801–812 (1978)Google Scholar

Copyright information

© Springer-Verlag 1990

Authors and Affiliations

  • Roland Freund
    • 1
    • 2
  1. 1.Institut für Angewandte Mathematik und StatistikUniversität WürzburgWürzburgFederal Republic of Germany
  2. 2.NASA Ames Research CenterRIACSMoffett FieldUSA

Personalised recommendations