Abstract
Preconditioned Krylov subspace methods are among the preferred iterative techniques for solving large sparse linear systems of equations. As computer architectures are evolving and problems are becoming more complex, iterative techniques are undergoing several mutations. In particular, there has been much recent work on new preconditioned that yield higher parallelism or on new implementations of the standard ones. In the past, a conservative and well understood approach has consisted of porting standard preconditioners to the new computers. However, in addition to their limited parallelism, such preconditioners have other drawbacks. The simple ILU(O) pre-conditioner may fail to converge for realistic problems arising from applications such as Computational Fluid Dynamics. The more robust analogues ILU(k) [30] or ILUT(k) [40] that allow more fill-in are not always a good alternative since they tend to be sequential in nature. In this paper we discuss these issues and give an overview of the standard approaches. Then we will propose a number of alternatives. It will be argued that some approaches based on multi-coloring can offer good compromises between generality and efficiency.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
L. Adams and H. Jordan. Is SOR color-blind? SIAM J. Sci. Statist. Comp, 6: 490–506, 1985.
L. Adams and J. Ortega. A multi-color SOR Method for Parallel Computers. In Proceedings 1982 Int. Conf. Par. Proc., pages 53–56, 1982.
L. M. Adams. Iterative algorithms for large sparse linear systems on parallel computers. PhD thesis, Applied Mathematics, University of Virginia, Charlottsville, VA, 22904, 1982. Also NASA Contractor Report 166027.
E. C. Anderson. Parallel implementation of preconditioned conjugate gradient methods for solving sparse systems of linear equations. Technical Report 805, CSRD, University of Illinois, Urbana, IL, 1988. MS Thesis.
E. C. Anderson and Y. Saad. Solving sparse triangular systems on parallel computers. Technical Report 794, University of Illinois, CSRD, Urbana, IL, 1988.
S. F. Ashby, T. A. Manteuffel, and P. E. Saylor. Adaptive polynomial preconditioning for Hermitian indefinite linear systems. BIT, 29: 583–609, 1989.
C. C. Ashcraft and R. G. Grimes. On vectorizing incomplete factorization and SSOR preconditioners. SIAM J. Sci. Statist. Comput., 9: 122–151, 1988.
O. Axelsson. A generalized conjugate gradient, least squares method. Num. Math., 51: 209–227, 1987.
D. Baxter, J. Saltz, M. H. Schultz, and S. C. Eisenstat. Preconditioned Krylov solvers and methods for runtime loop paral-lelization. Technical Report 655, Computer Science, Yale University, New Haven, CT, 1988.
D. Baxter, J. Saltz, M. H. Schultz, S. C. Eisenstat, and K. Crowley. An experimental study of methods for parallel preconditioned Krylov methods. Technical Report 629, Computer Science, Yale University, New Haven, CT, 1988.
M. Benantar and J. E. Flaherty. A Six color procedure for the parallel solution of Elliptic systems using the finite Quadtree structure. In J. Dongarra, P. Messina, D. C. Sorenson, and R. G. Voigt, editors, Proceedings of the fourth SIAM conference on parallel processing for scientific computing, pages 230–236, 1990.
I. S. Duff, A. M. Erisman, and J. K. Reid. Direct Methods for Sparse Matrices. Clarendon Press, Oxford, 1986.
I. S. Duff and G. A. Meurant. The effect of reordering on preconditioned conjugate gradients. BIT, 29: 635–657, 1989.
H. C. Elman. Iterative Methods for Large Sparse Nonsymmetric Systems of Linear Equations. PhD thesis, Yale University, Computer Science Dept., New Haven, CT., 1982.
H. C. Elman and E. Agron. Ordering techniques for the preconditioning conjugate gradient method on parallel computers. Technical Report UMIACS-TR-88-53, UMIACS, University of Maryland, College Park, MD, 1988.
H. C. Elman and G. H. Golub. Iterative methods for cyclically reduced non-self-adjoint linear systems. Technical Report CS-TR-2145, Dept. of Computer Science, University of Maryland, College Park, MD, 1988.
R. M. Ferencz. Element-by-element preconditioning techniques for large scale vectorized finite element analysis in nonlinear solid and structural mechanics. PhD thesis, Applied Mathematics, Stanford, CA, 1989.
R. Freund, M. H. Gutknecht, and N. M. Nachtigal. An implementation of the Look-Ahead Lanczos algorithm for non-Hermitian matrices, Part I. Technical Report 90-11, Massachusetts Institute of Technology, Cambridge, Massachusetts, 1990.
R. Freund and N. M. Nachtigal. An implementation of the look-ahead lanczos algorithm for non-Hermitian matrices, Part II. Technical Report 90-11, Massachusetts Institute of Technology, Cambridge, Massachusetts, 1990.
R. S. Varga G. H. Golub. Chebyshev semi iterative methods successive overrelaxation iterative methods and second order Richardson iterative methods. Numer. Math., 3: 147–168, 1961.
A. Greenbaum. Solving triangular linear systems using fortran with parallel extensions on the nyu ultracomputer prototype. Technical Report 99, Courant Institute, New York University, New York, NY, 1986.
A. L. Hageman and D. M. Young. Applied Iterative Methods. Academic Press, New York, 1981.
M. A. Heroux, P. Vu, and C. Yang. A parallel preconditioned conjugate gradient package for solving sparse linear systems on a cray Y-MP. Technical report, Cray Research, Eagan, MN, 1990. Proc. of the 1990 Cray Tech. Symp.
T. J. R. Hughes, R. M. Ferencz, and J. O. Hallquist. Large-scale vectorized implicit calculations in solid mechanics on a cray x-mp/48 utilizing ebe preconditioning conjugate gradients. Computer Methods in Applied Mechanics and Engineering, 61: 215–248, 1987.
K. C. Jea and D. M. Young. Generalized conjugate gradient acceleration of nonsymmetrizable iterative methods. Linear Algebra Appl., 34: 159–194, 1980.
M. T. Jones and P. E. Plassmann. Parallel iterative solution of sparse linear systems using ordering from graph coloring heuristics. Technical Report MCS-P198-1290, Argonne National Lab., Argonne, IL, 1990.
T. I. Karush, N. K. Madsen, and G. H. Rodrigue. Matrix multiplication by diagonals on vector/parallel processors. Technical Report UCUD, Lawrence Livermore National Lab., Livermore, CA, 1975.
T. A. Manteuffel. The Tchebychev iteration for nonsymmetric linear systems. Numer. Math., 28: 307–327, 1977.
T. A. Manteuffel. Adaptive procedure for estimation of parameter for the nonsymmetric Tchebychev iteration. Numer. Math., 28: 187–208, 1978.
J. A. Meijerink and H. A. van der Vorst. An iterative solution method for linear systems of which the coefficient matrix is a symmetric M-matrix. Math. Comp., 31 (137): 148–162, 1977.
R. Melhem. Solution of linear systems with striped sparse matrices. Parallel Comput., 6: 165–184, 1988.
T. C. Oppe, W. Joubert, and D. R. Kincaid. Nspcg user’s guide, a package for solving large linear systems by various iterative methods. Technical report, The University of Texas at Austin, 1988.
T. C. Oppe and D. R. Kincaid. The performance of ITPACK on vector computers for solving large sparse linear systems arising in sample oil reservoir simulation problems. Communications in applied numerical methods, 2: 1–7, 1986.
J. M. Ortega. Introduction to Parallel and Vector Solution of Linear Systems. Plenum Press, New York, 1988.
G. A. Paolini and G. Radicati di Brozólo. Data structures to vectorize CG algorithms for general sparsity patterns. BIT, 29: 4, 1989.
E. L Poole and J. M. Ortega. Mullticolor ICCG methods for vector computers. SIAM J. Numer. Anal., 24: 1394–1418, 1987.
Y. Saad. Least squares polynomials in the complex plane and their use for solving sparse nonsymmetric linear systems. SIAM J. Num. Anal., 24: 155–169, 1987.
Y. Saad. SPARSKIT: A basic tool kit for sparse matrix computations. Technical Report 90-20, Research Institute for Advanced Computer Science, NASA Ames Research Center, Moffet Field, CA, 1990.
Y. Saad. A flexible inner-outer preconditioned GMRES algorithm. Technical Report 91-279, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, Minnesota, 1991.
Y. Saad. ILUT: a dual strategy accurate incomplete ilu factorization. Technical Report -, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, 1991. in preparation.
Y. Saad. Massively parallel preconditioned Krylov subspace methods. Technical Report -, Minnesota Supercomputer Institute, Minneapolis, Minnesota, 1991. In preparation.
Y. Saad and M. H. Schultz. Conjugate gradient-like algorithms for solving nonsymmetric linear systems. Mathematics of Computation, 44 (170): 417–424, 1985.
Y. Saad and M. H. Schultz. Parallel implementations of preconditioned conjugate gradient methods. Research report 425, Dept Computer Science, Yale University, 1985.
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.
J. H. Saltz. Automated problem scheduling and reduction of synchronization delay effects. Technical Report 87-22, ICASE, Hampton, VA, 1987.
F. Shakib. Finite element analysis of the compressible Euler and Navier Stokes Equations. PhD thesis, Aeronautics Dept., Stanford, CA, 1989.
T. E. Tezduyar, M. Behr, S. K. A. Abadi, and S. E. Ray. A mixed CEBE/CC preconditionning for finite element computations. Technical Report UMSI 91/160, University of Minnesota, Minnesota Supercomputing Institute, Mineapolis, Minnesota, June 1991.
H. A. van der Vorst. The performance of FORTRAN implementations for preconditioned conjugate gradient methods on vector computers. Parallel Comput., 3: 49–58, 1986.
H. A. van der Vorst. Large tridiagonal and block tridiagonal linear systems on vector and parallel computers. Par. Comp., 5: 303–311, 1987.
H. A. van der Vorst. High performance preconditioning. SIAM j. Scient. Stat. Comput., 10: 1174–1185, 1989.
R. S. Varga. Matrix Iterative Analysis. Prentice Hall, Englewood Cliffs, NJ, 1962.
V. Venkatakrishnan. Preconditioned Conjugate Gradient methods for the compressible Navier Stokes equations. AIAA Journal, 29: 1092–1100, 1991.
V. Venkatakrishnan and D. J. Mavriplis. Implicit solvers for unstructured grids. In Proceedings of the AIAA 10th CFD Conference, June, 1991, HI., 1991.
V. Venkatakrishnan, H. D. Simon, and T. J. Barth. A MIMD Implementation of a Parallel Euler Solver for Unstructured Grids. Technical Report RNR-91-024, NASA Ames research center, Moffett Field, CA, 1991.
P. K. W. Vinsome. Orthomin, an iterative method for solving sparse sets of simultaneous linear equations. In Proceedings of the Fourth Symposium on Resevoir Simulation, pages 149–159. Society of Petroleum Engineers of AIME, 1976.
O. Wing and J. W. Huang. A computation model of parallel solution of linear equations. IEEE Transactions on Computers, C-29: 632–638, 1980.
C. H. Wu. A multicolour SOR method for the finite-element method. J. of Comput and App. Math., 30: 283–294, 1990.
D.M. Young. Iterative solution of large linear systems. Academic Press, New-York, 1971.
D. M. Young, T. C. Oppe, D. R. Kincaid, and L. J. Hayes. On the use of vector computers for solving large sparse linear systems. Technical Report CNA-199, Center for Numerical Analysis, University of Texas at Austin, Austin, Texas, 1985.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1993 Springer-Verlag New York, Inc.
About this paper
Cite this paper
Saad, Y. (1993). Supercomputer Implementations of Preconditioned Krylov Subspace Methods. In: Hussaini, M.Y., Kumar, A., Salas, M.D. (eds) Algorithmic Trends in Computational Fluid Dynamics. ICASE/NASA LaRC Series. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-2708-3_8
Download citation
DOI: https://doi.org/10.1007/978-1-4612-2708-3_8
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4612-7638-8
Online ISBN: 978-1-4612-2708-3
eBook Packages: Springer Book Archive