Parallel Hybrid Sparse Solvers Through Flexible Incomplete Cholesky Preconditioning (original) (raw)

Abstract

We consider parallel preconditioning schemes to accelerate the convergence of Conjugate Gradients (CG) for sparse linear system solution. We develop methods for constructing and applying preconditioners on multiprocessors using incomplete factorizations with selective inversion for improved latency-tolerance. We provide empirical results on the efficiency, scalability and quality of our preconditioners for sparse matrices from model grids and some problems from practical applications. Our results indicate that our preconditioners enable more robust sparse linear system solution.

This work was supported in part by the National Science Foundation through grants ACI-0102537, EIA-0221916, and DMR-0205232.

Preview

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Balay, S., Gropp, W.D., Mcinnes, L.C., Smith, B.F.: PETSc users manual. Tech. Rep. ANL-95/11 - Revision 2.1.1, Argonne National Laboratories (2002)
    Google Scholar
  2. Chow, E.: Parallel implementation and practical use of sparse approximate inverse preconditioners with a priori sparsity patterns. Int. J. High Perf. Comput. Apps. 15, 56–74 (2001)
    Article MATH Google Scholar
  3. George, A., Liu, J.W.-H.: Computer Solution of Large Sparse Positive Definite Systems. Prentice-Hall, Englewood Cliffs (1981)
    MATH Google Scholar
  4. Grote, M.J., Huckle, T.: Parallel preconditioning with sparse approximate inverses. SIAM J. Sci. Comput. 18(3), 838–853 (1997)
    Article MATH MathSciNet Google Scholar
  5. Hestenes, M.R., Stiefel, E.: Methods of conjugate gradients for solving linear systems. National Bureau Standard J. Res. 49, 409–436 (1952)
    MATH MathSciNet Google Scholar
  6. Jones, M., Plassmann, P.: The efficient parallel iterative solution of large sparse linear systems. In: George, A., Gilbert, J.R., Liu, J.W.H. (eds.) Graph Theory and Sparse Matrix Computations. IMA, vol. 56, pp. 229–245. Springer, Heidelberg (1994)
    Google Scholar
  7. Karypis, G., Kumar, V.: A parallel algorithm for multilevel graph partitioning and sparse matrix ordering. Journal of Parallel and Distributed Computing 48(1), 71–95 (1998)
    Article MathSciNet Google Scholar
  8. Kolotilina, L.Y., Yeremin, A.Y.: Factorized sparse approximate inverse preconditionings. I. Theory. SIAM J. Matrix Anal. Appl. 14(1), 45–58 (1993)
    Article MATH MathSciNet Google Scholar
  9. Liu, J.W.-H.: The role of elimination trees in sparse factorization. SIAM J. Matrix Anal. Appl. 11(1), 134–172 (1990)
    Article MATH MathSciNet Google Scholar
  10. Liu, J.W.-H., Ng, E., Peyton, B.W.: On finding supernods for sparse matrix ccomputation. SIAM J. Matrix Anal. Appl. 14(1), 242–252 (1993)
    Article MATH MathSciNet Google Scholar
  11. Ng, E.G., Peyton, B.W.: Block sparse Cholesky algorithmon advanced uniprocessor computers. SIAM J. Sci. Comput. 14, 1034–1056 (1993)
    Article MATH MathSciNet Google Scholar
  12. Raghavan, P.: Efficient parallel triangular solution using selective inversion. Parallel Processing Letters 9(1), 29–40 (1998)
    Article MathSciNet Google Scholar
  13. Raghavan, P.: DSCPACK:Domain-separator codes for solving sparse linear systems. Tech. Rep.CSE-02-004,Department ofComputer Science and Engineering, The Pennsylvania State University (2002)
    Google Scholar
  14. Raghavan, P., Teranishi, K., Ng, E.G.: A latency tolerant hybrid sparse solver using incomplete Cholesky factorization. Numer. Linear Algebra Appl. 10, 541–560 (2003)
    Article MATH MathSciNet Google Scholar
  15. Saad, Y.: Iterative method for sparse linear systems, 2nd edn. SIAM, Philadelphia (2003)
    Book Google Scholar
  16. Teranishi, K.: Scalable Hybrid Sprase Lienar Solvers. PhD thesis, Department of Computer Science and Engineering, The Pennsylvania State University (2004)
    Google Scholar
  17. Teranishi, K., Raghavan, P., Ng, E.: A new data-mapping scheme for latencytolerant distributed sparse triangular solution. In: SuperComputing 2002, pp. 238–247 (2002)
    Google Scholar
  18. Teranishi, K., Raghavan, P., Sun, J., Micharelis, P.: A hybrid preconditioner for sparse systems from thermo-mechanical applications. Intl. J. Numerical Methods in Engineering (submitted)
    Google Scholar
  19. Zlatev, Z.: Use of iterative refinement in the solution of sparse linear systems. SIAM J. Numer. Anal. 19, 381–399 (1982)
    Article MATH MathSciNet Google Scholar

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, The Pennsylvania State University, 111 IST Bldg, University Park, PA, 16802, USA
    Keita Teranishi & Padma Raghavan

Authors

  1. Keita Teranishi
  2. Padma Raghavan

Editor information

Editors and Affiliations

  1. Computer Science Department, University of Tennessee, 37996-3450, Knoxville, TN, USA
    Jack Dongarra
  2. Department of Informatics and Mathematical Modelling, Technical University of Denmark, DK-2800, Lyngby, Denmark
    Kaj Madsen
  3. Informatics & Mathematical Modeling, Technical University of Denmark, DK-2800, Lyngby, Denmark
    Jerzy Waśniewski

Rights and permissions

© 2006 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Teranishi, K., Raghavan, P. (2006). Parallel Hybrid Sparse Solvers Through Flexible Incomplete Cholesky Preconditioning. In: Dongarra, J., Madsen, K., Waśniewski, J. (eds) Applied Parallel Computing. State of the Art in Scientific Computing. PARA 2004. Lecture Notes in Computer Science, vol 3732. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11558958\_76

Download citation

Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Publish with us