On sparse matrix orderings in interior point methods

Abstract

The major computational task of most interior point implementations is solving systems of equations with symmetric coefficient matrix by direct factorization methods, therefore, the performance of Cholesky-like factorizations is a critical issue. In the case of sparse and large problems the efficiency of the factorizations is closely related to the exploitation of the nonzero structure of the problem. A number of techniques were developed for fill-reducing sparse matrix orderings which make Cholesky factorizations more efficient by reducing the necessary floating point computations. We present a variant of the nested dissection algorithm incorporating special techniques that are beneficial for graph partitioning problems arising in the ordering step of interior point implementations. We illustrate the behavior of our algorithm and provide numerical results and comparisons with other sparse matrix ordering algorithms.

This is a preview of subscription content, access via your institution.

References

  1. Andersen ED, Gondzio J, Mészáros C, Xu X (1996) Implementation of interior point methods for large scale linear programs. In: Terlaky T (ed) Interior point methods of mathematical programming. Kluwer Academic, Norwel, pp 189–252

    Google Scholar 

  2. Ashcraft C, Liu JWH (1997) Using domain decomposition to find graph bisectors. BIT Numer Math 37(3):506–534

    MathSciNet  Article  MATH  Google Scholar 

  3. Ashcraft C, Liu JWH (1998a) Applications of the Dulmage–Mendelsohn decomposition and network ow to graph bisection improvement. SIAM J Matrix Anal Appl 19:325–354

    MathSciNet  Article  MATH  Google Scholar 

  4. Ashcraft C, Liu JWH (1998b) Robust ordering of sparse matrices using multisection. SIAM J Matrix Anal Appl 19(3):816–832

    MathSciNet  Article  MATH  Google Scholar 

  5. Booth KS, Lipton RJ (1981) Computing extremal and approximate distances in graphs having unit cost edges. Acta Inform 15(4):319–328

    MathSciNet  MATH  Google Scholar 

  6. Duff IS, Erisman AM, Reid JK (1986) Direct methods for sparse matrices. Oxford University Press, New York

    Google Scholar 

  7. Fiduccia C, Mattheyses R (1982) A linear-time heuristic for improving network partition. Technical report, ACM IEEE 19th design and automation conference proceedings

  8. George JA (1973) Nested dissection of a regular finite element mesh. SIAM J Numer Anal 10:345–363

    MathSciNet  Article  MATH  Google Scholar 

  9. George A, Liu JWH (1981) Computer solution of large sparse positive definite systems. Prentice-Hall, Englewood Cliffs

    Google Scholar 

  10. George A, Liu JWH (1989) The evolution of the minimum degree ordering algorithm. SIAM Rev 31:1–19

    MathSciNet  Article  MATH  Google Scholar 

  11. Gupta A (1996) Watson graph partitioning package. Technical report RC 20453, IBM T. J. Watson Research Center

  12. Hendrickson B, Leland R (1995) The Chaco user’s guide, version 2.0. Technical report, Sandia National Laboratories

  13. Hendrickson B, Rothberg E (1998) Improving the run time and quality of nested dissection ordering. SIAM J Sci Comput 20(2):468–489

    MathSciNet  Article  Google Scholar 

  14. Karypis G, Kumar V (1995) Analysis of multilevel graph partitioning. Technical report, Supercomputing ’95, proceedings of the 1995 ACM/IEEE conference on supercomputing

  15. Karypis G, Kumar V (1998) Metis a software package for partitioning unstructured graphs, partitioning meshes, and computing fill-reducing orderings of sparse matrices version 3.0. Technical report, University of Minnesota

  16. Karypis G, Kumar V (1999) A fast and high quality multilevel scheme for partitioning irregular graphs. SIAM J Sci Comput 20(1):359–392

    MathSciNet  Article  MATH  Google Scholar 

  17. Kernighan BW, Lin S (1970) An efficient heuristic procedure for partitioning graphs. Technical report 49, Bell Syst Tech J

  18. Lustig IJ, Marsten RE, Shanno DF (1994) Interior point methods for linear programming: computational state of the art. ORSA J Comput 6(1):1–15

    MathSciNet  Article  MATH  Google Scholar 

  19. Mészáros C (1997) The augmented system variant of IPMs in two-stage stochastic linear programming. Eur J Oper Res 101(2):317–327

    Article  MATH  Google Scholar 

  20. Mészáros C (1998) Ordering heuristics in interior point LP methods. In: Gianessi F, Komlósi S, Rapcsák T (eds) New trends in mathematical programming. Kluwer Academic, Norwel, pp 203–221

    Google Scholar 

  21. Mészáros C (1999) The BPMPD interior-point solver for convex quadratic problems. Optim Methods Softw 11/12:431–449

    Article  Google Scholar 

  22. Mészáros C (2007) Detecting dense columns in linear programs for interior point methods. Comput Optim Appl 36(2–3):309–320

    MathSciNet  Article  MATH  Google Scholar 

  23. Mészáros C (2010) On the implementation of interior point methods for dual-core platforms. Optim Methods Softw 25(3):449–456

    MathSciNet  Article  MATH  Google Scholar 

  24. Mészáros C, Suhl UH (2004) Advanced preprocessing techniques for linear and quadratic programming. OR Spektrum 25:575–595

    Article  Google Scholar 

  25. Mittelmann HD, Spellucci P (1998) Decision tree for optimization software. World Wide Web. http://plato.la.asu.edu/guide.html

  26. Parter SV (1961) The use of linear graphs in Gaussian elimination. SIAM Rev 3:130–191

    MathSciNet  Google Scholar 

  27. Rothberg E (1996) Ordering sparse matrices using approximate minimum local fill. Technical report, Silicon Graphics Inc., Mountain View, CA 94043

  28. Rothberg E, Hendrickson B (1998) Sparse matrix ordering methods for interior point linear programming. INFORMS J Comput 10(1):107–113

    MathSciNet  Article  Google Scholar 

  29. Yannakakis M (1981) Computing the minimum fill-in is NP-complete. SIAM J Algebr Discrete Methods 2:77–79

    MathSciNet  Article  MATH  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Csaba Mészáros.

Additional information

This work was supported in part by Hungarian Research Fund OTKA K-77420 and K-60480.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Mészáros, C. On sparse matrix orderings in interior point methods. Optim Eng 14, 519–527 (2013). https://doi.org/10.1007/s11081-013-9233-7

Download citation

Keywords

  • Sparse matrix ordering
  • Nested dissection
  • Interior point methods