Skip to main content

A new load-prediction scheme based on algorithmic cost functions

  • Conference paper
  • First Online:
Parallel Processing: CONPAR 94 — VAPP VI (VAPP 1994, CONPAR 1994)

Abstract

Most load-sharing and load-balancing techniques rely on load indexes which only concentrate on the external behavior of parallel programs. For some applications, including symbolic computation methods, this amounts to make unrealistic assumptions about the stability of parallel programs. We present a new technique which provides reliable predictions on tasks completion times for a class of parallel applications including symbolic computations. A straightforward and simple annotation of the initial program is required. The stability and limits of our load index are also discussed. We finally show how this technique can improve the writing of portable and scalable parallel libraries, even for an heterogeneous parallel machine.

Besides, when the processors are shared between several users, no static assumption can be made on the number of under-loaded processors.

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

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. A. V. Aho, J. E. Hopcroft, and J. D. Ullman. The design and analysis of computer algorithms. Addison-Wesley, 1974.

    Google Scholar 

  2. P. E. Bernard. Répartition dynamique de charge. Master's thesis, LMC-IMAG, 1993.

    Google Scholar 

  3. D. Ferrari and S. Zhou. A load index for dynamic load balancing. In Proc. IEEE-ACM Fall Joint Comput. Conf., pages 684–690, Nov 1986.

    Google Scholar 

  4. K. O. Geddes, S. R. Czapor, and G. Labahn. Algorithms for computer algebra. Kluwer Academic Publishers, 1992.

    Google Scholar 

  5. K. K. Goswami, M. Devarakonda, and R.K. Iyer. Prediction-based dynamic load-sharing heuristics. IEEE Trans. on Parallel & Distrib. Syst., 4(6):638–648, Jun 1993.

    Google Scholar 

  6. M. Christaller J. Briat and J.L. Roch. Une maquette pour athapascan-0. In Actes de RenPar'6, ENS Lyon, France, pages 231–234, Jun 1994.

    Google Scholar 

  7. D. E. Knuth. The art of computer programming., volume 2. Addison-Wesley, 1981.

    Google Scholar 

  8. O. Kremien, J. Kramer, and J. Magee. Scalable, adaptive load sharing for distributed systems. IEEE Parallel & Distrib. Tech., 1(3):62–70, Aug 1993.

    Google Scholar 

  9. T. Kunz. The influence of different workload descriptions on a heuristic load balancing scheme. IEEE Trans. on Software Eng., 17(7):725–730, Jul 1991.

    Google Scholar 

  10. B. Plateau. Présentation d'apache. Technical report, LGI & LMC-IMAG, 1993.

    Google Scholar 

  11. J. L. Roch. An environment for parallel algebraic computation. In Springer-Verlag, editor, Computer Algebra and Parallelism, LNCS 584, pages 33–50, May 1990.

    Google Scholar 

  12. J. L. Roch, A. Vermeerbergen, and G. Villard. Cost prediction for load balancing: Application to algebraic computations. In Springer-Verlag, editor, Proc. of CONPAR 92-VAPP V, LNCS 634, pages 467–478, Sep 1992.

    Google Scholar 

  13. M. Snir. Scalable parallel computers and scalable parallel codes: from theory to practice. In Springer-Verlag, editor, Parallel Architecures and Their Efficient Use. LNCS 678, pages 176–184, Nov 1992.

    Google Scholar 

  14. A. Vermeerbergen. Les poly-algorithmes et la prévision de coûts pour une expression portable et extensible du parallélisme. In Actes de RenPar'6, ENS Lyon, France, pages 51–54, Jun 1994.

    Google Scholar 

  15. M. H. Willebeek-LeMair and A. P. Reeves. Strategies for dynamic load balancing on highly parallel computers. IEEE Trans. on Parallel. & Distrib. Sys., 4(9):979–993, Sep 1993.

    Google Scholar 

  16. S. Zhou. A trace-driven simulation study of dynamic load balancing. IEEE Trans. on Software Eng., 14(9):1327–1341, Sep 1988.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Bruno Buchberger Jens Volkert

Rights and permissions

Reprints and permissions

Copyright information

© 1994 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Roch, J.L., Vermeerbergen, A., Villard, G. (1994). A new load-prediction scheme based on algorithmic cost functions. In: Buchberger, B., Volkert, J. (eds) Parallel Processing: CONPAR 94 — VAPP VI. VAPP CONPAR 1994 1994. Lecture Notes in Computer Science, vol 854. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-58430-7_76

Download citation

  • DOI: https://doi.org/10.1007/3-540-58430-7_76

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-58430-8

  • Online ISBN: 978-3-540-48789-0

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics