Modeling of workload in MPPs

  • Joefon Jann
  • Pratap Pattnaik
  • Hubertus Franke
  • Fang Wang
  • Joseph Skovira
  • Joseph Riordan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1291)

Abstract

In this paper we have characterized the inter-arrival time and service time distributions for jobs at a large MPP supercomputing center. Our findings show that the distributions are dispersive and complex enough that they require Hyper Erlang distributions to capture the first three moments of the observed workload. We also present the parameters from the characterization so that they can be easily used for both theoretical studies and the simulations of various scheduling algorithms.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    A. Tucker and A. Gupta, “Process control and scheduling issues for multiprogrammed shared-memory multiprocessors,” in Proceedings of the 12th. ACM Symposium on Operating Systems Principle, pp. 159–166,1989.Google Scholar
  2. 2.
    D. Feitelson and L. Rudolph, “Mapping and scheduling in a shared parallel environment using distributed hierarchical control,” in Intl. Conf Parallel Processing, pp. 1–8, 1990.Google Scholar
  3. 3.
    D. G. Feitelson, “Packing schemes for gang scheduling,” in Job Scheduling Strategies for Parallel Processing Springer-Verlag, LNCS Vol. 1162, Apr 1996, pp. 89–110,1996.Google Scholar
  4. 4.
    F. Wang, H. Franke, M. Papaefthymiou, P Pattnaik, L. Rudolph, and M. S. Squillante, “A gang scheduling design for multiprogrammed parallel computing environments,” in Job Scheduling Strategies for Parallel Processing Springer-Verlag, LNCS Vol. 1162, Apr 1996, pp. 111–125,1996.Google Scholar
  5. 5.
    M. S. Squillante, F. Wang, and M. Papaefthymiou, “An analysis of gang scheduling for multiprogrammed parallel computing environments,” in Proceedings of the Annual ACM Symposium on Parallel Algorithms and Architectures (SPAA), pp. 89–98, 1996.Google Scholar
  6. 6.
    H. Franke, P Pattnaik, and L. Rudolph, “Gang scheduling for highly efficient distributed multiprocessor systems,” in Proceedings of the 6th Symposium on the Frontiers of Massively Parallel Computation, Annapolis, MD, pp. 1–9, 1996.Google Scholar
  7. 7.
    D. R. Cox, “A use of complex probabilities in the theory of stochastic process,” Proc. Cambridge Philos. Soc, 1955.Google Scholar
  8. 8.
    M. F. Neuts, Matrix-Geometric Solutions in Stochastic Models. The John Hopkins University Press, 1981.Google Scholar
  9. 9.
    M. A. Johnson and M. R. Taaffe, “Matching moments to phase distributions,” Commun. Stat. Stochastics Models, vol. 5, pp. 711–743, 1989.Google Scholar
  10. 10.
    R. Schassberger, “Insensitivity of steady-state distributions of generalized semi-markov processes i,” Ann. Prob., pp. 87–99,1977.Google Scholar
  11. 11.
    R. Schassberger, “Insensitivity of steady-state distributions of generalized semi-markov processes ii,” Ann. Prob., pp. 85–93, 1978.Google Scholar
  12. 12.
    R. Schassberger, “Insensitivity of steady-state distributions of generalized semi-markov processes with speeds,” Adv. Appl. Prob., pp. 836–851,1978.Google Scholar
  13. 13.
    R. Schassberger, “The insensitivity of stationary probabilities in networks of queues,” Adv. Appl. Prob., pp. 906–912, 1978.Google Scholar

Copyright information

© Springer-Verlag 1997

Authors and Affiliations

  • Joefon Jann
    • 1
  • Pratap Pattnaik
    • 1
  • Hubertus Franke
    • 1
  • Fang Wang
    • 2
  • Joseph Skovira
    • 3
  • Joseph Riordan
    • 3
  1. 1.IBM T. J. Watson Research CenterYorktown Heights
  2. 2.Computer Science DepartmentYale UniversityNew Haven
  3. 3.Cornell Theory Center, Cornell UniversityIthaca

Personalised recommendations