Objective-oriented algorithm for job scheduling in parallel heterogeneous systems

  • Pham Hong Hanh
  • Valery Simonenko
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1291)


This paper presents a new approach to solve the problem of job scheduling for parallel processing in heterogeneous systems. The optimization goals are: (i) minimum total execution time including communication costs and (ii) shortest response time for all jobs. We introduce a classification for the given scheduling problem by the heterogeneity of the systems, from the view of the schedulers' eyes. Then, according to this analysis, a new scheduling strategy for so-called “Strictly-Heterogeneous” systems is proposed. The key idea of the new approach is the use of the Hungarian method, which provides a quick and objective-oriented search for the best schedule by the given optimization criteria. In addition, by modifying this method into so-called Objective-Oriented Algorithm (OOA), the time complexity for scheduling is decreased to O(n(E+nlogn)). The simulation results show us that OOA provides better solution quality while scheduling time is less than the existing methods.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    C. Berge, Theorie des graphes et ses application, Dunod, Paris, 1958.Google Scholar
  2. [2]
    J. Blazevicz, M. Drozdowski, G. Schmidt, and D. De Werra, “Scheduling independent multiprocessor tasks on a uniform k-processor system”, Journal of Parallel Computer 20, pp. 15–28, 1994.CrossRefGoogle Scholar
  3. [3]
    T. Bultan and C. Aykanat, “A new mapping heuristic based on mean field annealing”, Journal of Parallel and Distributed Computing, Vol. 16, N4, December 1992.Google Scholar
  4. [4]
    T.L. Casavant and J.G. Kuhl, “A taxonomy of scheduling in general-purpose distributed computing systems”, IEEE Trans. Softw.Eng.14, pp. 141–154, 1988.CrossRefGoogle Scholar
  5. [5]
    K. Efe, “Heuristic models for task assignment scheduling in distributed systems”, IEEE Computer, June 1982.Google Scholar
  6. [6]
    H. El-Rewini and T.G. Lewis, “Scheduling Parallel tasks onto Arbitrary Target Machines”, Journal of Par. and Distr. Com.,Vol.9, pp. 138–153, 1990.CrossRefGoogle Scholar
  7. [7]
    A. A. Elsadek and B.E Wells, “Heuristic model for task allocation in a heterogeneous distributed systems”, Proceeding of PDPTA'96, California USA, Vol.2, pp. 659–671, August 1996.Google Scholar
  8. [8]
    R. F Freund, B.R.Carter, Daniel Watson, et al., “Generational Scheduling for Heterogeneous Computing Systems”, Proceeding of PDPTA'96, California-USA, Vol.2, pp769–778, August 1996.Google Scholar
  9. [9]
    M.R. Garey and D.S. Johnson, Computer and Intractability—A guide to the Theory of NP-completeness, Freeman New York, 1979.Google Scholar
  10. [10]
    A. Kaufmann, Introduction a la combinatorique en vue des aplications, Dunod, Paris, 1968.Google Scholar
  11. [11]
    S. Kirkpatrick, C.D. Gelatt, and M.P. Vecchi, “Optimization by simulated annealing”, Journal of Science, Vol.220, N.4589, May 1983.Google Scholar
  12. [12]
    X. Papadimitry, K. Stayglitsh, Combinatory optimization, algorithm and complexity, Moscow-Mir, 1985.Google Scholar
  13. [13]
    Hanh H. Pham and Valery Simonenko, “A new algorithm and simulation for task assignment in parallel distributed systems”, Proceeding of the l1 th European Simulation Multiconference '96, Budapest-Hungary, pp. 95–99, June 1996.Google Scholar
  14. [14]
    Hanh H. Pham and Valery Simonenko, “Adaptation of algorithms for Job-Resource Assignment in Heterogeneous Distributed Systems”, Proceeding of PDPTA'96, California-USA, Vol.2, pp. 835–845, August 1996.Google Scholar
  15. [15]
    Riedl Reinhard and Richter Lutz, “Classification of Load Distribution Algorithms”, Proceeding of IEEE PDP'96, pp. 404–413, 1996.Google Scholar
  16. [16]
    P.Shroff, D.W Watson, N.F. Flann, and R.F. Freund, “Genetic simulated annealing for scheduling data-dependent tasks in heterogeneous environments”, Proceeding of Heterogeneous Computing Workshop '96, pp.98–104, April 1996.Google Scholar
  17. [17]
    M. Tan, J.K Antonio, et. al., “Scheduling and data relocation for sequentially executed subtasks in a heterogeneous computing system”, Proceeding of Heterogeneous Computing Workshop '95, pp 109–120, 1995.Google Scholar
  18. [18]
    Salleh Shaharuddin et. al., “A Mean-field Annealing Model For Task Scheduling in Multi-processor Systems”, Proceedings of PDPTA'96, California-USA, Vol.2, pp. 189–198, August 1996.Google Scholar
  19. [19]
    Shen S. Wu and David Sweeting, “Heuristic algorithms for task assignment and scheduling in a processor network”, Journal of Parallel Computing 20, pp. 1–14, 1994.CrossRefGoogle Scholar
  20. [20]
    Honbo Zhou, Scheduling DAGs on a Bounded number of Processors, Proceedings of PDPTA'96, Vol.2, pp. 823–834, August 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Pham Hong Hanh
    • 1
  • Valery Simonenko
    • 1
  1. 1.Department of Computer ScienceNational Technical University of UkraineKievUkraine

Personalised recommendations