The Journal of Supercomputing

, Volume 16, Issue 3, pp 217–235 | Cite as

Memory Conscious Scheduling for Cluster-based NUMA Multiprocessors

  • Takahiro Koita
  • Tetsuro Katayama
  • Keizo Saisho
  • Akira Fukuda
Article

Abstract

For Non-Uniform Memory Access (NUMA) multiprocessors, memory access overhead is crucial to system performance. Processor scheduling and page placement schemes, dominant factors of memory access overhead, are closely related. In particular, if the processor scheduling scheme is dynamic space-sharing, it should be considered together with the page placement scheme for efficient process execution. Most research in this area, however, has focused exclusively on either the processor scheduling scheme or the page placement scheme alone without considering the interaction between the two. This paper proposes several policies for cluster-based NUMA multiprocessors that are combinations of a processor scheduling scheme and a page placement scheme and investigates the interaction between them. The simulation results show that policies that cooperate to employ the home-cluster concept achieve the best performance. The paper also compares the best of the proposed policies with other existing dynamic processor scheduling policies. Based on our study reported here, the best policy is found to perform better than other existing policies.

processor scheduling dynamic space-sharing page placement cluster-based NUMA multiprocessors multiprogrammed environments 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    W. J. Bolosky, M. L. Scott, R. P. Fitzgerald, R. J. Fowler, and A. L. Cox. NUMA policies and their relation to memory architecture. In Proceedings of the 4th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), pp. 212–221. ACM, New York, 1991.Google Scholar
  2. 2.
    T. B. Brecht. An experimental evaluation of processor pool-based scheduling for shared-memory NUMA multiprocessors. In '97 Workshop, Job Scheduling Strategies for Parallel Processing, pp. 139–163. Springer-Verlag, New York, 1997.Google Scholar
  3. 3.
    R. Chandra, S. Devine, B. Verghese, and A. Gupta. Scheduling and page migration for multiprocessor compute servers. In Proceedings of the 6th International Conference on Architectural Support for Programming Language and Operating Systems (ASPLOS), pp. 12–24. ACM, New York, 1994.Google Scholar
  4. 4.
    M. Crovella, P. Das, C. Dubnicki, T. LeBlanc, and E. Markatos. Multiprogramming on multiprocessors. In Proceedings of the 3rd IEEE Symposium on Parallel and Distributed Processing, pp. 590–597. IEEE Computer Society Press, Los Alamitos, Calif., 1991.Google Scholar
  5. 5.
    D. G. Feitelson. A survey of scheduling in multiprogrammed parallel systems. Technical report, IBM Research Division, Yorktown Heights, NY, 1994.Google Scholar
  6. 6.
    A. Fukuda, R. Fujiki, and H. Kai. Two-level processor scheduling for multiprogrammed NUMA multiprocessors. In Proceedings of the 17th International Computer Software and Applications Conference, pp. 343–351. IEEE Computer Society Press, Los Alamitos, Calif., 1994.Google Scholar
  7. 7.
    A. Gupta, A. Tucker, and S. Urushibara. The impact of operating system scheduling policies and synchronization methods on the performance of parallel applications. In Proceedings of the 1991 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems, pp. 120–132. ACM, New York, 1991.Google Scholar
  8. 8.
    H. Kai, R. Fujiki, and A. Fukuda. Two-level scheduling in collaboration with memory management. In Proceedings of the 6th Joint Symposium on Parallel Processing (JSPP'94), pp. 319–326. Information Processing Society of Japan, 1994.Google Scholar
  9. 9.
    H. Kai, R. Fujiki, and A. Fukuda. Two-level scheduling for multiprogrammed multiprocessors--its structure and performance. Transactions of the Information Processing Society Japan, 35(10): 2115–2127, 1994.Google Scholar
  10. 10.
    T. Koita, T. Katayama, K. Saisho, and A. Fukuda. Performance evaluation of two-level processor scheduling in collaboration with memory management for NUMA multiprocessors. In Proceedings of the 16th IASTED International Conference on Applied Informatics (AI'98), pp. 265–272. ACTA Press, 1998.Google Scholar
  11. 11.
    T. Koita, T. Katayama, K. Saisho, and A. Fukuda. Processor scheduling with page placement for cluster-based NUMA multiprocessors. In Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA'98), pp. 851–858. CSREA Press, 1998.Google Scholar
  12. 12.
    R. P. LaRowe and C. S. Ellis. Experimental comparison of memory management policies for NUMA multiprocessors. ACM Transactions on Computer Systems, 9(4): 319–363, 1991.Google Scholar
  13. 13.
    R. P. LaRowe, C. S. Ellis, and L. S. Kaplan. The robustness of NUMA memory management. Operating System Review, 25(5): 137–151, 1991.Google Scholar
  14. 14.
    D. Lenoski, J. Laudon, T. Joe, D. Nakahira, L. Stevens, A. Gupta, and J. Hennessy. The DASH prototype: logic overhead and performance. IEEE Transactions on Parallel and Distributed Systems, 4(1): 41–61, 1993.Google Scholar
  15. 15.
    E. Markatos. Scheduling for locality in shared-memory multiprocessors. Ph.D. Thesis, University of Rochester, Rochester, NY, 1993.Google Scholar
  16. 16.
    C. McCann, R. Vaswanim, and J. Zahorjan. A dynamic processor allocation policy for multiprogrammed shared-memory multiprocessors. ACM Transactions on Computer Systems, 11(2): 146–178, 1993.Google Scholar
  17. 17.
    Y. Nobukuni, T. Matsumoto, and K. Hiraki. Resource management methods for general purpose massively parallel OS SSS-CORE. In Proceedings of the International Symposium on High Performance Computing, pp. 255–266. Springer-Verlag, New York, 1997.Google Scholar
  18. 18.
    Y. Ohishi, K. Saisho, and A. Fukuda. Performance evaluation of two-level scheduling algorithms for NUMA multiprocessors. IEICE Transactions on Information and System, J80–D1(1): 31–41, 1997.Google Scholar
  19. 19.
    J. Ramanathan and L. M. Ni. Critical factors in NUMA memory management. In Proceedings of the 11th International Conference on Distributed Computing Systems, pp. 500–507. IEEE Computer Society Press, Los Alamitos, Calif., 1991.Google Scholar
  20. 20.
    B. Verghese, S. Devine, A. Gupta, and M. Rosenblum, Operating system support for improving data locality on CC-NUMA compute servers. In Proceedings of the 7th International Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS), pp. 279–289. ACM, New York, 1996.Google Scholar
  21. 21.
    J. Zahorjan and C. McCann, Processor scheduling in shared memory multiprocessors. In Proceedings of the 1990 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems, pp. 214–215. ACM, New York, 1990.Google Scholar

Copyright information

© Kluwer Academic Publishers 2000

Authors and Affiliations

  • Takahiro Koita
    • 1
  • Tetsuro Katayama
    • 2
  • Keizo Saisho
    • 3
  • Akira Fukuda
    • 4
  1. 1.Nara Institute of Science and TechnologyIkoma, NaraJapan
  2. 2.Nara Institute of Science and TechnologyIkoma, NaraJapan
  3. 3.Nara Institute of Science and TechnologyIkoma, NaraJapan
  4. 4.Nara Institute of Science and TechnologyIkoma, NaraJapan

Personalised recommendations