Gang Scheduling Extensions for I/O Intensive Workloads

  • Yanyong Zhang
  • Antony Yang
  • Anand Sivasubramaniam
  • Jose Moreira
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2862)

Abstract

Scientific applications are becoming more complex and more I/O demanding than ever. For such applications, the system with dedicated I/O nodes does not provide enough scalability. Rather, a serverless approach is a viable alternative. However, with the serverless approach, a job’s execution time is decided by whether it is co-located with the file blocks it needs. Gang scheduling (GS), which is widely used in supercomputing centers to schedule parallel jobs, is completely not aware of the application’s spatial preferences.

In this paper, we show that gang scheduling does not do a good job scheduling I/O intensive applications. We extend gang scheduling by adding different levels of I/O awareness, and propose three schemes. We show that all these three new schemes are better than gang scheduling for I/O intensive jobs. One of them, with the help of migration, outperforms the others significantly for all the workloads we look at.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Anderson, T., Dahlin, M., Neefe, J., Roselli, D., Patterson, D., Wang, R.: Server-less Network File Systems. ACM Transactions on Computer System 14(l), 41–79 (1996)CrossRefGoogle Scholar
  2. 2.
    Bolosky, W.J., Douceur, J.R., Ely, D., Theimer, M.: Feasibility of a serverless distributed file system deployed on an existing set of desktop PCs. In: Proceedings of the ACM SIGMETRICS 2000 Conference on Measurement and Modeling of Computer Systems, pp. 34–43 (2000)Google Scholar
  3. 3.
    Corbett, P.F., Feitelson, D.G.: The Vesta parallel file system. A CM Transactions on Computer System 14(3), 225–264 (1996)CrossRefGoogle Scholar
  4. 4.
    Feitelson, D.G.: A Survey of Scheduling in Multiprogrammed Parallel Systems. Technical Report RC 19790 (87657), IBM T. J. Watson Research Center (October 1994)Google Scholar
  5. 5.
    Feitelson, D.G., Jette, M.A.: Improved Utilization and Responsiveness with Gang Scheduling. In: Feitelson, D.G., Rudolph, L. (eds.) IPPS-WS 1997 and JSSPP 1997. LNCS, vol. 1291, pp. 238–261. Springer, Heidelberg (1997)Google Scholar
  6. 6.
    Feitelson, D.G., Rudolph, L., Schwiegelshohn, U., Sevcik, K.C., Wong, P.: Theory and Practice in Parallel Job Scheduling. In: Feitelson, D.G., Rudolph, L. (eds.) IPPS-WS 1997 and JSSPP 1997. LNCS, vol. 1291, pp. 1–34. Springer, Heidelberg (1997)Google Scholar
  7. 7.
    Feitelson, D.G., Weil, A.M.: Utilization and predictability in scheduling the IBM SP2 with backfilling. In: 12th International Parallel Processing Symposium, (April 1998), pp. 542-546 (1998)Google Scholar
  8. 8.
    Franke, H., Jann, J., Moreira, J.E., Pattnaik, P.: An Evaluation of Parallel Job Scheduling for ASCI Blue-Pacific. In: Proceedings of SC 1999, Portland, OR, IBM Research Report RC21559 (November 1999)Google Scholar
  9. 9.
    Franke, H., Pattnaik, P., Rudolph, L.: Gang Scheduling for Highly Efficient Multiprocessors. In: Sixth Symposium on the Frontiers of Massively Parallel Computation, Annapolis, Maryland (1996)Google Scholar
  10. 10.
    Gorda, B., Wolski, R.: Time Sharing Massively Parallel Machines. In: International Conference on Parallel Processing, August 1995, vol. II, pp. 214–217 (1995)Google Scholar
  11. 11.
    Islam, N., Prodromidis, A.L., Squillante, M.S., Fong, L.L., Gopal, A.S.: Extensible Resource Management for Cluster Computing. In: Proceedings of the 17th International Conference on Distributed Computing Systems, pp. 561–568 (1997)Google Scholar
  12. 12.
    Jann, J., Pattnaik, P., Franke, H., Wang, F., Skovira, J., Riordan, J.: Modeling of Workload in MPPs. In: Proceedings of the 3rd Annual Workshop on Job Scheduling Strategies for Parallel Processing, (April 1997), pp. 95–116, InConjuction with IPPS1997, Geneva, Switzerland (1997)Google Scholar
  13. 13.
    Karatza, H.D.: A Simulation-Based Performance Analysis of Gang Scheduling in a Distributed System. In: Proceedings 32nd Annual Simulation Symposium, San Diego, CA, April 11-15, pp. 26–33 (1999)Google Scholar
  14. 14.
    Lifka, D.: The ANL/IBM SP scheduling system. In: Feitelson, D.G., Rudolph, L. (eds.) IPPS-WS 1995 and JSSPP 1995. LNCS, vol. 949, pp. 295–303. Springer, Heidelberg (1995)Google Scholar
  15. 15.
    Ma, X., Jiao, J., Campbell, M., Winslett, M.: Flexible and efficient parallel i/o for large-scale multi-component simulations. In: Proceedings of The 4-th Workshop on Parallel and Distributed Scientific and Engineering Computing with Applications, in conjunction with the 2003 International Parallel and Distributed Processing Symposium (2003)Google Scholar
  16. 16.
    Moreira, J.E., Franke, H., Chan, W., Fong, L.L., Jette, M.A., Yoo, A.: A Gang-Scheduling System for ASCI Blue-Pacific. In: Sloot, P.M.A., Hoekstra, A.G., Bubak, M., Hertzberger, B. (eds.) HPCN-Europe 1999. LNCS, vol. 1593, pp. 831–840. Springer, Heidelberg (1999)CrossRefGoogle Scholar
  17. 17.
    Nieuwejaar, N., Kotz, D.: Low-level interfaces for high-level parallel I/O. In: Proceedings of the IPPS 1995 Workshop on Input/Output in Parallel and Distributed Systems,(April 1995), pp. 47–62 (1995)Google Scholar
  18. 18.
    Ousterhout, J.K.: Scheduling Techniques for Concurrent Systems. In: Third International Conference on Distributed Computing Systems, pp. 22–30 (1982)Google Scholar
  19. 19.
    Schwiegelshohn, U., Yahyapour, R.: Improving First-Come-First-Serve Job Scheduling by Gang Scheduling. In: IPPS 1998 Workshop on Job Scheduling Strategies for Parallel Processing (March 1998)Google Scholar
  20. 20.
    Skovira, J., Chan, W., Zhou, H., Lifka, D.: The EASY-LoadLeveler API project. In: Feitelson, D.G., Rudolph, L. (eds.) IPPS-WS 1996 and JSSPP 1996. LNCS, vol. 1162, pp. 41–47. Springer, Heidelberg (1996)CrossRefGoogle Scholar
  21. 21.
    Suzaki, K., Walsh, D.: Implementation of the Combination of Time Sharing and Space Sharing on AP/Linux. In: IPPS 1998 Workshop on Job Scheduling Strategies for Parallel Processing (March 1998)Google Scholar
  22. 22.
    Wiseman, Y., Feitelson, D.G.: Paired Gang Scheduling. IEEE Transactions on Parallel and Distributed Systems.Google Scholar
  23. 23.
    Yue, K.K., Lilja, D.J.: Comparing Processor Allocation Strategies in Multi-programmed Shared-Memory Multiprocessors. Journal of Parallel and Distributed Computing 49(2), 245–258 (1998)MATHCrossRefGoogle Scholar
  24. 24.
    Zhang, Y., Franke, H., Moreira, J., Sivasubramaniam, A.: The Impact of Migration on Parallel Job Scheduling for Distributed Systems. In: Bode, A., Ludwig, T., Karl, W.C., Wismüller, R. (eds.) Euro-Par 2000. LNCS, vol. 1900, pp. 245–251. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  25. 25.
    Zhang, Y., Franke, H., Moreira, J., Sivasubramaniam, A.: Improving Parallel Job Scheduling by Combining Gang Scheduling and Backfilling Techniques. In: Proceedings of the International Parallel and Distributed Processing Symposium,(May 2000), pp. 133–142 (2000)Google Scholar
  26. 26.
    Zhang, Y., Franke, H., Moreira, J., Sivasubramaniam, A.: An Integrated Approach to Parallel Scheduling Using Gang-Scheduling, Backfilling and Migration. IEEE Transactions on Parallel and Distributed Systems 14(3), 236–247 (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Yanyong Zhang
    • 1
  • Antony Yang
    • 1
  • Anand Sivasubramaniam
    • 2
  • Jose Moreira
    • 3
  1. 1.Department of Electrical & Computer Engg.Rutgers, The State University of New JerseyPiscatawayUSA
  2. 2.Department of Computer Science & Engg.The Pennsylvania State UniversityUniversity ParkUSA
  3. 3.IBM T. J. Watson Research CenterYorktown HeightsUSA

Personalised recommendations