Time-Sharing Parallel Jobs in the Presence of Multiple Resource Requirements

  • Fabrizio Petrini
  • Wu-chun Feng
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1911)

Abstract

Buffered coscheduling is a new methodology that can substantially increase resource utilization, improve response time, and simplify the development of the run-time support in a parallel machine. In this paper, we provide an in-depth analysis of three important aspects of the proposed methodology: the impact of the communication pattern and type of synchronization, the impact of memory constraints, and the processor utilization. The experimental results show that if jobs use non-blocking or collectivecommunication patterns, the response time becomes largely insensitive to the job communication pattern. Using a simple job access policy, we also demonstrate the robustness of bu.ered coscheduling in the presence of memory constraints. Overall, bu.ered coscheduling generally outperforms back.lling and back.lling gang scheduling with respect to response time, wait time, run-time slowdown, and processor utilization. Keywords: Parallel Job Scheduling, Distributed Operating Systems, Communication Protocols, Performance Evaluation.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Andrea C. Arpaci-Dusseau, David Culler, and Alan M. Mainwaring. Scheduling with Implicit Information in Distributed Systems. In Proceedings of the 1998 ACM Sigmetrics International Conference on Measurement and Modeling of Computer Systems, Madison, WI, June 1998.Google Scholar
  2. 2.
    Douglas C. Burger and David A. Wood. Accuracy vs. Performance in Parallel Simulation of Interconnection Networks. In Proceedings of the 9th International Parallel Processing Symposium, IPPS’95, Santa Barbara, CA, April 1995.Google Scholar
  3. 3.
    Keith Diefendor. Compaq Chooses SMT for Alpha: Simultaneous Multithreading Exploits Instruction-and Thread-Level Parallelism. Microprocessor Report, 13(16), December 1999.Google Scholar
  4. 4.
    Andrea C. Dusseau, Remzi H. Arpaci, and David E. Culler. Effective Distributed Scheduling of Parallel Workloads. In Proceedings of the 1996A CM Sigmetrics International Conference on Measurement and Modeling of Computer Systems, Philadelphia, PA, May 1996.Google Scholar
  5. 5.
    Susan J. Eggers, Henry M. Levy, and Jack L. Lo. Multithreading: A Platform for Next-Generation Processors. IEEE Micro, 17(5), September/October 1997.Google Scholar
  6. 6.
    Fabrizio Petrini and Wu-chun Feng. Buffered Coscheduling: A New Methodology for Multitasking Parallel Jobs on Distributed Systems. In Proceedings of the International Parallel and Distributed Processing Symposium 2000, IPDPS2000, Cancun, MX, May 2000.Google Scholar
  7. 7.
    Dror G. Feitelson and Morris A. Jette. Improved Utilization and Responsiveness with Gang Scheduling. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 1291 of Lecture Notes in Computer Science. Springer-Verlag, 1997.Google Scholar
  8. 8.
    Dror G. Feitelson and Larry Rudolph. Parallel Job Scheduling: Issues and Approaches. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 949 of Lecture Notes in Computer Science. Springer-Verlag, 1995.Google Scholar
  9. 9.
    Dror G. Feitelson and Larry Rudolph. Toward Convergence in Job Schedulers for Parallel Supercomputers. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 1162 of Lecture Notes in Computer Science. Springer-Verlag, 1996.Google Scholar
  10. 10.
    Alex Gerbessiotis and Fabrizio Petrini. Network Performance Assessment under the BSP Model. In International Workshop on Constructive Methods for Parallel Programming, CMPP’98, Marstrand, Sweden, June 1998.Google Scholar
  11. 11.
    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, pages 120– 132, May 1991.Google Scholar
  12. 12.
    Anshul Gupta and Vipin Kumar. The Scalability of FFT on Parallel Computers. IEEE Transactions on Parallel and Distributed Systems, 4(8):922– 932, August 1993.CrossRefGoogle Scholar
  13. 13.
    Adolfy Hoisie, Olaf Lubeck, and Harvey Wasserman. Scalability Analysis of Multidimensional Wavefront Algorithms on Large-Scale SMP Clusters. In The Ninth Symposium on the Frontiers of Massively Parallel Computation (Frontiers’99), Annapolis, MD, February 1999.Google Scholar
  14. 14.
    Joefon Jann, Pratap Pattnaik, Hubertus Franke, Fang Wang, Joseph Skovira, and Joseph Riordan. Modeling of Workload in MPPs. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 1291 of Lecture Notes in Computer Science, pages 95– 116. Springer-Verlag, 1997.Google Scholar
  15. 15.
    Vijay Karamcheti and Andrew A. Chien. Do Faster Routers Imply Faster Communication? In First International Workshop, PCRCW’94, volume 853 of LNCL, pages 1– 15, Seattle, Washington, USA, May 1Google Scholar
  16. 16.
    Stephen W. Keckler, Andrew Chang, Whay S. Lee, Sandeep Chatterje, and William J. Dally. Concurrent Event Handling through Multithreading. IEEE Transactions on Computers, 48(9):903– 916, September 1999.Google Scholar
  17. 17.
    Mario Lauria and Andrew Chien. High-Performance Messaging on Workstations: Illinois Fast Messages (FM) for Myrinet. In Proceedings of Supercomputing ’95, November 1995.Google Scholar
  18. 18.
    Walter Lee, Matthew Frank, Victor Lee, Kenneth Mackenzie, and Larry Rudolph. Implications of I/O for Gang Scheduled Workloads. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 1291 of Lecture Notes in Computer Science. Springer-Verlag, 1997.Google Scholar
  19. 19.
    Shailabh Nagar, Ajit Banerjee, Anand Sivasubramaniam, and Chita R. Das. A Closer Look At Coscheduling Approaches for a Network of Workstations. In Eleventh ACM Symposium on Parallel Algorithms and Architectures, SPAA’99, Saint-Malo, France, June 1999.Google Scholar
  20. 20.
    Fabrizio Petrini. Network Performance with Distributed Memory Scientific Applications. Submitted to the Journal of Parallel and Distributed Computing, September 1998.Google Scholar
  21. 21.
    Fabrizio Petrini. Total-Exchange on Wormhole k-ary n-cubes with Adaptive Routing. In Proceedings of the 12th International Parallel Processing Symposium, IPPS’98, Orlando, FL, March 1998.Google Scholar
  22. 22.
    Fabrizio Petrini and Marco Vanneschi. Efficient Personalized Communication on Wormhole Networks. In The 1997 International Conference on Parallel Architectures and Compilation Techniques, PACT’97, San Francisco, CA, November 1997.Google Scholar
  23. 23.
    Fabrizio Petrini and Marco Vanneschi. Latency and Bandwidth Requirements of Massively Parallel Programs: FFT as a Case Study. Future Generation Computer Systems, 1999. Accepted for publication.Google Scholar
  24. 24.
    Ian R. Philp and Y. Liong. The Scheduled Transfer (ST) Protocol. In Proceedings of Workshop on Communication, Architecture, and Applications for Network-based Parallel Computing, January 1999.Google Scholar
  25. 25.
    D. B. Skillicorn, Jonathan M. D. Hill, and W. F. McColl. Questions and Answers about BSP. Journal of Scientific Programming, 1998.Google Scholar
  26. 26.
    Joseph Skovira, Waiman Chan, Honbo Zhou, and David Lifka. The EASYLoadLeveler API Project. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 1162 of Lecture Notes in Computer Science, pages 41– 47. Springer-Verlag, 1996.Google Scholar
  27. 27.
    Patrick Sobalvarro, Scott Pakin, William E. Weihl, and Andrew A. Chien. Dynamic Coscheduling on Workstation Clusters. In Dror G. Feitelson and Larry Rudolph, editors, Job Scheduling Strategies for Parallel Processing, volume 1459 of Lecture Notes in Computer Science, pages 231– 256. Springer-Verlag, 1998.Google Scholar
  28. 28.
    Patrick Sobalvarro and William E. Weihl. Demand-Based Coscheduling of Parallel Jobs on Multiprogrammed Multiprocessors. In Proceedings of the 9th International Parallel Processing Symposium, IPPS’95, Santa Barbara, CA, April 1995.Google Scholar
  29. 29.
    Leslie G. Valiant. A Bridging Model for Parallel Computation. Communications of the ACM, 33(8):103– 111, August 1990.CrossRefGoogle Scholar
  30. 30.
    Yanyong Zhang, Hubertus Franke, Josée Moreira, and Anand Sivasubramaniam. Improving Parallel Job Scheduling by Combining Gang Scheduling and Backfilling Techniques. In Proceedings of the International Parallel and Distributed Processing Symposium 2000, IPDPS2000, Cancun, MX, May 2000.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Fabrizio Petrini
    • 1
  • Wu-chun Feng
    • 1
  1. 1.Computing, Information, & Communications DivisionLos Alamos National LaboratoryUSA

Personalised recommendations