Advertisement

OStrich: Fair Scheduling for Multiple Submissions

  • Joseph Emeras
  • Vinicius Pinheiro
  • Krzysztof Rzadca
  • Denis Trystram
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8385)

Abstract

Campaign Scheduling is characterized by multiple job submissions issued from multiple users over time. This model perfectly suits today’s systems since most available parallel environments have multiple users sharing a common infrastructure. When scheduling individually the jobs submitted by various users, one crucial issue is to ensure fairness. This work presents a new fair scheduling algorithm called OStrich whose principle is to maintain a virtual time-sharing schedule in which the same amount of processors is assigned to each user. The completion times in the virtual schedule determine the execution order on the physical processors. Then, the campaigns are interleaved in a fair way by OStrich. For independent sequential jobs, we show that OStrich guarantees the stretch of a campaign to be proportional to campaign’s size and the total number of users. The stretch is used for measuring by what factor a workload is slowed down relative to the time it takes on an unloaded system. The theoretical performance of our solution is assessed by simulating OStrich compared to the classical FCFS algorithm, issued from synthetic workload traces generated by two different user profiles. This is done to demonstrate how OStrich benefits both types of users, in contrast to FCFS.

Keywords

Job scheduling Fairness Job campaigns Multi-user Workload traces 

Notes

Acknowledgement

Krzysztof Rzadca is partly supported by Polish National Research Center SONATA grant UMO-2012/07/D/ST6/02440 Work partly supported by the French-Polish scientific cooperation program POLONIUM. Vinicius Pinheiro is partly supported by the CAPES/COFECUB Program (project number 4971/11-6).

References

  1. 1.
    Agnetis, A., Mirchandani, P.B., Pacciarelli, D., Pacifici, A.: Scheduling problems with two competing agents. Oper. Res. 52(2), 229–242 (2004)CrossRefzbMATHMathSciNetGoogle Scholar
  2. 2.
    Beaumont, O., Carter, L., Ferrante, J., Legrand, A., Marchal, L., Robert, Y.: Centralized versus distributed schedulers for bag-of-tasks applications. IEEE Trans. Parallel Distrib. Syst. 19(5), 698–709 (2008)CrossRefGoogle Scholar
  3. 3.
    Bender, M.A., Chakrabarti, S., Muthukrishnan, S.: Flow and stretch metrics for scheduling continuous job streams. In: Proceedings of the Ninth Annual ACM-SIAM Symposium on Discrete Algorithms, SODA’98, pp. 270–279. Society for Industrial and Applied Mathematics, Philadelphia, PA, USA (1998). http://dl.acm.org/citation.cfm?id=314613.314715
  4. 4.
    Bender, M.A., Muthukrishnan, S., Rajaraman, R.: Improved algorithms for stretch scheduling. In: Proceedings of the Thirteenth Annual ACM-SIAM Symposium on Discrete Algorithms, SODA’02, pp. 762–771. Society for Industrial and Applied Mathematics, Philadelphia, PA, USA (2002). http://dl.acm.org/citation.cfm?id=545381.545482
  5. 5.
    Bruno, J., Coffman, J.E.G., Sethi, R.: Scheduling independent tasks to reduce mean finishing time. Commun. ACM 17(7), 382–387 (1974). http://doi.acm.org/10.1145/361011.361064
  6. 6.
    Casanova, H., Desprez, F., Suter, F.: Minimizing stretch and makespan of multiple parallel task graphs via malleable allocations. In: 2010 39th International Conference on Parallel Processing (ICPP), September 2010, pp. 71–80 (2010)Google Scholar
  7. 7.
    Celaya, J., Marchal, L.: A fair decentralized scheduler for bag-of-tasks applications on desktop grids. In: 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing (CCGrid), May 2010, pp. 538–541 (2010)Google Scholar
  8. 8.
    Donassolo, B., Legrand, A., Geyer, C.: Non-cooperative scheduling considered harmful in collaborative volunteer computing environments. In: Proceedings of the 2011 11th IEEE/ACM International Symposium on Cluster, Cloud and Grid, Computing, CCGRID’11, pp. 144–153 (2011)Google Scholar
  9. 9.
    Emeras, J., Pinheiro, V., Rzadca, K., Trystram, D.: Fair scheduling for multiple submissions. Research Report RR-LIG-033, LIG, Grenoble, France (2012)Google Scholar
  10. 10.
    Feitelson, D.: Workload modeling for computer systems performance evaluation (2005). http://www.cs.huji.ac.il/feit/wlmod/wlmod.pdf
  11. 11.
    Ghodsi, A., Sekar, V., Zaharia, M., Stoica, I.: Multi-resource fair queueing for packet processing. ACM SIGCOMM Comput. Commun. Rev. 42(4), 1–12 (2012)CrossRefGoogle Scholar
  12. 12.
    Graham, R.L.: Bounds on multiprocessing timing anomalies. SIAM JOURNAL ON APPLIED MATHEMATICS 17(2), 416–429 (1969)CrossRefzbMATHMathSciNetGoogle Scholar
  13. 13.
    Iosup, A., Jan, M., Sonmez, O.O., Epema, D.H.J.: The Characteristics and Performance of Groups of Jobs in Grids. In: Kermarrec, A.-M., Bougé, L., Priol, T. (eds.) Euro-Par 2007. LNCS, vol. 4641, pp. 382–393. Springer, Heidelberg (2007) Google Scholar
  14. 14.
    Iosup, A., Sonmez, O., Anoep, S., Epema, D.: The performance of bags-of-tasks in large-scale distributed systems. In: Proceedings of the 17th International Symposium on High Performance Distributed Computing, pp. 97–108. ACM (2008)Google Scholar
  15. 15.
    Lee, C.Y.: Parallel machines scheduling with nonsimultaneous machine available time. Discrete Appl. Math. 30, 53–61 (1991). http://dx.doi.org/10.1016/0166-218X(91)90013-M
  16. 16.
    Legrand, A., Su, A., Vivien, F.: Minimizing the stretch when scheduling flows of biological requests. In: Proceedings of the Eighteenth Annual ACM Symposium on Parallelism in Algorithms and Architectures, SPAA’06, pp. 103–112. ACM, New York, NY, USA (2006). http://doi.acm.org/10.1145/1148109.1148124
  17. 17.
    Mehrzadi, D., Feitelson, D.G.: On extracting session data from activity logs. In: Proceedings of the 5th Annual International Systems and Storage Conference, SYSTOR’12, pp. 3:1–3:7 (2012)Google Scholar
  18. 18.
    Pinheiro, V., Rzadca, K., Trystram, D.: Campaign scheduling. In: IEEE International Conference on High Performance Computing (HiPC), Proceedings (2012, accepted for publication)Google Scholar
  19. 19.
    Raz, D., Levy, H., Avi-Itzhak, B.: A resource-allocation queueing fairness measure. SIGMETRICS Perform. Eval. Rev. 32(1), 130–141 (2004)CrossRefGoogle Scholar
  20. 20.
    Sabin, G., Kochhar, G., Sadayappan, P.: Job fairness in non-preemptive job scheduling. In: Proceedings of the 2004 International Conference on Parallel Processing, ICPP’04, pp. 186–194 (2004)Google Scholar
  21. 21.
    Saule, E., Trystram, D.: Multi-users scheduling in parallel systems. In: Proceedings of IEEE International Parallel and Distributed Processing Symposium, May 2009, pp. 1–9. Washington, DC, USA (2009)Google Scholar
  22. 22.
    Shmueli, E., Feitelson, D.: Using site-level modeling to evaluate the performance of parallel system schedulers. In: 14th IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, 2006, MASCOTS 2006, September 2006, pp. 167–178 (2006)Google Scholar
  23. 23.
    Wu, Y., Cao, G.: Stretch-optimal scheduling for on-demand data broadcasts. In: Proceedings of Tenth International Conference on Computer Communications and Networks, pp. 500–504 (2001)Google Scholar
  24. 24.
    Zakay, N., Feitelson, D.G.: On identifying user session boundaries in parallel workload logs. In: Proceedings of the 16th Workshop on Job Scheduling Strategies for Parallel Processing. The Hebrew University, Israel (May 2012). http://www.cs.huji.ac.il/feit/parsched/jsspp12/p12-zakay.pdf

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Joseph Emeras
    • 1
  • Vinicius Pinheiro
    • 2
  • Krzysztof Rzadca
    • 3
  • Denis Trystram
    • 4
    • 5
  1. 1.Laboratoire d’Informatique de Grenoble CEA - CNRSGrenobleFrance
  2. 2.Laboratory for Parallel and Distributed ComputingUniversity of São PauloSão PauloBrasil
  3. 3.Institute of InformaticsUniversity of WarsawWarsawPoland
  4. 4.Grenoble Institute of TechnologyGrenobleFrance
  5. 5.Institut Universitaire de FranceVesoulFrance

Personalised recommendations