CHARACTERISTICS OF FAIRNESS METRICS AND THEIR EFFECT ON PERCEIVED SCHEDULER EFFECTIVENESS

J. Ngubiri and M. van Vliet

References

  1. [1] D.G. Feitelson, L. Rudolph, E. Frachtenberg, &U. Schwiegelshohn (Eds.), The job scheduling strategies forparallel processing workshop series. http://www.cs.huji.ac.il/feit/parsched/
  2. [2] E. Frachtenberg & D.G. Feitelson, Pitfalls in parallel jobscheduling evaluation, Proc. 11th Workshop on Job SchedulingStrategies for Parallel Processing, LNCS 3834, Cambridge,MA, 2005, 257–282.
  3. [3] D.G. Feitelson, L. Rudolph, & U. Schwiegelshohn, Parallel jobscheduling – A status report, Proc. 10th Workshop on JobScheduling Strategies for Parallel Processing, LNCS 3277, NewYork, NY, 2004, 1–16.
  4. [4] R. Jain, D. Chiu, & W.R. Hawe, A quantitative measure offairness and discrimination for resource allocation in sharedcomputer systems, Technical Report TR-301, Digital Equip-ment Corporation, 1984.
  5. [5] G. Sabin, G. Kochar, & P. Sadayappan, Job fairness in non-preemptive job scheduling, Proc. 2004 International Conf. onParallel Processing, Montreal, Canada, 2004, 186–194.
  6. [6] G. Sabin & P. Sudayappan, Unfairness metrics for space-sharingparallel job schedulers, Proc. 11th Workshop on Job SchedulingStrategies for Parallel Processing, LNCS 3834, Cambridge,MA, 2005, 238–256.
  7. [7] U. Schwiegelshohn & R. Yahyapour, Fairness in parallel jobscheduling, Journal of Scheduling, 3 (5), 2000, 297–320.
  8. [8] S. Vasupongayya & S.-H. Chiang, On job fairness in non-preemptive parallel job scheduling. Proc. 17th IASTED Conf.on Parallel and Distributed Computing and Systems, Phoenix,AZ, 2005.
  9. [9] D. Raz, H. Levy, & B. Avi-itzak, A resource-allocation queu-ing fairness measure, SIGMETRICS Performance EvaluationReview, 32 (1), 2004, 130–141.
  10. [10] B. Avi-Itzhak, H. Levy, & D. Raz, David. A resource allocationqueueing fairness measure: Properties and bounds. QueueingSystems 56(2), 2007, 65–71.
  11. [11] D.G. Feitelson, Metric and workload effects on computersystems evaluation. Computers, 36 (9), 2003, 18–25.
  12. [12] J. Ngubiri & M. van Vliet, Group-wise performance evaluationof processor co-allocation in multi-cluster systems, Proc. 13thWorkshop on Job Scheduling Strategies for Parallel Processing,Seattle, WA, 2007, 1–13.
  13. [13] S. Srinivasan, R. Kettimuthu, V. Subramani, & P. Sadayap-pan, Selective reservation strategies for backfill job scheduling,Proc. 8th Workshop on Job Scheduling Strategies for Paral-lel Processing, LNCS 2537, Edinburgh, Scotland, UK, 2002,55–71.
  14. [14] S. Srinivasan, R. Kettimuthu, V. Subramani, & P. Sadayap-pan, Characterization of Backfilling strategies for parallel jobscheduling, Proc. 2002 Int. Conf. on Parallel Processing Work-shops, Vancouver, BC, Canada, 2002, 514–520.
  15. [15] J. Ngubiri & M. van Vliet, The impact of jobs’ physicalcharacteristics on their schedulability in a multi-cluster system,Proc. 2nd International Conf. on Computing and ICT Research,Makerere University – Kampala, Uganda, 2006, 39–49.
  16. [16] S. Srinivasan, S. Krishnamoorthy, & P. Sadayappan, A robustscheduling technology for moldable scheduling of parallel jobs,Proc. IEEE Conf. on Cluster Computing, Kowloon, HongKong, 2003, 92–99.
  17. [17] A. Rafaeli, E. Kedmi, D. Vsluli, & G. Barron, Queuesand fairness: A multiple study experimental investigation,http://queues-fairness.rafaeli.net/.
  18. [18] B. Avi-Itzhak & H. Levy, On measuring fairness in queues,Advances of Applied Probability, 36 (3), 2004, 667–971.
  19. [19] A.W. Mu’alem & D.G. Feitelson, Utilization, predictability,workloads, and user runtime estimates in scheduling the IBMSP2 with backfilling, IEEE Transaction Parallel & DistributedSystems, 12 (6), 2001, 529–543.
  20. [20] D. Lifka, The ANL/IBM SP scheduling system, Proc. 1stWorkshop on Job scheduling Strategies for Parallel Processing,LNCS 949 Santa Barbara, CA, 1995, 295–303.
  21. [21] E. Shmueli & D.G. Feitelson, Backfilling with lookahead tooptimize the performance of parallel job scheduling, Proc. 9thWorkshop on Job Scheduling Strategies for Parallel Processing,LNCS 2862, Seattle, WA, 2003, 228–251.
  22. [22] K. Aida, H. Kasahara, & S. Narita, Job scheduling scheme forpure space sharing among rigid jobs, Proc. 4th Workshop onJob Scheduling Strategies for Parallel Processing, LNCS 1459,Orlando, FL, 1998, 98–121.
  23. [23] S.-H. Chiang, A. Arpaci-Dusseau, & M.K. Vernon, The im-pact of more accurate requested runtimes on production jobscheduling performance, Proc. 8th Workshop on Job Schedul-ing Strategies for Parallel Processing, LNCS 2537, Edinburgh,UK, 103–127, 2002.
  24. [24] D. Tsafrir & D.G. Feitelson, Instability in parallel job schedul-ing simulation: The role of workload flurries, Proc. 20th In-ternational Parallel and Distributed Computing Symposium,Rhodes, Greece, 2006.
  25. [25] J. Ngubiri & M. van Vliet, Using the greedy approach toschedule jobs in a multi-cluster systems, Proc. InternationalConf. on Parallel and Distributed Processing Techniques andApplications, Las Vegas, NV, 2006, 663–668.

Important Links:

Go Back