Unfairness Metrics for Space-Sharing Parallel Job Schedulers
Sociology, computer networking and operations research provide evidence of the importance of fairness in queuing disciplines. Currently, there is no accepted model for characterizing fairness in parallel job scheduling. We introduce two fairness metrics intended for parallel job schedulers, both of which are based on models from sociology, networking, and operations research. The first metric is motivated by social justice and attempts to measure deviation from arrival order, which is perceived as fair by the end user. The second metric is based on resource equality and compares the resources consumed by a job with the resources deserved by the job. Both of these metrics are orthogonal to traditional metrics, such as turnaround time and utilization.
The proposed fairness metrics are used to measure the unfairness for some typical scheduling policies via simulation studies. We analyze the fairness of these scheduling policies using both metrics, identifying similarities and differences.
KeywordsSocial Justice Schedule Policy Arrival Order Resource Equality First Come First Serve
Unable to display preview. Download preview PDF.
- 1.Talby, D., Feitelson, D.: Supporting priorities and improving utilization of the IBM SP scheduler using slack-based backfilling. In: Proceedings of the 13th International Parallel Processing Symposium (1999)Google Scholar
- 5.Feitelson, D.: Workshops on job scheduling strategies for parallel processing, http://www.cs.huji.ac.il/~feit/parsched/
- 7.Srinivasan, S., Kettimuthu, R., Subramani, V., Sadayappan, P.: Characterization of backfilling strategies for job scheduling. In: 2002 Intl. Workshops on Parallel Processing (2002); held in conjunction with the 2002 Intl. Conf. on Parallel Processing, ICPP 2002Google Scholar
- 8.Raz, D., Levy, H., Avi-Itzhak, B.: A resource-allocation queueing fairness measure. In: Proceedings of Sigmetrics 2004/Performance 2004 Joint Conference on Measurement and Modeling of Computer Systems, New York, NY, pp. 130–141 (2004); Also appears as Performance Evaluation Review Special Issue 32(1), 130–141Google Scholar
- 9.Avi-Itzhak, B., Levy, H., Raz, D.: Quantifying fairness in queueing systems: Principles and applications. Technical Report RRR-26-2004, RUTCOR, Rutgers University (2004)Google Scholar
- 10.Raz, D., Levy, H., Avi-Itzhak, B.: RAQFM: a resource allocation queueing fairness measure. Technical Report RRR-32-2004, RUTCOR, Rutgers University (2004)Google Scholar
- 13.Gordon, E.S.: Slips and Skips in Queues. PhD thesis, Massachusetts Institute of Technology (1989)Google Scholar
- 15.Rafaeli, A., Kedmi, E., Vashdi, D., Barron, G.: Queues and fairness: A multiple study experimental investigation, http://queues-fairness.rafaeli.net/
- 17.Demers, A., Keshav, S., Shenker, S.: Analysis and simulation of a fair queueing algorithm. Internetworking Research and Experience 1, 3–26 (1990)Google Scholar
- 19.Wierman, A., Harchol-Balter, M.: Classifying scheduling policies with respect to unfairness in an M/GI/1. In: Proceedings of the 2003 ACM SIGMETRICS international conference on Measurement and modeling of computer systems, pp. 238–249 (2003)Google Scholar
- 20.Bansal, N., Harcol-Balter, M.: Analysis of SRPT scheduling: Investigating unfairness. In: SIGMETRICS (2001)Google Scholar
- 21.Harchol-Balter, M., Sigman, K., Wierman, A.: Asymptotic convergence of scheduling policies with respect to slowdown. In: IFIP WG 7.3 International Symposium on Computer Modeling, Measurement and Evaluation (2002)Google Scholar
- 23.Sabin, G., Sahasrabudhe, V., Sadayappan, P.: On fairness in distributed job scheduling across multiple sites. In: Cluster (2004)Google Scholar
- 24.Sabin, G., Kochhar, G., Sadayappan, P.: Job fairness in non-preemptive job scheduling. In: International Conference on Parallel Processesing (2004)Google Scholar
- 25.Feitelson, D.G.: Logs of real parallel workloads from production systems, http://www.cs.huji.ac.il/labs/parallel/workload/
- 26.Hansen, B.: An analysis of response ratio. In: IFIP Congress (1971)Google Scholar
- 27.Weisstein, E.W.: Spearman rank correlation coefficient, http://mathworld.wolfram.com/SpearmanRankCorrelationCoefficient.html From MathWorld–A Wolfram Web Resource