Job fairness in non-preemptive job scheduling

Sabin, Gerald; Kochhar, G.; Sadayappan, P.

doi:10.1109/icpp.2004.1327920

Cited by 35 publications

(29 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is due to the scheduler allowing a job to backfill. The apparently benign backfill can delay earlier jobs when other running jobs have not accurately estimated their runtimes [25]. The unfairness in a SJF schedule is possibly caused by temporarily starving some of the longer jobs while giving preferential treatment to shorter jobs.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Unfairness Metrics for Space-Sharing Parallel Job Schedulers

Sabin

Sadayappan

2005

Job Scheduling Strategies for Parallel Processing

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Fair start time (FST) [25] based metrics are proposed as a means to measure social justice for parallel job schedulers. In order to motivate our approach, we use an analogy.…”

Section: Fair Start Time Analysismentioning

confidence: 99%

Unfairness Metrics for Space-Sharing Parallel Job Schedulers

Sabin

Sadayappan

2005

Job Scheduling Strategies for Parallel Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…The approach in [5] used only ∆R 3 k to evaluate fairness. Like shown in (10), other factors are also involved.…”

Section: Fairness By Fair Start Time Analysismentioning

confidence: 99%

“…This is because in some cases, good 'performance' is got by starving some jobs at the expense of others (unfairness). Measures of dispersion [6][4], Fair start-time analysis [5] and the Resource Allocation Queuing Measure (RAQFM) [17] [18] are common fairness metrics. These measures can, sometimes, give misleading deductions for parallel job schedulers.…”

Section: Introductionmentioning

confidence: 99%

Characteristics of Fairness Metrics and Their Effect on Perceived Scheduler Effectiveness

Ngubiri

Vliet

2010

International Journal of Computers and Applications

View full text Add to dashboard Cite

Parallel job schedulers are mostly evaluated using performance metrics. Deductions however can be misleading due to selective job starvation (unfairness). To choose a better scheduler, therefore, there is a need to compare schedulers for fairness as well. Performance and fairness, however, have mostly been studied independently. We examine characteristics of three approaches to fairness evaluation in parallel job scheduling. We examine how they represent job starvation and other aspects of discrimination. We show that the implied unfairness is not always starvation/discrimination in practice. We use simultaneous consideration of performance and fairness and compare deductions with scheduler effectiveness derived from group-wise performance evaluation. We observe that due to possible misrepresentation of starvation by fairness metrics, schedulers shown as superior may not be so in practice.

show abstract

“…al. [25,26,27] have introduced multiple fairness metrics for parallel job schedulers. The first metric is based on defining a fair start time (FST), similar to the CONS P metric defined above.…”

Section: Fairness Metrics For Parallel Job Schedulingmentioning

confidence: 99%

Parallel job scheduling policies to improve fairness : a case study.

Leung

Sabin

Sadayappan

2008

View full text Add to dashboard Cite

Balancing fairness, user performance, and system performance is a critical concern when developing and installing parallel schedulers. Sandia uses a customized scheduler to manage many of their parallel machines. A primary function of the scheduler is to ensure that the machines have good utilization and that users are treated in a "fair" manner. A separate compute process allocator (CPA) ensures that the jobs on the machines are not too fragmented in order to maximize throughput.Until recently, there has been no established technique to measure the fairness of parallel job schedulers. This paper introduces a "hybrid" fairness metric that is similar to recently proposed metrics. The metric uses the Sandia version of a "fairshare" queuing priority as the basis for fairness. The hybrid fairness metric is used to evaluate a Sandia workload. Using these results, multiple scheduling strategies are introduced to improve performance while satisfying user and system performance constraints.3

show abstract

Job fairness in non-preemptive job scheduling

Cited by 35 publications

References 15 publications

Unfairness Metrics for Space-Sharing Parallel Job Schedulers

Unfairness Metrics for Space-Sharing Parallel Job Schedulers

Characteristics of Fairness Metrics and Their Effect on Perceived Scheduler Effectiveness

Parallel job scheduling policies to improve fairness : a case study.

Contact Info

Product

Resources

About