Simulation Modeling and Analysis
Simulation Modeling and Analysis
An Object-Oriented Random-Number Package with Many Long Streams and Substreams
Operations Research
Classifying scheduling policies with respect to unfairness in an M/GI/1
SIGMETRICS '03 Proceedings of the 2003 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
A resource-allocation queueing fairness measure
Proceedings of the joint international conference on Measurement and modeling of computer systems
Theory, Volume 1, Queueing Systems
Theory, Volume 1, Queueing Systems
| Hi-index | 0.00 |
We apply simulation to evaluate the performance of scheduling according to an alternating FCFS/SJF service order, in particular concerning performance metrics other than delay. Metrics related to social justice, also referred to as quantitative fairness measures, are investigated. We focus on two types of situations that have been previously identified to provoke and increase queued customers' personal perceptions of social injustice: violations of the FCFS service discipline implying so-called slips and skips or overtaking, and long waiting for single customers. In order to balance the extreme behavior of pure FCFS and pure SJF service order and to improve perceived justice of customers, we propose to alternate between these traditional scheduling policies. We also examine the prize that has to be paid to improve justice oriented queue metrics at the cost of higher expected delay or higher delay variability. That is we compare the potential increase of expected delay and of delay variability to the potential increase of social justice. Discrete-event simulation studies are performed for different interarrival and service time distributions in general single server queues. Our results show that alternating FCFS/SJF scheduling can significantly improve social justice without much worsening the delay metrics.