Article contents
Optimal dynamic scheduling of a general class of parallel-processing queueing systems
Published online by Cambridge University Press: 01 July 2016
Abstract
In this paper we develop policies for scheduling dynamically arriving jobs to a broad class of parallel-processing queueing systems. We show that in heavy traffic the policies asymptotically minimize a measure of the expected system backlog, which we call system work. Our results yield succinct, closed-form expressions for optimal system work in heavy traffic.
Keywords
MSC classification
- Type
- General Applied Probability
- Information
- Copyright
- Copyright © Applied Probability Trust 1998
Footnotes
This is a revision of a working paper dated 17 November 1995.
References
- 4
- Cited by