Intelligent job selection for distributed scheduling
- 30 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
A key issue in distributed scheduling is selecting appropriate jobs to transfer. A job selection policy that considers the diversity of job behaviors is proposed. A mechanism used in artificial neural networks, called weight climbing, is employed. Using this mechanism, a distributed scheduler can learn the behavior of a job from its past executions and make a correct prediction about whether transferring the job is worthwhile. A scheduler using the proposed job selection policy has been implemented and experimental results show that it is able to learn job behaviors fast, make decisions accurately and adjust itself promptly when system configuration or program behaviors are changed. In addition, the selection policy introduces only negligible time and space overhead.Keywords
This publication has 12 references indexed in Scilit:
- A comparison of preemptive and non-preemptive load distributingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- The Stealth distributed schedulerPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Two adaptive location policies for global scheduling algorithmsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- History, an intelligent load sharing filterPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Transparent process migration: Design alternatives and the sprite implementationSoftware: Practice and Experience, 1991
- The limited performance benefits of migrating active processes for load sharingPublished by Association for Computing Machinery (ACM) ,1988
- Adaptive load sharing in homogeneous distributed systemsIEEE Transactions on Software Engineering, 1986
- A comparison of receiver-initiated and sender-initiated adaptive load sharingPerformance Evaluation, 1986
- Simulations of three adaptive, decentralized controlled, job scheduling algorithmsComputer Networks (1976), 1984
- Landmark learning: An illustration of associative searchBiological Cybernetics, 1981