To obtain access to full text of journal and articles you must register!
- Article name
- PARALLEL GENETIC ALGORITHM USAGE IN MULTIPROCESSOR COMPUTING SYSTEM JOB SCHEDULING
- Authors
- Abakumov E. M., , abakumov@vniia.ru, Federal State Unitary Enterprise "All-Russia Research Institute of Automatics", Moscow, Russia
Petunin S. A., , petunin@vniia.ru, FSUE "All-Russian Research Institute of Automatics named after N. L. Duhov", Moscow, Russia
Novikov A. B., , alexey.b.novikov@gmail.com, FSUE "All-Russian Research Institute of Automatics named after N. L. Duhov", Moscow, Russia
- Keywords
- metaheuristic / genetic algorithm / parallel algorithm / scheduling
- Year
- 2015 Issue 4 Pages 8 - 12
- Code EDN
- Code DOI
- Abstract
- This article describes current issues with user jobs control in high performance computing system. User job extended interpretation cannot be effectively processed by simple heuristics. Special parallel genetic algorithm has been developed to resolve this issue. It consider jobs and system addition information. This information allow to scheduling quality. Article contents descriptions of several approaches to design parallel genetic algorithm. Fitness function is based on simulation of system work with user application models.
- Text
- To obtain access to full text of journal and articles you must register!
- Buy
