Recent breakthroughs in the mathematical estimation of parallel genetic algorithm parameters by Cantu-Paz (2000) are applied to the NP-complete problem of scheduling multiple tasks on a cluster of computers connected by a shared bus. Experiments reveal that the parallel scheduling algorithm develops very accurate schedules when the parameter guidelines are used.
Published in:
Parallel and Distributed Processing Symposium, 2003. Proceedings. International
Date of Conference: 22-26 April 2003