Heterogeneous multi-processors scheduling by coevolutionary genetic algorithm
Qiuxi Zhong, Yue Qi · 2003
The efficacy and efficiency of conventional single population-based evolutionary algorithms (CEAs) decrease with the number of independent tasks in heterogeneous multi-processors systems. Based on computational model of cooperative coevolution, a task matching and scheduling algorithm is proposed for heterogeneous multi-processors systems, and the computation of individual's fitness is given according to the cooperative interactions among species. Simulation results show that the coevolutionary genetic algorithm is more effective than CEA for multi-tasks matching and scheduling, and the algorithm is of practical use in engineering.