基于用户满意度的网格资源调度遗传算法

Genetic Algorithm for Grid Resource Schedule Based on User Satisfaction

传统的网格资源调度研究注重调度的性能,很少考虑调度的服务质量。针对上述问题,设计3种类型的用户满意度评估方法衡量调度的服务质量,提出一种基于用户满意度的网格资源调度遗传算法,以用户满意度作为遗传变异的依据,实现网格资源调度过程的优化。实验表明,该算法能在保证较优调度性能的同时大幅度提高调度的服务质量。

Traditional research on grid resource schedule attaches importance to the performance, but rarely considers the quality of service. To scale the quality of schedule service, this paper designs an evaluating method for three kinds of user satisfaction, and proposes a Genetic Algorithm（GA） based on user satisfaction for grid resource schedule. The algorithm uses user satisfaction as the gist for gene evolution, and realizes the optimization of grid resource scheduling. Simulation results show that this algorithm can ensure a good schedule performance, and improves the quality of service.