混合资源调度的遗传算法研究

The Research on the Genetic Algorithms of Mixed Resource Scheduling

针对多机群协同计算背景下的混合资源调度的NP难题,采用一种新型的遗传算法和编码规则,较好解决分布式并行计算环境下,多任务与多资源的调度分配问题,并给出了实例计算,结果表明,该算法能较好地解决资源之间的冲突问题,而且遗传算法操作简单,每个染色体都对应一个可行调度,同时避免了非法染色体的产生。

To NP problems of the mixed resource scheduling, using a new type of genetic algorithms and coding rules can better solve the problem of the multi-task scheduling and the allocation of resources under the distributed parallel computing environment, and also an example was given. The results show that the algorithms can better solve the contradict problem of resources, and the genetic operation is simple, each chromosome corresponds to a feasible scheduling, while avoiding the formation of the illegal chromosome.