云计算中主机和任务的数量都是十分庞大的,如何通过任务分配调度来减少成本开销和降低能耗是当前云计算和绿色计算领域研究的热点问题。根据云计算任务以及运行环境的特点,将云计算任务分配问题抽象为多维多背包求解问题,并采用改进的混合遗传算法对该问题进行求解。实验结果表明,改进的混合遗传算法能够在较短的时间内找到问题的优化解,并且根据该算法实现的任务分配策略能够有效地减少任务执行的成本开销和能耗。
The number of hosts and tasks in the cloud computing is very large, how to reduce costs and energy consumption through task scheduling is the hot issue of the current cloud computing and green computing research. According to the characteristics of cloud computing tasks and operating environment, the paper abstracts cloud computing task assignment problem for the multidimensional multiple knapsack problem solving, and adopts improved hybrid genetic algorithm to solve the problem. Experimental results show that the improved hybrid genetic algorithm can find the optimal solution for the problem in a short period of time, and task allocation strategy based on this algorithm can effectively reduce costs of task ex- ecution and energy consumption.