云计算平台网络公共资源应急调度仿真研究

Emergency Dispatch of Network Public Resources on Cloud Computing Platform

对云计算平台网络公共资源的应急调度,能够有效提升云计算平台网络应急救援能力。对网络公共资源的调度,需要通过海明距离约束产生初始种群,改进目标函数为适应度函数,完成网络公共资源的应急调度。传统方法从公共资源公平性、优先级视角映射为图的构建问题,采用增量式优化算法实现公共资源应急调度,忽略了目标函数的改进过程,导致调度效果不理想。提出基于遗传算法的云计算平台网络公共资源应急调度方法。以应急公共资源调度的时间成本和运输成本最小化为目标,通过海明距离约束产生初始种群,并将应急调度模型相应的目标函数改进为适应度函数,对遗传算法个体选取方式以及交叉和变异概率进行优化,完成公共资源应急调度。实验证明,所提方法能够有效提升应急资源调度效率,且应急调度可靠性较优。

This paper puts forward a method for emergency scheduling of network common resource on cloud com- puting platform based on genetic algorithm. The initial population is generated via Hamming distance constrain targe- ting minimizing time cost and transportation cost of the emergency scheduling, then objective function corresponding to emergency scheduling model is modified to fitness function. The individual selection mode of genetic algorithm and crossover and mutation probabilities is optimized. Thus, the emergency scheduling is completed. Simulation results show that the method can improve scheduling efficiency of emergency resource and has superior reliability of emergen- cy scheduling.