改进GA的边缘计算任务卸载与资源分配策略

Computing unloading and resource allocation strategy based on improved GA in mobile edge computing environment

针对现有边缘计算计算卸载算法存在的延迟较大且负载不均衡的问题,提出一种移动边缘计算中基于改进遗传算法的计算卸载与资源分配算法。基于提出的移动边缘计算网络构建系统模型,其中包括能耗、平均服务延迟、执行时间以及负载均衡模型。以能耗、延迟、负载均衡最小化为优化目标,利用改进的遗传算法进行求解,其中采用染色体一维表现形式、交叉和变异算子提高算法的性能。利用iFogSim和Google集群对所提算法进行模拟仿真实验,结果表明,算法种群数量和最大迭代次数的合理值分别是60和25,所提算法得到的计算卸载和资源分配策略在能耗、负载均衡、延迟和网络使用率方面的表现均优于其它算法。

Aiming at the problem of large delay and unbalanced load existing in the existing edge computing offloading algorithm,a computing unloading and resource allocation algorithm based on improved genetic algorithm in mobile edge computing was proposed.A system model was built based on the proposed mobile edge computing network,including energy consumption,average service delay,execution time and load balancing model.To minimize energy consumption,delay and load balancing,an improved genetic algorithm was used to solve the problem,in which one-dimensional chromosome representation,crossover and mutation operators were used to improve the performance of the algorithm.The proposed algorithm was simulated by iFogSim and Google cluster.The results show that the reasonable values of the number of population and the maximum number of iterations are 60 and 25 respectively,and the proposed algorithm outperforms other algorithms in terms of energy consumption,load balancing,delay and network utilization.