Evolutionary computation method for solving cloud task scheduling

The invention belongs to the field of intelligent manufacturing cloud computing, and particularly relates to an evolutionary computation method for solving cloud task scheduling. According to the method, firstly, a mathematical model of cloud task scheduling is constructed, and then an improved genetic algorithm is designed to solve the mathematical model. According to the method, the crossover operator is redesigned, and the thought of the local greedy algorithm is combined, so that three indexes corresponding to each task set in the filial generation are balanced as much as possible while aneffective solution is ensured. The fitness value is described as an evaluation function after normalization processing of the reciprocal of the sum of variance of CPU occupied resources, memory occupied resources and bandwidth occupied resources of S virtual machines and the reciprocal of the total execution time. The algorithm is helpful for load balancing of each virtual machine and balancing of CPU utilization rate, m