Research On Cloud Computing Task Scheduling Algorithm Based On Differential Immunity

With the rapid development of computer and Internet technology,cloud computing as a commercial computing model has been concerned by many Internet enterprises.The quantitative development of cloud computing scale makes the 1 utilization of cloud computing virtual resources at low cost and the efficient processing of large-scale cloud tasks become one of the key issues in cloud computing technology research.The performance of cloud computing task scheduling algorithm directly affeets the efficiency and cost of cloud computing virtual resource processing tasks.At present,there arc still many deticiencies in the research on cloud computing task scheduling algorithm,which results in the waste of virtual resources in cloud computing and reduces the performance of cloud computing platfonn.Therefore,it is necessary to rnrther study reasonable and efficient cloud computing task scheduling algorithm.The performance of cloud computing task scheduling algorithm directly affects the efficiency and cost of cloud computing virtual resource processing tasks.At present,there are still many deticiencies in the research on cloud computing task scheduling algorithm,which results in the waste of virtual resources in cloud computing and reduces the performance of cloud computing platfonn.Therefore,it is necessary to further study reasonable and efficient cloud computing task scheduling algorithm.In order to allocate a large number of tasks submitted by users to cloud collmputing resources elliciently and reasonably,a differential immune-based cloud computing task scheduling algorithm,TMIDE is proposed,which considers both task completion time and cost.TMIDE algorithm maps the general continuous solution proble1 to the discrete cloud computing task scheduling problem through reasonable encoding and decoding schenies.The adaptive mutation factor F is designed in the variation process of the differential evolution algorithm,which accelerates the convergence speed of the algorithm at the early iteration stage and avoids the algorithm falling into the local optimal solution at the late iteration stage.Since the number of cloud tasks is very large,in order to increase the accuracy ot the algorithm,the vaccination mechanism of the immune algorithm is introduced into the traditional differential evolution algorithm.In the iterative process of the algorithm,individuals in the population are vaccinated according to the vaccination probability,so as to increase the number of optimal solutions of the population and improve the algorithm's accuracy.In order to avoid the destruction of the excellent individuals of the population by crossover operation after mutation and the waste of computing resources caused by unnecessary computing costs,this paper introduces the crossover judgment mechanism and effectively improves the operation efficiency of the algorithm.In the process of solving the problem,the fitness value function with dual objectives of time and cost is designed.In order to balance the order of time and cost,the adjustment formula is designed so that the optimal solution can consider both task completion time and task completion cost.To verify the effectiveness of the TMIDE algorithm,simulation tests were carried out on CloudSim cloud computing platform.The performance verification scheme was designed.The TMIDE algorithm was compared with the traditional differential evolution algorithm,genetic algorithm and Min-Min algorithm,and the experimental results were statistically analyzed.Experiments show that TMIDE algorithm can get better task scheduling sequence in a short time and achieve the balance between task completion time and calculation cost.