Research On Swarm Intelligence Optimization Algorithm For Cloud Computing Task Scheduling

In recent years,the whole society has been undergoing digital transformation,which has brought about a large amount of data calculation and processing with industrial upgrading.The rapid development of cloud computing effectively solves the increasing demand for computing in large-scale application programs.There are abundant computing resources in cloud computing environments,and reasonable allocation and scheduling are conducive to improving resource utilization and reducing economic costs.How to allocate computing resources reasonably has become a hot topic in current research.Cloud computing task scheduling is an NP-hard problem.With the deepening of research,it has been shown that swarm intelligence algorithms in heuristic algorithms can often achieve better scheduling effects in this field.However,existing swarm intelligence optimization algorithms have problems such as slow convergence speed and easy fall into local optimal solutions.In view of the above problems,this paper proposes two improved swarm intelligence optimization algorithms to solve the problem of cloud computing task scheduling:(1)The Improved Whale Optimization Algorithm combining Levy flight and evolutionary population dynamics(WOALE).This paper makes 5 improvements to the standard Whale Optimization Algorithm to optimize its convergence accuracy and speed.First,the segmented Logistic chaotic mapping strategy is used for population initialization.Second,the nonlinearly decreasing adaptive inertia weight strategy is used to balance the global search ability and local search ability of the algorithm in the early and late stages.Third,the dynamic convergence coefficient strategy is used to control the iteration step size.Fourth,the Levy flight mechanism is incorporated to extract a population individual for Levy flight during each iteration process of the algorithm,which enhances the perturbation of the algorithm.Fifth,the idea of evolutionary population dynamics(EPD)is introduced to make whale individuals have mutation ability.Experimental results show that the improved algorithm improves both convergence speed and accuracy compared with the original algorithm.(2)The Improved Grey Wolf Optimization Algorithm combining Bald Eagle Optimization Algorithm(IGWO-BES).This paper makes 3 improvements to the standard Grey Wolf Optimization Algorithm to optimize its convergence accuracy and speed.First,the reverse learning strategy is used to initialize the population.Second,the nonlinear control parameter strategy is used to dynamically adjust the algorithm iteration step size.Third,the Bald Eagle Optimization Algorithm is incorporated.The algorithm after fusion has the characteristics of fast convergence of Bald Eagle Optimization Algorithm and also retains the characteristics of the original Grey Wolf Optimization Algorithm.Experimental results show that the improved algorithm improves both convergence speed and accuracy compared with the original algorithm.This paper selects 7 high-dimensional single-peak functions and 5 high-dimensional multi-peak functions to test the performance of the proposed algorithms,and the test results prove the superiority of the algorithms.In addition,this paper applies the algorithms to both the single-objective optimization model and the multi-objective optimization model of cloud computing task scheduling.And experimental simulations were conducted on the Matlab platform.The results show that the two algorithms proposed in this paper have better convergence speed and accuracy compared to the original algorithm,and perform well in solving the cloud computing task scheduling problem.