Research On Multi-objective Cloud Computing Task Scheduling Based On Improved Beetle Swarm Algorithm

With the rapid development of social informatization,new forms of the In ternet such as cloud mobile Internet and Internet of Things will rely on cloud computing technology.In cloud computing,cloud task scheduling is the most i mportant thing to make full use of cloud platform resources and provide more efficient and reasonable services for cloud users.An excellent scheduling strate gy can not only bring good experience to cloud users,but also meet the intere sts of cloud service providers.This paper focused on multi-objective optimizati on of scheduling time and usage cost in the process of cloud computing task s cheduling after a series of improvements to the beetle swarm intelligence algori thm.The main research content is as follows:(1)Firstly,the background and research status of task scheduling in cloud environment were introduced,and then the cloud computing technology and the concept of cloud task scheduling,architecture were explained,the principle of cloud task scheduling,optimization parameters and classical scheduling algorithm were analyzed,and the cloud task scheduling simulation software was introduced.(2)In this paper,we put forward a multi-objective optimal task scheduling algorithm with the optimization objectives of completion time and usage cost in the process of cloud task scheduling based on the improved beetle swarm algorithm.On the basis of the beetle swarm algorithm,the search accuracy of the algorithm is enhanced by adaptively modifying the speed weight factor and step size in the algorithm;the ability of the algorithm to jump out of the local optimal dilemma is improved by adding the odd-player exploration tactics;and the search efficiency of the beetle swarm is improved by using adaptation of the group and individual learning factors.The empirical results show that when the quantity of cloud tasks and the number of virtual machines gradually increases,the improved beetle swarm algorithm outperforms the comparison algorithm in terms of task completion time and scheduling utilization cost,and has a more significant optimization effect.(3)Aiming at the characteristic that the real-time computing ability of virtual machine in cloud task scheduling will constantly change with the increase or decrease of cloud tasks it receives,this paper takes the completion time and cost of all cloud users as the optimization goal,and proposes the beetle fireworks fusion algorithm to optimize the cloud task scheduling process with Nash equilibrium theory.The explosion factor and explosion radius in the fireworks algorithm were introduced into the beetle swarm algorithm,and the search ability of the fusion algorithm was improved by reasonable selection strategy.In addition,the Nash equilibrium principle is introduced in the scheduling process of cloud tasks,and each cloud user is regarded as a player,and the dynamic change of virtual machine real-time computing power is used to feedback the results,so that the Beetle fireworks algorithm provides reasonable scheduling strategies for each cloud user to play the game.The experimental results show that with the increasing number of virtual machines and heterogeneous values,the improved algorithm can effectively reduce the load balancing of the system while taking into account the completion time and cost in the process of cloud task scheduling.