Research Of Multi-Objective Task Scheduling Based On Chaos Cat Swarm Optimization In Cloud Computing

Cloud computing is an Internet-based computing method that developed on the basis of technologies such as distributed computing,parallel computing,and grid computing.In this way,shared hardware and software resources and information can be provided to computers and other devices on-demand.In order to optimize the service computing and meet the needs of users and service providers,it is necessary to improve the efficiency of task scheduling in cloud computing.There are some problems with current task scheduling strategies.Firstly,the optimization goal in the task scheduling strategy is too singular,and the impact of execution time and load balancing on task scheduling cannot be well taken care of.Secondly,the task scheduling strategy mostly use a single optimization algorithm or a combination of multiple algorithms,but all of them are essentially focused on local or global optimization.This paper proposes a multi-objective task scheduling model based on chaos cat swarm optimization,which takes execution time and system load balancing as optimization targets,Two search modes in the chaos cat swarm optimization can perform local optimization and global optimization at the same time.Chaotic mapping ensures a more balanced distribution of the two optimization modes and also increases the diversity of the population.To verify the optimization effect of the algorithm,Comparing the chaos cat swarm optimization and the optimization algorithm commonly used in cloud computing,namely genetic algorithm and particle swarm optimization algorithm on the Cloudsim platform.The experimental results show that the chaos cat swarm optimization is more stable and has better optimization effect on algorithm execution time and system load.To further improve the efficiency of the algorithm,using the MapReduce programming model in Hadoop to realize the parallelization improvement of the chaos cat swarm optimization.In the experiment,comparing the execution time of the algorithm under the single-machine condition and cluster mode.The result show that the improved algorithm execution time is lower when the large-scale task scheduling is performed.In the cluster mode,the algorithm execution time is compared by changing the number of nodes in the cluster,and the experimental result show that the execution time suddenly decreases with the increase of the number of nodes in the cluster.This proves that the chaos cat swarm optimization has good scheduling applicability in the cluster.