Research On Resource Management Of Cloud Computing Based On Swarm Intelligence Algorithm

Cloud computing is an emerging technology developed on the basis of distributed computing,network computing,parallel computing and other technologies.Cloud computing platform can manage a large number of computing resources scattered in different places and different devices through the network to form a virtual resource pool,and cloud computing users can obtain various resources according to their own needs through the cloud computing platform.At present,the overall scale of cloud computing continues to expand,resulting in the decline of the overall quality of cloud computing service,resource utilization continues to decrease,so the effective management of cloud computing resources can not only ensure the quality of service of cloud computing,but also improve the resource utilization of cloud computing platform,reduce energy consumption and save costs.As the mainstream cloud computing resource management algorithm,the traditional swarm intelligence algorithm still has many problems in resource management.For example,in the process of task scheduling,the task completion time is long,which leads to the low quality of cloud computing service.And the problem of low resource utilization in the process of virtual machine placement.Based on this background,this paper improves the existing swarm intelligence algorithm,and based on this algorithm,it makes an in-depth study on cloud computing resource management from two aspects of task scheduling and virtual machine placement,as follows:(1)Improvement of swarm intelligence fusion algorithmIn this paper,the main defects of particle swarm optimization algorithm in cloud computing resource management are analyzed,and a swarm intelligence fusion algorithm is proposed.Aiming at the problem that the quality of the initial population will affect the overall iteration effect of the particle swarm optimization algorithm,the opposition-based learning method is used to improve the quality of the initial population.Particle swarm optimization(PSO)is easy to fall into local optimum problem in the process of optimization.Three kinds of moving behaviors in artificial fish swarm algorithm are introduced to make the particles avoid falling into local optimal solution in the process of moving,and the particles trapped in local optimal solution have the ability to jump out of local optimal solution.At the same time,the inertia weight and learning factor in the process of particle movement were adjusted nonlinearly,which could not only accelerate the convergence speed of particles,but also balance the optimization ability of particles in different stages and improve the optimization quality of the algorithm.(2)Resource management based on swarm intelligence fusion algorithmIn order to improve the efficiency of cloud computing resource management,this paper applies the proposed swarm intelligence fusion algorithm to the two stages of task scheduling and virtual machine placement in cloud computing resource management,establishes a mapping relationship between the swarm intelligence fusion algorithm and the two stages,and optimizes for different objectives to establish the corresponding fitness function.In the task scheduling phase,in order to improve the quality of service of cloud computing platform,the task completion time was selected as the optimization goal.In the virtual machine placement stage,in order to reduce the energy consumption and cost of cloud computing platform,the utilization of various resources in the physical machine is optimized as the main goal.