Virtual Machine Deployment Strategy Based On Improved Particle Swarm Algorithm

With the rapid development of information and communication technology,the computing mode has experienced a mode of centralizing tasks to large processors from the beginning.The computing model developed into a distributed task processing model based on the network,and then to a cloud computing model based on on-demand processing.Through cloud computing,users can quickly apply for or release resources according to their business load,and pay for the resources used in an on-demand manner,improving service quality while reducing operation and maintenance costs.However,with the development of cloud computing technology,user demand has increased,and the scale of data centers has continued to expand.Data center operation costs are high and consume a lot of power during operation.Energy consumption has become an important cost in cloud systems.Therefore,energy conservation has become one of the main problems faced by cloud systems.The average CPU utilization rate of cloud computing data center servers is low.How to allocate and utilize the resources of the data center reasonably and reduce energy consumption is crucial.The data center uses virtualization technology to construct computing resources,storage resources,and network resources into a dynamic virtual resource pool,and uses virtual resource management technology to realize automatic deployment,dynamic expansion,and on-demand allocation of cloud computing resources.Users use on-demand and pay-as-you-go Use the method to obtain resources.Therefore,virtual resource management has become a hot and difficult point in current cloud computing research.Virtual machine deployment is a combinatorial optimization problem,and it is necessary to find a set of approximate optimal solutions.Particle swarm algorithm has been widely used in the approximate solution of optimization problems.It can effectively deal with combinatorial optimization problems.But particle swarm algorithm has the problem of easily falling into local optimal and single goal when solving virtual machine deployment.Aiming at the above problems,this paper first designs an improved single-objective particle swarm optimization algorithm.The algorithm first uses chaotic mapping,linearly decreasing inertia weight,and random grouping strategy to make the group have a higher search ability in the early stage,improve the ergodicity and diversity of the population,and improve the convergence accuracy and optimization ability in the later stage.Finally,the effectiveness of the algorithm is verified based on six benchmark functions,and the defect that the particle swarm algorithm is easy to fall into the local optimum is improved.Based on the improved single-objective particle swarm algorithm,a multi-objective particle swarm algorithm is designed for virtual machine deployment.Firstly,a data center resource model is established,and the resource utilization of the physical machine and the number of virtual machine migrations are used as the objective function to monitor the load situation of the three resources of the physical machine CPU,memory and bandwidth.Through multi-target selection strategy and virtual machine coding,the group is dispersed in the Pareto frontier of resource utilization and virtual machine migration,which improves the effectiveness of coding.Finally,using the Cloudsim simulation experiment tool,the experimental results show that the algorithm is highly efficient in improving physical machine resource utilization and reducing resource waste,achieving the purpose of energy saving and optimizing the performance of cloud computing data centers.