Research On Virtual Machine Placement And Migration In Cloud Computing

As a new type of computing model,cloud computing has the advantages of low price,high reliability,on-demand services,scalability,etc.,and has received widespread attention from the society.The cloud data center is the most important infrastructure of cloud computing.Using virtualization technology to integrate physical resources can efficiently manage data center resources.As the scale of cloud data centers continues to expand,the energy consumption of equipment and network resource consumption in cloud data centers are also increasing.Reasonable virtual machine placement strategies and virtual machine migration strategies can effectively reduce the energy consumption and bandwidth resource consumption of cloud data centers.Therefore,the virtual machine placement and virtual machine migration problems in the cloud environment have become hot research issues in the cloud computing field.The specific work is as follows:This paper proposes two algorithms to solve the virtual machine placement problem and the virtual machine migration problem,and uses the CloudSim cloud simulation platform to test the proposed algorithm.(1)For the virtual machine placement problem,this paper constructs a virtual machine placement problem model based on energy consumption and network bandwidth.In order to reduce the energy consumption and bandwidth consumption of cloud data centers,this paper proposes a CPSO algorithm to solve the problem model.This algorithm improves the local search ability of the algorithm by adding a convergence factor to the particle swarm optimization algorithm.This paper compares the proposed algorithm with some heuristic and meta-heuristic algorithms.The experimental results show that as for the virtual machine placement problem,the algorithm proposed in this paper can significantly reduce data center energy consumption and network bandwidth consumption.(2)For the problem of virtual machine migration,this paper constructs a virtual machine migration problem model based on communication cost.In order to reduce the communication cost and energy consumption of the cloud data center,this paper proposes a CEVM virtual machine migration algorithm considering communication cost,energy consumption and Qo S.This algorithm selects the virtual machine with the highest communication cost among the overloaded physical machines as the virtual machine to be migrated out,and selects the physical machine with the largest weighted sum of the communication cost and energy consumption reduction rate after the migration as the destination physical machine to be migrated in.This paper uses the CloudSim cloud simulation platform to compare the proposed migration algorithm with the other two migration algorithms.Experimental results show that the proposed virtual machine migration algorithm can effectively reduce the network traffic and energy consumption of the cloud data center when solving the problem of virtual machine migration.