Research On Virtual Machine Migration Timing And Migration Scheduling Strategy In Cloud Environment

Cloud computing is a computing model officially proposed in 2006,which provides pay-as-you-go cloud services for users.With the development of cloud computing,a large number of data centers have been established around the world,which generate a lot of energy consumption while providing infrastructure services for cloud computing.In order to solve the problem of high energy consumption in data center,virtual machine migration technology is proposed by researchers.However,the additional resource utilization cost in the process of virtual machine migration will also generate energy cost.In addition,frequent virtual machine migration will violate the service level agreement,thus affecting the service quality.Therefore,in view of the frequent migration of virtual machines in the cloud data center,the dynamic migration timing and migration scheduling strategy of virtual machines are studied in this paper,and a new migration strategy of virtual machines is proposed on the basis of the existing research.The main research work of this paper is as follows:Firstly,in order to solve the migration timing problem of virtual machine,a virtual machine adaptive migration strategy based on improved exponential smoothing prediction is proposed.On the one hand,this strategy combines threshold-based and predictive-based methods to provide a dual guarantee for virtual machine migration timing,effectively avoiding frequent virtual machine migration due to virtual machine transient overload;On the other hand,based on the existing exponential smoothing prediction method,the adaptive prediction mechanism is introduced to adapt to the special environment in which the data center load is constantly changing dynamically,and the prediction accuracy is improved effectively.Experiments show that this method improves the migration efficiency of virtual machines and has advantages in reducing energy consumption.Secondly,aiming at the migration process of virtual machine,a virtual machine migration scheduling strategy based on improved firefly algorithm is proposed.First,the strategy simplifies the virtual machine migration process by simulating the virtual machine migration process as a glowing firefly attracting companions.Secondly,the simulation annealing algorithm is introduced to solve the problem that the search for the source host using the firefly algorithm may fall into the local optimal solution.Finally,the impact of bandwidth on energy consumption is considered.Experiments show that this method can effectively reduce the energy consumption of the data center while ensuring the service quality of the data center.Finally,CloudSim is used to verify the proposed strategy.The experimental results show that the first strategy not only improves the efficiency of virtual machine migration,but also achieves the goal of reducing energy consumption.The second strategy not only reduces the energy consumption of the data center,but also ensures the quality of service.