The Research On Virtual Migration Migration Based On Software-Defined Networks In Cloud Computing

Cloud computing and Software-defined Networks are the key issues of current research.Using virtualization technology,Cloud computing can provide IT resources as services to users and these services can be requested at any time and be paid on-demand.The virtual machine(VM)is a carrier of the virtualization technology,which can make full use of the server's effective resources,effectively reduce operating costs and improve server operating efficiency.Among them,VM migration is one of the key technologies to realize the dynamic management in data center networks.However,the cloud data center needs to manage a great number of VMs followed by the expanding of the data center's scale.Therefore,how to choose the target server to the VM and how to reduce the impact on the whole data center by the VM migration have become the current focus in recent years.On the basis of in-depth study and summarize the existing achievements in related fields,in this thesis,surrounding VM placement problem and VM migration planning problems,we mainly carried out the following research work.Since data centers hosting Cloud applications consume huge amounts of energy,we need Green Cloud computing solutions that can not only optimize resource configuration but also reduce operational costs.Therefore,a heuristic algorithm based on best fit(HABF)is proposed in this thesis.Based on the relevant resource constraints,the HABF algorithm elects a placement scheme with the goal of minimizing the number of servers by considering the impact on the server's resource utilization after the VM has been placed in the server.Finally,the simulation results show that the algorithm can not only significantly reduce the number of servers used in the data center and the configuration time of VMs,but also decrease energy consumption effectively.When multiple VMs need to be migrated at the same time,the migration order and migration bandwidth has a great impact on migration performance.In order to effectively reduce the total time of VM migration and the impact on the service,a grouping-based heuristic algorithm(GBHA)is proposed in this thesis.The GBHA algorithm finds the optimal migration plan by considering the cumulative effect of a group of parallel migrations VMs on the migration time of other VMs and updating resource constraints such as network bandwidth in time.The simulation results show show that the algorithm can significantly reduce total migration time of VMs compared with existing algorithms and improve migration performance effectively.