Research On Virtual Machine Scheduling Methods In Cloud Computing Environment

In recent years,cloud computing has drawn wide attention from industry and academia,becoming a research hotspot and trend in the field of information technology.The implementation of virtualization technology makes the virtual machine(VM)scheduling largely affect the performance of cloud computing environment.At present,most of the researches on VM scheduling are based on individual VMs,ignoring the frequent communication between VMs.These scheduling results will not only generate a lot of communication costs,but also lead to an increase in data center energy consumption.In order to solve these problems,this paper studies from two aspects of virtual machine placement and migration respectively.The main contents are as follows:Aiming at the problem of high energy consumption of network equipment and service level agreement(SLA)violation caused by resource aggregation during VM placement,a VM placement algorithm based on graph partitioning is proposed to optimize energy consumption.For the VM group submitted by the user,we reconstruct the VM associated graph according to the traffic and load correlation between VMs,and partition the graph using the improved multilevel k-way partitioning algorithm.Combined with the data center topology,the two-layer mapping relationship of VMs and physical machines(PMs)is determined by extending PM clusters.The experimental results show that our proposed algorithm can guarantee better resource utilization,control SLA violation and offer a significant savings of energy compared with other related algorithms.Aiming at the problem of high communication costs caused by neglecting the association between VMs during VM migration,a multi-target VM migration algorithm based on group selection is proposed.The appropriate VM groups are selected as migration options,and a comprehensive optimization method that combines the migration cost,the communication cost and the VM heat is designed to get the optimal migration plan.Extensive experiments show that our algorithm can effectively reduce the migration cost and communication cost,improve the system reliability compared with other related algorithms.Based on the above mentioned placement and migration algorithms,a VM scheduling system is designed and implemented.The system can effectively place the VM group,monitor resource utilization status of PMs in real time,and migrate VMs for overloaded PMs.The test verifies that VMs can be efficiently placed and migrated in a cloud computing environment and the system performance of cloud computing can be improved.