A Live Migration Strategy for Virtual Machine Based on Performance Predicting

The live migration for virtual machine has been widely utilized. This paper proposes a live migration strategy for virtual machine, which combines with performance predicting algorithm. According to the average utilization of the CPU, memory, I/O and network bandwidth, the live migration strategy of this paper utilizes the performance predicting, and make a serial of judgements about migration, such as whether a migration is triggered, what virtual machine should be migrated, and where is the destination host of the virtual machine. In this paper, we describe the algorithm of realizing this strategy in details, and carry out some experiments which demonstrate that this strategy is able to maintain the system stability and achieve the load balancing.