Virtual Machine migrating based on Markow chain
Jian Zhang, Caixin Zhu, Shengbo Chen, Xianyang Liu, Hongguang Du · 2016
Cloud computing has become one of the most dominant features in the computing area nowadays. The hardware of the datacenter can be get virtualized and create multiple Virtual Machine (VM) instances on a single Physical Machine (PM) in the Cloud computing environment. It is necessary to migrate the VMs from over utilized hosts to ensure the quality of service. Meanwhile the VMs loading on underutilized hosts need to be integrated into a physical machine which usage is in a reasonable range and the system will deactivate that hosts in order to saving power consumption. The key points of VM migration include issues like selecting appropriate VMs for migration at proper time and placement of VMs to suitable hosts. For solving the energy saving and migration issues the Markov-VMMS method is introduced in the paper. In this work the whole migration process is based on Markov chain which is used to predict the value of resource requirement. Extensive simulations have been performed on the cloud computing simulation platform named CloudSim to evaluate the proposed approach. The simulation results show that our approach can better control the energy consumption of the datacenter and enable the PMs provide better service quality.