CPU Usage Prediction Technique for Live Migration of Virtual Machines

Abstract:Virtualization is a very important technology of cloud computing which partition the physical host into several Virtual Machines (VMs). The number of active host and their power consumption is easily managed by migrate the virtual machines based on their resource requirement and current status of the particular host. In cloud environment, Service Level Agreement (SLA) is necessary for maintaining reliable Quality of Service (QoS) between the user and data centers. The main objective is reduction of power cost and SLA violation. Live migration of the virtual machines manages the over load and under loaded situation of host which gives the ability of dynamic resource allocation on different host. The proposed technique will provide the ability of dynamic virtual machine consolidation using adaptive utilization threshold value based on CPU usage prediction which can easily manage the better SLA and reduces the number of VM migrations in between the host. The validation of the proposed technique on multiple workload traces of the Planet lab servers. Index Terms – Virtualization, Resource, CPU, Prediction, Live migration, Consolidation, Virtual Machine.

Computer Science,Engineering