Automated Live Migration of Virtual Machines in Cloud Data Center

Abstract

The thesis entitled “Automated Live Migration of Virtual Machines in Cloud Data Center” involves the study and implementation of load balancing algorithm for Virtual Machine Management in the cloud infrastructure. The study is focused on dynamic resource scaling and live migration algorithm of virtual machines to achieve the load balancing in the cloud infrastructure. As data centers grow sharply, they find themselves accommodating an increasing number of Physical Machines and Virtual Machines. This development requires an effective resource management in data centers. As a result, the load is evenly distributed and service level agreements (SLAs) are met. From this point of view, load balancing and live migration are becoming essential processes for data center management. In cloud, computing resources are provided as a service to its clients across the globe based on demand. Huge demand for cloud resources results in overutilization by degrading the performance of the servers whenever there is a heavy load. It is necessary to distribute the load across the servers in cloud by taking into consideration of allocating the right amount of resources dynamically based on the load to improve the performance of applications running inside virtual machines. Experiment is conducted on Xen servers in the real physical hardware and in the virtualized environment and implemented dynamic resource scaling and live migration technique to manage load in the virtual machines running on those Xen servers. The response time of virtual machine is used as a metric to measure the performance of algorithms. After the implementation of dynamic resource scaling and live migration technique, the improved performance of virtual machine in terms of response time and efficient utilization of Xen server resources is observed.