Authors : K. Aruna Kumari, K. Raja Sekhara Rao and J.K.R. Sastry
Abstract: Virtualization is a concept implemented as a major part of cloud computing infrastructure to offer everything as a service. Hypervisors are supported within the cloud computing platform using which many of the virtual machines can be created as per the demand of the end users. The load on the Virtual Machines (VM) greatly varies while some are over used, some may be under used or average used. Migration of a process, data and sometimes both needs to be done from one VM to other in such a way that the load on the servers is well balanced leading to high performance computing. Load balancing helps in a big way especially in controlling server sprawling and efficient use of the computing resources. Live migration has to be undertaken seamlessly without the initiation of any process from the client application. During migration, movement of data from one server to another has to be undertaken which generally consume lot of time. Data stored on a server by a user is historical and live and enormous amount of duplicated data is stored on the server especially for maintaining different versions of the same data. Elimination of duplicated data before migration of the same from one server to the other will greatly enhance the speed of migration. This study outlines some of the strategies for implementing data de-duplication within migration processes.
K. Aruna Kumari, K. Raja Sekhara Rao and J.K.R. Sastry, 2017. On Speeding up Virtual Machine Migration Through Integrated Data De-Duplication Methods. Research Journal of Applied Sciences, 12: 131-138.