State-Full Virtual Machine Live Data Migration For Improved Load Balancing

M. SHAIKH, A. SHAIKH, M. A. MEMON, A. A. SHAH, R. H. SHAH

Abstract


Cloud resembles with Virtualization to achieve the shared environment and shared resources. In virtualized shared environment, the availability of the shared resources for the shared multiple clusters/Virtual Machines is always matters. To provide the better performance to the end users of the cloud VMs and for the always availability of shared resources, virtual machine migration is always a challenging task in virtualized environment. The aim of this research is to investigate various parameters of live data migration between the shared clusters/virtual machines in state-full context and overall load balance level. The proposed migration strategy ensures the avoidance of attacks/anomalies which causes the load increase on the VMs which effects the entire system. This could be done by the resource allocation from one cluster to another via checkpointing method of VM to maintain the state of the VM, load of the VM including resource availability. Using Xen hypervisor memory technique which dynamically allows the migration of the configured memory while the allocated memory could be discarded for a while. By this the bad memory remains un-migrated only the good memory consisting the used data would be migrated by means of Real-time.


Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Copyright (c) 2018 Sindh University Research Journal - SURJ (Science Series)

 Copyright © University of Sindh, Jamshoro. 2017 All Rights Reserved.
Printing and Publication by: Sindh University Press.