Design Issues and Performance Analysis of Automatic Resource Allocation Scheme on Virtualization Environment

Abstract

Recently, virtualization is designed to isolate the system resource such like computing power, memory space, and application. Also, virtualization technique can make the resource usage more efficient and dynamically allocate for utilization. Hence, the virtualization can make virtual machine (VM) un-interrupt to be executed via direct migration between Physical servers and also make hardware maintenance and load balancing much better. Virtual architecture is to implement virtual layer on the host operating system, therefore the server can efficiently perform multiple VMs and also dynamically share the system resource between these VMs. Based on the RedHat Virtual Machine Manager system, we propose and design new automatic resource allocation technique called ARAS (Automatic Resource Allocation Scheme). Hence, the primary features of ARAS model will cover: (1) to perform the automatic allocation of resource such as CPU, Memory, and disk on the VM; and (2) to indentify and analyze the resource availability on physical servers for migration task on the VM; (3) to perform live migration methods of VM for automatic balancing. This paper will validate the efficiency and effect of the ARAS, and also perform simulations for its performance analysis under using ARAS on the VM. The final simulation results indicate that increase the resource utilization and enhance the system throughput on the VM with ARAS.