POWER-DELAY AWARE ALLOCATION OF VIRTUAL MACHINES IN CLOUD DATACENTERS

Abstract

Recently, enormous high demand of computational power created by modern compute-intensive business and scientific applications has driven the IT infrastructures to create computing infrastructures such as large-scale datacenters. However, these large scale datacenters consumes enormous amount of electrical power. Which, not only increases the operational cost but also reduce the system reliability and devices lifetime due to overheating. Another problem with this is emission of carbon dioxide which adds to the green house effect. Virtualization is a promising approach to reduce this power consumption by consolidating multiple virtual servers onto a smaller number of computing resources. But modem Cloud computing environments have to provide high Quality of Service for their customers resulting in the necessity to deal with power-performance trade-off. In this paper an efficient resource management policy for virtualized Cloud data centers is proposed. Power consumption by data-centers can be reduced by leveraging live migration of VMs and switch off idle nodes. Again dynamic reallocation incurs network propagation delay that can affect the operation cost. The objective is to optimize the power consumption as well as network delay.