Optimizing Energy Consumption of Cloud Computing IaaS

Abstract

Cloud computing infrastructures are designed to support the accessibility and availability of various services to consumers over the Internet. Datacenters hosting Cloud applications consume massive amounts of power, contributing to high carbon footprints to the environment. Hence, Green Cloud computing solutions are needed within the Cloud datacenters that optimize the energy consumption. The main objective of this thesis is to address the problem of power and carbon efficient resource management in a Cloud datacenter. This work focuses on the development of a dynamic task scheduling algorithm to enhance the datacenter power efficiency over time. To achieve this objective, a formal optimization model is proposed using Integer Linear Programming (ILP) that minimizes the energy consumption in a Cloud datacenter. The model is verified using exact techniques and the Genetics Algorithm (GA) heuristic-based technique. Furthermore, an adaptive GA is proposed to reflect the dynamic nature of the Cloud computing environment. The proposed adaptive GA is validated by simulating an IaaS Cloud environment and conducting a set of performance and quality evaluation study in this environment. The results demonstrate that the proposed solution offers performance gains with regards to response time and in reducing the power consumption in the Cloud datacenter.