Energy Efficient Data Center for Cloud Computing

Wenjie Zhuang

Energy Efficient Data Center for Cloud Computing

2015

New York Institute of Technology

Data processing service centers--Energy conservation
Data centers--Energy conservation
Cloud computing

English

New York Institute of Technology 2015. All rights reserved.

PDF

Text

School of Engineering and Computing Sciences

Master of Science in Computer Science

Ziqian (Cecilia) Dong

Cloud computing provides users with on-demand access to a pool of configurable computing resources that can be dynamically allocated and released. Data center is the backend infrastructure that made cloud computing possible. Today’s datacenters are providing a huge amount of computing power while consuming a considerable amount of energy, which are used in powering cooling system, switches and servers. Virtualization technology utilize one physical server to run mutiple virtualized operating systems, which is essential for dynamic computing resource allocation in the cloud computing paradigm.
This thesis starts with a thorough analysis of energy consumption in data centers and the problems raised by it. Studies in this thesis focus on reducing energy consumption in data centers, especially energy consumed on servers. An online virtual machine (VM) placement heuristic is proposed to reduce dynamic energy consumption. The technique of delay scheduling is then discussed and proposed as a way to improve energy efficiency. Experimental results are presented to show the impact of delay scheduling on energy efficiency.