Investigation on energy consumption problem in data centre systems

• Main Author: Xiong, Xiao.
• Other Authors: Zhong Wende
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/54394
• Institution: Nanyang Technological University

Data centres which are powered by green energy, such as wind and solar would benefit the environment significantly. More and more IT companies have started to incorporate green energy to power their data centres. For example, Google are currently using renewable energy to power over 30% of their operations and they are looking to replace all the brown energy with green energy in the future. Scheduling the workloads across the distributed data centres by considering the presence of the green energy can reduce the brown power consumption in the data centres significantly. In this project, we simulate distributed data centre environment where job arrivals follow the Poisson process and job service times follow an exponential distribution. Our scheduling algorithm is green-aware and leverages on the task migration across data centres in different locations. Our simulations demonstrated that our scheduling algorithm have saved up to 30% brown energy compared with the conventional round robin algorithm which is one of the most widely used scheduling algorithm.