Student research opportunities
Operation Cost Minimization in Large Scale Distributed Data Centers
Project Code: CECS_828
This project is available at the following levels:
CS single semester, Honours, Summer Scholar, Masters, PhD
Keywords:
power management, reusable energy, approximation algorithm, energy optimization, cloud computing
Supervisor:
Assoc Professor Weifa LiangOutline:
Electricity expenditure comprises a significant fraction of the total operating cost in data centers. Data centers now pay more for electricity bills than servers. A typical data center consumes as much energy as 25,000 households per year. The total electricity bill for data centers in 2010 was estimated over $11 billion and this cost is doubled every five years. Cloud service providers are required to reduce electricity cost as much as possible.
In this project we aim to minimize the operation cost of the cloud providers by developing efficient algorithms for allocating user requests to the data centers that are geographically located at different regions while meeting their negotiated service level agreements (SLA), by incorporating both the temporal and spatial variations of electricity prices under wholesale electricity markets,
for which we will propose different cost modeling, devise fast approximation algorithms, and conduct extensive experiments to evaluate the performance of the proposed algorithms, using the real electricity price data sets.
Goals of this project
The objective of this project is to develop an optimization framework to allocate requests from various web portals to geographically distributed data centers such that the operation cost of such cloud service provide is minimized
while the user SLA requirements are met.
Requirements/Prerequisites
Excellent programming skills, basic algorithm backgrounds, and distributed computing knowledge.
