76. MANAGING DISTRIBUTED UPS ENERGY FOR EFFECTIVE POWER CAPPING IN DATA CENTERS

Department: Computer Science & Engineering
Research Institute Affiliation: Center for Networked Systems (CNS)
Faculty Advisor(s): Dean M. Tullsen | Tajana Simunic-Rosing

Primary Student
Name: Vasileios Kontorinis
Email: vkontori@ucsd.edu
Phone: 858-534-6181
Grad Year: 2012

Student Collaborators
Baris Aksanli, baksanli@cs.ucsd.edu

Abstract
Power over-subscription is a well-known technique to reduce both one-time capital costs and operating costs for modern data centers. However, designing the power infrastructure for a lower power point than the aggregated peak power of all servers requires dynamic techniques to avoid tripping breakers under worst-case power scenarios. This work considers the use of distributed per-server batteries to store energy during low activity periods and use this energy during power spikes. Researchers leverage the distributed nature of the batteries and design algorithms to prolong the duration of their usage. They find that they can reduce data center peak power by 19% and support 23.5% more servers with the same power infrastructure. Their approach reduces data center total cost of ownership per server by 6.2% with zero impact on performance.

« Back to Posters or Search Results