Journal article
Dynamic right-sizing for power-proportional data centers
M Lin, A Wierman, LLH Andrew, E Thereska
IEEE ACM Transactions on Networking | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | Published : 2013
Abstract
Power consumption imposes a significant cost for data centers implementing cloud services, yet much of that power is used to maintain excess service capacity during periods of low load. This paper investigates how much can be saved by dynamically 'right-sizing' the data center by turning off servers during such periods and how to achieve that saving via an online algorithm. We propose a very general model and prove that the optimal offline algorithm for dynamic right-sizing has a simple structure when viewed in reverse time, and this structure is exploited to develop a new 'lazy' online algorithm, which is proven to be 3-competitive. We validate the algorithm using traces from two real data-..
View full abstractGrants
Awarded by NSF
Awarded by ARC
Funding Acknowledgements
Manuscript received February 20, 2012; revised July 27, 2012; accepted October 04, 2012; approved by IEEE/ACM TRANSACTIONS ON NETWORKING Editor Y. Bejerano. Date of publication November 22, 2012; date of current version October 11, 2013. This work was supported by the NSF under Grants CCF 0830511 and CNS 0846025, Microsoft Research, the Lee Center for Advanced Networking, and the ARC under Grant FT0991594.