Simple and Effective Dynamic Provisioning for Power-Proportional Data Centers

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journal

47 Scopus Citations
View graph of relations



Original languageEnglish
Article number6269026
Pages (from-to)1161-1171
Journal / PublicationIEEE Transactions on Parallel and Distributed Systems
Issue number6
Online published14 Aug 2012
Publication statusPublished - Jun 2013
Externally publishedYes


Energy consumption represents a significant cost in data center operation. A large fraction of the energy, however, is used to power idle servers when the workload is low. Dynamic provisioning techniques aim at saving this portion of the energy, by turning off unnecessary servers. In this paper, we explore how much gain knowing future workload information can bring to dynamic provisioning. In particular, we develop online dynamic provisioning solutions with and without future workload information available. We first reveal an elegant structure of the offline dynamic provisioning problem, which allows us to characterize the optimal solution in a divide-and-conquer manner. We then exploit this insight to design two online algorithms with competitive ratios (2-α) and ((e-1+α )), respectively, where (0≤ α ≤ 1) is the normalized size of a look-ahead window in which future workload information is available. A fundamental observation is that future workload information beyond the full-size look-ahead window (corresponding to (α =1)) will not improve dynamic provisioning performance. Our algorithms are decentralized and easy to implement. We demonstrate their effectiveness in simulations using real-world traces. © 1990-2012 IEEE.

Research Area(s)

  • Cloud computing, data center, dynamic provisioning, energy efficiency, online algorithms