Frequency-Adaptive VDC Embedding to Minimize Energy Consumption of Data Centers

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

1 Scopus Citations
View graph of relations

Author(s)

  • Zhiyuan Wang
  • Chao Guo
  • Sanjay K. Bose
  • Gangxiang Shen

Related Research Unit(s)

Detail(s)

Original languageEnglish
Pages (from-to)447-461
Number of pages14
Journal / PublicationIEEE Transactions on Green Communications and Networking
Volume6
Issue number1
Online published19 Aug 2021
Publication statusPublished - Mar 2022

Abstract

The increasing popularity of cloud computing would require Data Centers (DCs) to be scaled up rapidly, as needed, to provide adequate computing and storage infrastructure with rapidly growing demands. Since energy costs would be crucially important for these DCs, various approaches have been proposed to operate them efficiently. Virtual data centers (VDCs) are a promising approach for this as they can efficiently provide computing and storage resources to users over a shared physical infrastructure. In the context of VDC services, this paper focuses on improving the energy efficiency of DCs by applying a Dynamic Frequency Scaling (DFS) mechanism to provision VDC services. Specifically, the frequencies applied in each hardware are adaptively adjusted as per the given service requirements. This is done to minimize the overall energy consumption in the DC hardware when embedding a specific VDC. To the best of our knowledge, this is the first work that incorporates this DFS mechanism in the VDC provisioning problem. To minimize the overall energy consumption, we develop both an integer linear programming (ILP) optimization model and efficient heuristic algorithms. Extensive simulations are conducted to show that incorporating the DFS mechanism in VDC service provisioning significantly improves the energy efficiency of a DC when compared with a scheme where this mechanism is not applied. The proposed heuristic algorithms are also efficient and perform almost as well as the optimum ILP model.

Research Area(s)

  • Cloud computing, Computational modeling, Data center, Data centers, DFS, Energy consumption, Energy efficiency, Heuristic algorithms, Servers, VDC embedding