Title: Energy-efficiency enhanced virtual machine deployment policy for data-intensive applications in cloud platforms
Authors: Xiao Peng; Chen Runtong
Addresses: College of Computer Science and Technology, Hunan Institute of Engineering, Xiangtan City 411104, China ' College of Computer Science and Technology, Hunan Institute of Engineering, Xiangtan City 411104, China
Abstract: In cloud platforms, data-intensive workflows are widely applied for solving non-trivial applications. However, extra performance and energy consumption costs have to be spent because of using virtualisation technology. In this paper, we present a novel virtual machine deployment policy, which is aiming at improving the energy-efficiency of executing data-intensive workflows in virtualised datacentres. The proposed deployment policy consists of two phases: firstly, it uses a novel heuristic for deploying virtual machines; secondly, it schedules workflow activities to an energy-aware priority. In this way, both the execution performance and energy-efficiency are fully taken into consideration in the proposed algorithm. Extensive experiments are conducted by using a real-world workflow application as workload, and the results show that the proposed policy can significantly reduce the energy consumption of intermediate data transferring. In addition, it exhibits better robustness than existing approaches when cloud systems are in presence of I/O-intensive workloads.
Keywords: cloud computing; resource virtualisation; green data centres; data-intensive workflow; energy efficiency; virtual machine deployment; cloud platforms; energy consumption; deployment policy; virtual machines; workflow scheduling; energy-aware priority.
International Journal of Internet Protocol Technology, 2014 Vol.8 No.4, pp.181 - 189
Available online: 23 Mar 2015 *Full-text access for editors Access for subscribers Purchase this article Comment on this article