Achieving better Cloud access experience with server virtualization

Issue title: Special Supplement Issue in Section A and B: Selected Papers from the ISCA International Conference on Software Engineering and Data Engineering, 2011

Guest editors: Narayan C. Debnath

Article type: Research Article

Authors: Hao, Wei^{a; *} | Fu, Jicheng^b | Trenkamp, Chris^a | Prapatanant, Pochara^a

Affiliations: [a] Department of Computer Science, Northern Kentucky University, Highland Heights, KY, USA | [b] Department of Computer Science, University of Central Oklahoma, Edmond, OK, USA | Department of Computer Science, Winona State University, Winona, MN, USA

Correspondence: [*] Corresponding author: Wei Hao, Department of Computer Science, Northern Kentucky University, Highland Heights, KY 41099, USA. Tel.: +1 859 572 5659; E-mail: [email protected].

Abstract: Cloud computing is changing the way we work and the way we live. Cloud providers host their Clouds at their data centers. However, the data centers might be far away from end users. The end users would experience long response latency. Client-side proxy servers are close to the end users. They are widely deployed around the Internet. They are the ideal platforms to cache/replicate Cloud applications. However, current proxy servers do not support caching and execution of Cloud applications. In this paper, we utilize virtualization technology to expand the functionalities of proxy servers to support caching and execution of Cloud applications. Specifically, we propose a system model that consists of a virtualization server, a Cloud access request interception mechanism, proxy server extension modules, and a migration protocol. Heterogeneous services hosted by different Cloud providers can be migrated from the Clouds to the virtualization server at the proxy server side. Also, a migration cost function is designed to assist service migration decision making. With our approach, Cloud applications can be moved close to the end users. Our approach can reduce the user perceived response latency and thus it can achieve better user experience during Cloud access. We have utilized the industrial standard e-commerce benchmark TPC-W to conduct experimental studies. The experimental results show that our approach significantly reduces client response latency.

Keywords: Cloud computing, software virtualization, proxy server, service migration

DOI: 10.3233/JCM-2012-0434

Journal: Journal of Computational Methods in Sciences and Engineering, vol. 12, no. s1, pp. S29-S37, 2012