Authors: Hong Min; Jinman Jung; Junyoung Heo
Addresses: School of Computer and Information Engineering, Hoseo University, Cheonan, South Korea ' Department of Information and Communication Engineering, Hannam University, Daejeon, South Korea ' Department of Computer Engineering, Hansung University, Seoul, South Korea
Abstract: Mobile cloud computing is the combination of cloud computing and mobile computing, and provides rich computational resources to a mobile computer. In mobile cloud computing, computation offloading techniques are used to overcome the limitations of resource-constrained mobile devices. Offloading techniques perform some parts of a job of mobile devices in the cloud in behalf of mobile devices. If the cost of the operation in that part of the job in a mobile device is larger than the cost associated with offloading, the part is executed in the cloud. Traditional cost analysis models for deciding which parts of a job to execute in the cloud or a mobile device were estimated by using only the cost of offloading, which is composed of data transfer and response time required for the function call. In this paper, we propose a novel offloading cost analysis model based on the data synchronisation rate and the data exchange rate for the input of the function to improve the accuracy of offloading cost estimation. We confirm through experiments that the offloading technique with the proposed model can reduce the execution time of a job and consequentially improve the energy efficiency when compared to previous techniques.
Keywords: mobile cloud computing; profiling; offloading; remote procedure call; RPC.
International Journal of Information and Communication Technology, 2018 Vol.13 No.3, pp.316 - 328
Received: 02 Sep 2015
Accepted: 06 Dec 2015
Published online: 10 Apr 2018 *