A Lightweight Algorithm for Collaborative Task Execution in Mobile Cloud Computing

Mobile cloud computing (MCC) combines mobile internet and cloud computing to improve the performance of mobile applications. In MCC, the performance of mobile devices (MDs) can be significantly improved by offloading the mobile applications to the remote cloud. However, the data, which is transmitted on wireless networks, is increasing rapidly since users' mobile applications have to get support from the remote cloud, therefore, these applications offloading face the problem of energy efficiency because of stochastic wireless channel. In this paper, we investigate collaborative task execution between MD and cloud side for mobile applications. In our study we assume the mobile application is composed by a sequence of tasks that are independent of each other, and can be executed on the MD or on the cloud side. We aim to minimize the energy consumption on the MD while meeting a deadline, by offloading a part of tasks of mobile application to the cloud. We formulate this collaborative task execution problem as an energy optimization problem. Then, we derive several offloading thresholds by characterizing the optimal solution and propose several algorithms for the collaborative task execution. Further, using Lagrange duality principle and these algorithms, we propose a collaborative task execution scheduling (CTES) algorithm to solve the optimization problem approximately. Simulation results suggest that our proposed CTES algorithm is energy efficient for the MCC environment. Moreover, compared to the local execution and the cloud execution, our proposed CTES algorithm can significantly save the energy consumption on the MD.