An Effective Offloading Model Based on Genetic Markov Process for Cloud Mobile Applications

摘要

Mobile Cloud Computing (MCC) has drawn significant research attention as mobile devices' capability has been improved in recent years. MCC forms the platforms for a broad range of mobile cloud solutions. MCC's key idea is to use powerful back-end computing nodes to enhance the capabilities of small mobile devices and provide better user experiences. In this paper, we propose a novel idea for solving multisite computation offloading in dynamic mobile cloud environments that considers the environmental changes during applications' life cycles and relationships among components of an application. Our proposal, called Genetic Markov Mobile Cloud Computing (GM-MCC), adopts a Markov Decision Process (MDP) framework to determine the best offloading decision that assigns components of the application to the target site by consuming the minimum amount of mobile's energy through determining the cost metrics to identify overhead on each the component. Furthermore, the suggested model utilizes a genetic algorithm to tune the MDP parameters to achieve the highest benefit. Simulation results demonstrate that the proposed model considers the different capabilities of sites to allocate appropriate components. There is a lower energy cost for data transfer from the mobile to the cloud.