Task Assignment Algorithms In Data Shared Mobile Edge Computing Systems

The appearance of the Mobile Edge Computing (MEC) successfully solves the bottlenecks of traditional Cloud based networks as computation ability of each mobile edge is sufficiently utilized. Since the mobile edges, including mobile devices and base stations, have certain data processing abilities, it is not necessary to offload all the computation tasks to the remote could for handling. Therefore, it is quite important to decide the optimal task assignment in a MEC system, and a series of algorithms have been proposed to solve it. However, the existing algorithms ignored the data distribution during task assignment, so that the applied ranges of these algorithms are quite limit. Considering the data sharing is very important and common in a MEC system, this paper studies the task assignment algorithm in Data Shared Mobile Edge Computing Systems, and three algorithms are proposed to deal with holistic tasks and divisible tasks, respectively. The theoretical analysis on the hardness of the problem, the correctness, complexities and ratio bounds of the algorithms are also provided. Finally, the extensive experiment results were carried out. Both of the theoretical analysis and experiment results show that our algorithms have high performance in terms of latency and energy consumption.