|
As an emerging computing paradigm, Mobile Edge Computing (MEC) can dramatically reduce latency and system energy consumption, which provides better quality of service for mobile users.One of the main challenges of MEC is determining a feasible task offloading decision of a mobile application to minimize the makespan or energy consumption. However, previous works on task offloading often ignore the interdependency between tasks, which is common in applications.In this paper, we focus on the problem of task offloading for a mobile application with several dependent tasks in an edge-cloud collaboration architecture. We aim to find a decision to minimize the weighted sum of makespan and total energy consumption. Since the problem is NP-hard, we propose a multi-objective optimization based heuristic algorithm to tackle this problem. Specifically, the algorithm first obtains a sub-optimal solution by eliminating some constraints and relaxing the original problem. Then, it allocates a priority for each task under dependency constraints. Finally, in each iteration, it generates several intermediate solutions and selects the first few solutions greedily.Extensive simulation results using both real applications and randomly generated applications show that our algorithm substantially outperform other alternatives. |
|
Keywords:Computer Software and Theory ; Computer Network; Mobile Edge Computing; Task Offloading; Optimization Theory |
|