Multiobjective African Vulture Scheduling Algorithm in Green Mobile Edge Computing

With the proliferation of versatile mobile applications, offloading compute-intensive tasks to the mobile edge cloud (MEC) becomes a dramatic technique. With the introduction of energy harvesting (EH) technology, hybrid energy supply has been considered an effective way to save on-grid energy (brown energy). However, most current work on their offloading schemes does not consider both the task dependencies and the hybrid energy supply. In this paper, we proposed a novel model that is able to use EH technique to reduce brown energy while equipping energy storage to maintain stable system performance. We then investigate the dependent task scheduling problem in this model and formulate it as a multiobjective task scheduling problem. Later, an improved multiobjective African vulture optimization algorithm (IMOAVOA) is proposed to solve this problem with the goal of minimizing the brown energy consumption and execution latency of the application. Simulation results demonstrate that IMOAVOA outperforms the other three benchmark algorithms, which can obtain better Pareto fronts and provide optimal task scheduling policy for this system.