Ultra-Low Latency Multi-Task Offloading in Mobile Edge Computing

With the development of computer technology, computational-intensive and delay-sensitive applications are emerging endlessly, they limited by computing power battery life Smart Mobile Devices (SMDs). edge (MEC) is a computation model with great potential to meet application requirements alleviate burdens on SMDs through offloading. However, device mobility server status variability in multi-server multi-task scenario bring challenges To cope these challenges, we first propose parallel task offloading small area-based scheme MEC. Then, formulate optimization problem minimize completion time all tasks, transform into deep reinforcement learning-based Markov decision approach. Furthermore, present deterministic policy gradient (DDPG) approach for obtaining strategy. Experimental results demonstrate that DDPG- based improves long-term performance at least 19% ultra-low latency, efficient usage servers, frequent over traditional strategies.