Task offloading strategy to maximize task completion rate in heterogeneous edge computing environment

Abstract

With the rapid growth of mobile devices, more and more time sensitive tasks are emerging. This kind of task often has strict requirements on delay. Edge computing solves this problem to a great extent. However, the computing resources of the edge server are limited. When the number of tasks is too large, some tasks may not be completed within the specified time. In this paper, we aim to maximize the task completion rate. We propose a priority task scheduling scheme based on coherent beamforming(CB) technique and successive interference cancellation(SIC) technique. CB technique enables edge devices to transmit to distant edge servers, and SIC technique ensures that multiple tasks can be received simultaneously by edge servers. We first build the system model, and show that the model is a NP-hard problem. For solving the problem, in our algorithm, we first determine the transmission tasks in each time slot, and select a suitable server for each task. After tasks arrive at servers, we insert them into task queues according to their priorities. In simulations, we compare our schemes with other four schemes. Simulation results show that our scheme can improve the task completion rate and reduce completion delay.