Scheduling Security-Critical Workflow Tasks in Mobile Edge Computing

Abstract

Mobile edge computing can deliver high-throughput and low-latency computing services by means of offloading computation tasks of mobile devices to edge servers. This paper aims at scheduling security-critical workflow tasks, which require data encryption and decryption during the offloading procedure and have precedence relations among each other, in a mobile edge computing environment. The scheduling problem is formulated as an optimization problem that minimizes workflow execution time and total energy consumption under precedence constraints upon security-critical tasks. The optimization model takes into account the time and energy overhead for data encryption and decryption. We propose a new particle swarm optimization algorithm to solve the resulting optimization problem. This algorithm uses a position-based mapping operator to convert each particle into a high-quality feasible solution, which is represented by a task sequence. For each converted solution, we develop a greedy search strategy to offload workflow tasks onto edge servers such that the completion time and energy consumption can be evaluated. In addition, we incorporate Levy flight into the standard movement scheme for updating particle positions, to improve the computational efficiency of the particle updating procedure. Experimental results show that the proposed algorithm outperforms baseline approaches in terms of both effectiveness and efficiency when solving the considered scheduling problem.