Research on Computing Task Offloading Mechanism Based on Cloud-Edge Collaboration

Mobile edge computing (MEC) as an emerging and prospective computing paradigm, can significantly enhance computation capability and save energy of smart mobile devices (SMDs) by offloading computation-intensive tasks from resource-constrained SMDs to the resource-rich cloud. Most edge computing offloading algorithms only consider latency and energy consumption, while neglecting the load balancing on edge servers. Therefore, in this paper a load balancing model is introduced. The load balance model is based on the utilization of edge servers. At the same time, in order to achieve load balancing goals, cloud-edge collaboration is needed. In addition, the traditional particle swarm optimization algorithm cannot be applied to discrete solution spaces, an improved multi-dimensional discrete particle swarm optimization (MDPSO) algorithm is proposed to solve multi-objective optimization problems based on our load balancing model. At the same time, the improved algorithm MDPSO can reduce the degree of manual intervention in the selection process of learning factors’, while accelerating the convergence speed of the algorithm. Finally, a simulation platform was built to verify the algorithm, and the simulation results demonstrate that the improved algorithm effectively reduces network latency and energy consumption, achieving global optimization while maintaining load balance on edge servers.