A Two-Stage Intelligent Task Offloading Approach for Cloud-Edge Collaborative Computing Based on Particle Swarm Algorithm

Abstract

Cloud-edge collaborative computing plays a crucial role in the massive computing demands and enhancing application performance. However, this computing method faces numerous challenges, including network latency, resource imbalances, and changes in requirements. In this paper, we propose an intelligent two-level scheduling approach for tasks based on cloud-edge collaborative computing to meet the demand for real-time and low-latency in applications like IoT. First, the approach assigns tasks to edge or cloud according to the task types. Furthermore, it intelligently offloads some tasks to the cloud when the load on the edge is too high. Finally, it finds the optimal time and cost balance point through multi-objective optimization by particle swarm algorithm. Experimental results show that this method can fully utilize the edge and cloud resources, reduce data uploading, and control the cost while ensuring low latency.