Elastic Task Offloading Through the Cooperation of Multiple Road-Side Unit (RSU)-Based Edge Servers

Deploying powerful servers on Road-Side Units (RSUs) is a typical solution to provisioning computation offloading service to the vehicles on road. However, the computation workload among the RSUs often varies spatially and temporally, which calls for cross-server cooperation. To this end, we propose a multi-RSU cooperation-assisted task offloading strategy using the auction mechanism. Specifically, we design the Online Learning-driven Resource Trading Algorithm (OLRTA) for task offloading across multiple RSU-based edge servers. Besides, we conduct trace-driven simulation experiments based on a real dataset, and the results demonstrate that our method improves the task processing success rate by at least 23% and reduces the processing delay by at least 25% compared with other popular methods.