Load Distribution and Service Time Analysis of a Vehicular Edge Computing System

Due to the evolution of vehicles, most of them have some amount of computational capacity. Vehicular Edge Computing (VEC) has been introduced in recent years due to its augmentation of computational capacity. With the advance of VEC, vehicles are used to reduce latency and improve the quality of service (QoS) of network-based information services. In this paper, vehicles are used as edge servers and are grouped as clusters depending on the service groups that contain popular application services. Whenever a mobile device requests a service to the base station, the edge gateway in the base station sends the request to the cluster that has the requested service. This paper provides the load distribution mechanism within the cluster using the Deep Q Network. Moreover, the total processing time on the edge servers is compared with that of the cloud servers. It was shown that processing in the edge server is better than the cloud server when the length of the request message is short, or the wireless transmission speed is fast.