Peer‐assisted content delivery network by vehicular clouds: Algorithm and evaluation

Abstract

In this paper, we study how to improve the performance of content delivery networks (CDNs) using the novel concept of vehicular clouds. Our main idea is to use a cluster of connected vehicles (ie, vehicular clouds) to act as virtual servers in CDNs in order to relieve the workload of physical CDN servers. The key benefits include guaranteed content availability/reliability in terms of mean time between failures and optimized data storage workload at vehicular clouds. Workload is an important metric as it is directly related to the latency of content delivery. Specifically, we present a content placement algorithm that has an optimal workload at vehicular clouds. Finally, we verify the system efficacy using our own simulator based on simulation of urban mobility (SUMO) and Python.