Energy- and performance-aware load-balancing in vehicular fog computing

Abstract

An IoT-enabled cluster of automobiles provides a rich source of computational resources, in addition to facilitating efficient collaboration with vehicle-to-vehicle and vehicle-to-infrastructure communication. This is enabled by vehicular fog computing where vehicles are used as fog nodes and provide cloud-like services to the Internet of things (IoT) and are further integrated with the traditional cloud to collaboratively complete the tasks. However, efficient load management in vehicular fog computing is a challenging task due to the dynamic nature of the vehicular ad-hoc network (VANET). In this context, we propose a cluster-enabled capacity-based load-balancing approach to perform energy- and performance-aware vehicular fog distributed computing for efficiently processing the IoT jobs. The paper proposes a dynamic clustering approach that takes into account the position, speed, and direction of vehicles to form their clusters that act as the pool of computing resources. The paper also proposes a mechanism for identifying a vehicle's departure time from the cluster, which allows predicting the future position of the vehicle within the dynamic network. Furthermore, the paper provides a capacity-based load-distribution mechanism for performing load-balancing at the intra- as well as the inter-cluster level of the vehicular fog network. The simulation results are obtained using the state-of-the-art NS2 network simulation environment. The results show that the proposed scheme achieves balanced network energy consumption, reduced network delay, and improved network utilization.