Vehicular fog computing (VFC) is envisioned as an extension of cloud and mobile edge computing to utilize the rich sensing and processing resources available in vehicles. We focus on slow-moving cars that spend a significant time in urban traffic congestion as a potential pool of on-board sensors, video cameras and processing capacity. For leveraging the dynamic network and processing resources, we utilize a stochastic mobility model to select nodes with similar mobility patterns. We then design two distributed applications that are scaled in real-time and placed as multiple instances on selected vehicular fog nodes. We handle the unstable vehicular environment by a), Using real vehicle density data to build a realistic mobility model that helps in selecting nodes for service deployment b), Using community-detection algorithms for selecting a robust vehicular cluster using the predicted mobility behaviour of vehicles. The stability of the chosen cluster is validated using a graph centrality measure, and c), Graph-based placement heuristics are developed to find optimal placement of service graphs based on a multi-objective constrained optimization problem with the objective of efficient resource utilization. The heuristic solves an important problem of processing data generated from distributed devices by balancing the trade-off between increasing the number of service instances to have enough redundancy of processing instances to increase resilience in the service in case of node or link failure, versus reducing their number to minimise resource usage. We compare our heuristic to an integer linear program solution and a first-fit heuristic. Our approach performs better than these comparable schemes in terms of resource utilization and/or has a lesser service latency, which is a crucial requirement for safety-related applications.

Many vehicles spend a significant amount of time in urban traffic congestion. Due to the evolution of autonomous cars, driver assistance systems, and in-vehicle entertainment, many vehicles have plentiful computational and communication capacity. How can we deploy data collection and processing tasks on these (slowly) moving vehicles to productively use any spare resources? To answer this question, we study the efficient placement of distributed services on a moving vehicle cluster. We present a macroscopic flow model for an intersection in Dublin, Ireland, using real vehicle density data. We show that such aggregate flows are highly predictable (even though the paths of individual vehicles are not known in advance), making it viable to deploy services harnessing vehicles' sensing capabilities. Our main contribution is a detailed mathematical specification for a task-based, distributed service placement model that scales according to the resource requirements and is robust to the changes caused by the mobility of the cluster. We formulate this as a constrained optimization problem, with the objective of minimizing overall processing and communication costs. Our results show that jointly scaling tasks and finding a mobility-aware, optimal placement results in reduced processing and communication costs compared to an autonomous vehicular edge computing-based naïve solution.