Abstract
Vehicular cloud computing is a new paradigm that makes use of cloud computing resources to overcome the restraints of vehicular computing. It allows the sharing of resources such as storage capacity, computational power, and Internet connectivity from those vehicles in which these resources remain idle for long hours (in parking or heavy traffic jams). This chapter presents an availability study of the vehicular cloud. Due to its multilayered architecture, composite modeling technique is desired for availability analysis of vehicular clouds. Distinct models are developed for each subsystem using reliability block diagrams and semi-Markov process, and the models are then combined to evaluate the availability of the complete system. Two different sensitivity analysis techniques (partial derivatives and percentage difference) to determine the parameters that cause the greatest impact on the availability of the vehicular cloud are also applied. The analysis reflects that the availability can be improved by aiming on reduced set of parameters that affect the availability most.
