Cost minimization scheduling for deadline constrained applications on vehicular cloud infrastructure

Abstract

Two categories of reasons have led to the emergence of vehicular cloud computing; on one hand vehicular networks and the new generation of well-equipped smart cars and on another, the advent of cloud computing and its maturity over a short time-span. Owing to these two facts, the concept of vehicular cloud computing has been shaped up in the recent years in which cloud services are offered through underutilized resources of vehicles that construct a dynamic groups of autonomous vehicles and therefore create a cloud. The difference between this set-up and conventional structures of cloud computing lies in the mobility of nodes which leads to change in the dynamics of resource availability over time. The goal of this study is to offer an application scheduling model for determining the optimum response needed for management of dynamic vehicular cloud resources in a way that tasks are completed with minimum cost; before their deadlines and within the lifetime of the cloud. To solve the aforementioned problem, a binary integer program model is formulated here, and the impact of changes in various parameters such as different tasks costs, application deadlines, types and lifetime of created clouds are analyzed and evaluated. The presented results, specify and highlight the factors that should be taken into consideration in the process of application scheduling and demonstrate how our proposed optimization model could result in reliable solutions for the vehicular computing optimization problems.