In this project I discussed two types of potential Vehicular Cloud (Vehicular Cloud) applications. The first group includes two value added applications that aggregate the computing capabilities of parked vehicles; these applications benefit from a Vehicular Cloud but could, potentially, be run on top of conventional clouds. The second group includes two traffic-related applications that are highly dynamic in nature and cannot be run on conventional clouds. More specifically, suggested ways in which under-utilized computing, networking and storage facilities can be used effectively and/or be shared between drivers or rented out to other customers over the Internet.
The distinguishing characteristic that sets the Vehicular Clouds apart from conventional clouds is the dynamically changing amount of available resources. It is, therefore, clear that the Vehicular Clouds can only achieve their full potential if their basic architecture is tailored to offer a seamless integration and decentralized management of the cyber physical resources of the participating vehicles. In particular, the Vehicular Cloud architecture must dynamically adapt its managed vehicular resources allocated to an application according to dynamically changing requirements and systems conditions. The main contribution of this work is to take the first step towards implementing a non-trivial form of a Vehicular Cloud. Specifically, I envision a Vehicular Cloud involving cars in the long-term parking lot of a typical international airport. The patrons of such a parking lot are typically on travel for several days, providing a rather stable pool of cars that can serve as the basis of a datacenter at the airport. I envision a “park and plug” scenario where the cars that participate in the Vehicular Cloud are plugged into a standard power outlet and are provided cable (i.e. Ethernet) connection to a central server at the airport.
In the proposed system the data center at the airport is implemented using the aggregating the computing capabilities of parked vehicles. The data center resources are the pooled resources of the individual vehicles parked at the airport for long time that are willing to participate in the vehicular cloud.
The proposal in this project presented a distributed optimal routing algorithm to balance the load along multiple paths for multiple multicast sessions. Our measurement-based algorithm does not assume the existence of the gradient of an analytical cost function and is inspired by the unicast routing algorithm based on Simultaneous Perturbation Stochastic Approximation (SPSA).