Cloud computing is gaining more and more popularity due to the wide use of the Internet. It can now be integrated into vehicle applications to enhance mobility, preventing accidents, improve safety, and improve driving comfort. In this paper, we propose a secure architecture of inter-vehicle communication which is obtained from a cloud services, including architecture level, communication level and associated security challenges. With the help of these cloud services, real time vehicle simulations are conducted to find the optimal route from a given source to the destination based on some factors such as distance, road width, traffic count, junction delay and velocity - all of which play an essential role in minimizing the travel time. Instead of deploying sensors and devices on each vehicle, authorized drivers will subscribe to the cloud-provided platform and applications as services on-demand with payment based on usage. On the other hand, this study tackles one of the most important and conical attacks in cloud vehicle offering services, which is the Shrew attack. A number of steps should be taken first in order to explore these attacks like activating a loss-based probe to identify the locations and capabilities of the current bottlenecks, making use the data centers low-tenancy in order to synchronize the attacks flow and analyzing the necessary and the sufficient traffic for an effective attack.