With the exponential development of cities, the traffic flow prediction has become an important step in the development of transport systems. The main purpose is to improve the logistic services and reduce the cost of the road congestion. In this paper, we propose an extended iCanCloud framework simulation to evaluate the performance of VANET-Cloud architectures. In addition, we propose a VANET-Cloud architecture uses connected traffic detectors (e.g., speed loop trap) to collect traffic data in real time. After that, these data are used to predict the traffic flow. In the simulation step, the proposed architecture was tested by the extended framework, where it provides better performances at peak hours in terms of service latency and data accuracy.