The need to extend the features of Cloud computing to the edge of the network has fueled the development of new computing architectures, such as Fog computing. When put together, the combined and continuous use of fog and cloud computing, lays the foundation for a new and highly heterogeneous computing ecosystem, making the most out of both, cloud and fog. Incipient research efforts are devoted to propose a management architecture to properly manage such combination of resources, such as the reference architecture proposed by the OpenFog Consortium or the recent Fog-to-Cloud (F2C). In this paper, we pay attention to such a combined ecosystem and particularly evaluate the potential benefits of F2C in dynamic scenarios, considering computing resources mobility and different traffic patterns. By means of extensive simulations we specifically study the aspects of service response time, network bandwidth occupancy, power consumption and service disruption probability. The results indicate that a combined fog-to-cloud architecture brings significant performance benefits in comparison with the traditional standalone Cloud, e.g., over 50% reduction in terms of power consumption.