Performance analysis of a fog computing architecture for the Internet of Things via case studies

Technological advancements and cheaper computational resources have allowed the integration of computing and communication resources into regular objects, even making it possible to get them connected to the Internet. Such heterogeneous scenario with things connected to the Internet is usually referred to as the Internet of Things (IoT). Conventionally, cloud computing is employed to enable the deployment of IoT solutions. However, factors impacting the connectivity of devices to the infrastructure in the cloud can affect device deployment and interaction. To bring the control closer to the objects themselves, the fog computing paradigm was introduced so that it could bring some of the main cloud capabilities closer to where the devices really are. This work compares the cloud and fog based solutions via case studies designed on an IoT platform. The results indicate that there are significant benefits when employing the fog approach, especially in terms of reducing the communication delay with the objects/devices.