How much can Fog Computing enhance performances of heterogeneous delay-sensitive services in Smart Cities?

The growing number of urban residents leads to an intensive sharing of the natural and industrial resources of a city, as well as the embeddedness of billions of equipments and systems in the urban environment. Thus, the integration of Information and Communication Technology (ICT) and the Internet of Things (IoT) paradigm results on a “smart” city where “smart” services and applications, like smart surveillance, smart healthcare and smart transportation, are performed. Moreover, the Cloud computing paradigm is expected to provide on-demand software and hardware resources for gathered data treatment in such environment. However, the IoT/Cloud integration may induce higher energy consumption level and increased response-time, which is not suitable especially for delay-sensitive applications. This is because of the data processing by distant cloud datacenters, located in the core of the network, far from end-users. Thus, Fog computing seems to be a suitable solution to enhance delay-sensitive applications performance in an IoT/Cloud integration for smart cities, as, with fog nodes, data is processed at the edge of the network, closer to end-users. Several previous works showed the benefit of fog computing over IoT/Cloud integration, simulating one specific application. However, in this paper, we propose a fog-based model for delay-sensitive services in a smart city, simulating four realtime applications simultaneously. Our model records better results in terms of latency, network usage and energy consumption, compared to a Cloud-only reference model. Simulation is carried with iFogSim tool to perform four different real-time services, representing healthcare, transportation, content processing and public safety fields in a smart city.