Methodical Analysis of a Fog Computing Assisted Animal-Welfare Software System in a Real-World Smart Dairy Farm IoT Deployment.

In the IoT era, the devices along the things-to-cloud continuum, present a unique opportunity to additionally serve as computing hubs. Termed Fog computing, this paradigm can be used to host applications and process data closer to the source. In this article, we present a methodical analysis of our fog enabled software system in an IoT enabled smart dairy farm. The developed software system uses locomotion data generated by wearables on cows’ feet to detect anomalies in their behaviour. We analyze the benefits of using a fog computing assisted approach for developing such IoT solutions. We use resource utilization as the performance metric for analyzing the benefits of leveraging the fog computing paradigm compared to the traditional cloud centric approach. The results suggest that a fog enabled software system brings benefits such as efficient utilization of computing resources, improved QoS etc. The evaluation indicates that there will be need of special design (including both low-level and high-level system design) re-configurations and also re-engineering of some components to provide higher scalability using less computational resources.