Design a Distributed Fog Computing Scheme to Enhance Processing Performance in Real-Time IoT Applications

Abstract

The technology evolution and application popularity in cloud computing has driven the rapid development of the Internet of Things (IoT) services and applications. When these real-time events in IoT applications are transmitted to the cloud for processing, the load on cloud computing becomes heavier, which can result in processing delays and exceeding the processing deadline. To improve upon and solve problems derived from many and more real-time events in the cloud computing process, some have proposed a fog computing concept. Previous studies on fog computing are mainly focused on accurately defining the fog computing concept and its possible applications. The nodes in fog computing are closer to edge devices, which can process such these real-time events in this layer. With this fog process, processing efficiency can be significantly increased and delays can also be solved in cloud computing.

This paper design a set of fog computing framework called D-FOG scheme that can improve the better processing performance of real-time IoT applications than conventional cloud computing processing architecture. The experimental results of KPIs indicate that in Expired event rate can be lower by 28%, 34% and 28.2%; Event success rate can be increased of 21.2%, 26.9% and 20.48%, and also Average processing time can be reduced by 436.87 ms, 558.72 ms and 1320.45 ms in 100, 200 and 500 real-time events, respectively. Consequently, the proposed D-FOG scheme is more effective and efficient in real-time events processing than conventional cloud computing processing architecture.