A Fog Node Architecture for Real-Time Processing of Urban IoT Data Streams

Abstract

Sensors and IoT devices in smart cities generate vast volumes of data that need to be harnessed to help smart city applications make informed decisions on the fly. However, time-sensitive applications cannot tolerate sending data streams to the cloud for processing because of the unacceptable latency and network bandwidth requirements. Cities require the ability to efficiently stream data and process data streams in real-time at the edge. This paper describes a fog node-based architecture that aims to deal with the streaming and processing of data generated by the various devices and equipment of the city to enable the creation of value-added services. The core components of the fog node are a powerful distributed messaging system and a powerful stream processing engine that can adequately scale with the data volumes.