Engineering Resource-Efficient Data Management for Smart Cities with Apache Kafka ^†

In terms of the calibre and variety of services offered to end users, smart city management is undergoing a dramatic transformation. The parties involved in delivering pervasive applications can now solve key issues in the big data value chain, including data gathering, analysis, and processing, storage, curation, and real-world data visualisation. This trend is being driven by Industry 4.0, which calls for the servitisation of data and products across all industries, including the field of smart cities, where people, sensors, and technology work closely together. In order to implement reactive services such as situational awareness, video surveillance, and geo-localisation while constantly preserving the safety and privacy of affected persons, the data generated by omnipresent devices needs to be processed fast. This paper proposes a modular architecture to (i) leverage cutting-edge technologies for data acquisition, management, and distribution (such as Apache Kafka and Apache NiFi); (ii) develop a multi-layer engineering solution for revealing valuable and hidden societal knowledge in the context of smart cities processing multi-modal, real-time, and heterogeneous data flows; and (iii) address the key challenges in tasks involving complex data flows and offer general guidelines to solve them. In order to create an effective system for the monitoring and servitisation of smart city assets with a scalable platform that proves its usefulness in numerous smart city use cases with various needs, we deduced some guidelines from an experimental setting performed in collaboration with leading industrial technical departments. Ultimately, when deployed in production, the proposed data platform will contribute toward the goal of revealing valuable and hidden societal knowledge in the context of smart cities.