Scalability of Real-Time IoT-based Applications for Smart Cities

Abstract

The Internet of Things (IoT) is getting momentum, which drives us to design solutions able to deal with huge amounts of data coming from different sorts of sensors in order to make decisions to adapt system behavior automatically. While in recent years many IoT-based reasoning systems have already been proposed, there are no comprehensive results reporting their performance, particularly in complex environments. As an answer to that challenge, developers often choose an architecture design based on previous experience that have an impact on the system performance and scalability. This paper shows experimental results of a performance analysis study of different implementations of context-aware management architectures for IoT-based smart cities. Results show that different architectural choices affect system scalability and that automatic real time decision-making is feasible in an environment composed of dozens of thousands of sensors continuously transmitting data.