IoT Protocol Selection for Smart Grid Applications: Merging Qualitative and Quantitative Metrics

The concept of Internet of Things (IoT) implies that objects are connected to share information with each other and/or with a platform based on standardized communication protocols such as MQTT and HTTP. Recently, their performance has received much attention due to more widespread utilization. But, the vital aspect of the adoption of these IoT protocols is not just performance per se but also their feature set. However, comparative analysis that takes into account both the features and performance aspects are rarely presented. This paper investigates a combination of a qualitative comparison model (based on communication, operational and security attributes) with quantitative metrics in terms of CPU/memory utilization and time consumption. Quantitative metrics are collected for two of the four IoT protocols. Our analysis shows that there is no onesize-fits-all solution in terms of application layer IoT protocols. Additionally we emphasize that, on a higher level, both qualitative and quantitative measures are necessary for IoT stakeholders to consider before selecting the most appropriate protocol for the specific application to be deployed. We illustrate the results using a smart grid example simulating a collection of 1000 smart meters connected to the grid.