Microservices In IoT-based Remote Patient Monitoring Systems: Redesign of a Monolith

Since over half of the world population in remote areas does not have access to medical practitioners, Remote Patient Monitoring (RPM) systems have proved to be a viable alternative. However, because these systems are powered by many IoT devices, they require efficient software architectures. Therefore, an increasing number of applications of microservice architectures in the RPM systems with different models have been proposed. On the other hand, there is still an open question on how to decompose and apply microservice principles in IoT-based systems, while retaining benefits of microservices like maintainability. Therefore, the goals of this study are to investigate how microservices should be applied in the RPM systems and whether they improve software maintainability. To achieve that we rebuild the existing SBIoT-MPH system for monitoring hypertension using microservice principles and decomposition patterns. Moreover, to validate the findings we evaluate the end artefact and discuss improved maintainability and the sensor data ingestion throughput. With this study, we aim to provide both researchers and practitioners with a general and comprehensive solution to build microservice-based RPM systems.