Scalable Real-Time Analytics for IoT Applications

Large-scale industrial internet of things (IIoT) applications usually access distributed equipment where high-volume sensor streams are processed. The building of scalable analytic queries and models over such streams could potentially enhance various industrial processes management tasks, e.g., distribution, delivery, and predictive online maintenance. To enable real-time and historical analytics over distributed IIoT applications, we have combined an edge data stream management system (EDSMS), sa.engine, with the highly distributed NoSQL database MongoDB. For supporting advanced analytics and high-volume stream injection into MongoDB, we integrated an extended query processing (EQP) system with sa.engine and MongoDB. This work demonstrates how EQP provides a holistic data management solution for IIoT based on a use case for sound anomaly detection of distributed equipment.