The initial focus in IoT projects often is on connecting sensors and devices to an IoT platform. While this is challenging and interesting in itself, connecting devices can only be a means to an end: doing something interesting and profitable with the data collected by the devices. This talk will focus on what these interesting activities can be. Data can be analyzed using event streams or in batches. In the streaming case, patterns of events can be detected and events can be aggregated into new events that again serve as input for the stream analysis. In the batch case, data is analyzed after the fact to create reports or detect interesting patterns for the future. Analysis can be done using hard-coded rules or using machine learning models. Regardless of the analysis approach, the results of the analysis will in most case trigger activity in a backend system (e.g. an ERP system) or on a device. This device might not necessarily be the originating device as e.g. a sensor might have detected high pressure and another device is instructed to close to corresponding valve. Furthermore, the talk will present the architecture of Cumulocity, a modern IoT platform, and how it can be extended using microservices running in Docker containers.

Talk Level:
INTERMEDIATE

Bio:
Harald is a Senior Principal Consultant and the IoT Lead Architect with Software AG. For the last 10 years, he has been advising Fortune 500 companies on digital transformation projects using Software AG's Digital Business Platform.