A Configurable Distributed Data Analytics Infrastructure for the Industrial Internet of things

Abstract

Executing analytics functionalities over data from highly distributed data sources and data streams is at the very core of the vast majority of Industrial Internet of Things (IIoT) applications. State of the art streaming engines provide the means for high performance analytics over high velocity IIoT streams, yet they still need significant programming and customization efforts when deployed in heterogeneous industrial environments. This paper introduces a configurable engine for distributed data analytics for IIoT applications. The engine leverages the performance of state of the art data streaming middleware platforms, which it augments with a set of digital models for configuring DDA operations. As such the introduced engine reduces the effort needed to implement and deploy distributed data analytics in IIoT environments. The engine is available as open source software and has been validated in a various real-life IIoT applications in different environments.