Time is money, time creates problems. Especially when the path from the site of data creation and processing is long - "long" in terms of distance, "long" in terms of poor data rates. In the era of networked Industry 4.0, an idea that is increasingly interesting for fast data processing is thus being revived: instead of sending data and information from the end device to the cloud and back to the end device, it is transmitted directly from end device to end device or primarily processed there. This gives such systems the advantage, among other things, of remaining fully operational even in case of a jittery, weak or faulty connection. On the other hand, not every valve terminal can easily be upgraded to a stand-alone server. Edge computing is the magic solution.

At its Connect(); developer conference, Microsoft announced the general beta version of Azure IoT Edge for this problematic situation. Azure IoT Edge is essentially a container system with its own runtime environment that makes advanced IoT functions small and compact enough to be deployed at the field level. The Azure tools enable companies to handle complex data pipelines on Azure IoT Edge, among other things, and connect them to Azure Machine Learning, Azure Stream Analytics, Azure Functions, or any third-party code. Azure IoT Edge runs on multiple platforms and hardware architectures, and allows developers to write their own code in C#, C, and Python. Protocol adapters for OPC UA and Modbus are already available as ready-made modules.