Canonical Azure IoT architectures include several key components:
Data collection from devices in buildings, equipment, or the environment
Data transmission to Azure
Additional analytics, reporting, and communication with downstream applications such as Dynamics 365 Field Service.
One of the tools available from Microsoft for the on-premises components is Azure IoT Edge, on which you can deploy modules remotely to perform various actions. In an IoT scenario, you might deploy a device gateway such as ICONICS IoTWorX to get data from devices using standard protocols such as BACnet or Modbus, perform some transformations or analytics, and push that data to Azure IoT Hub. You can read about how to do this, for example, here. Similarly, you might deploy a Microsoft analytical tool such as Azure Stream Analytics to perform certain actions. You can read about how to do this here. Both of these examples show how to deploy a single application from Azure onto IoT Edge as part of an IoT architecture, communicating between the application and Microsoft Azure. Currently missing from the published literature, however, is how to get modules running on IoT Edge to communicate with each other, thereby enabling much more powerful solutions to be created on the edge.
Enter a new how-to article with an example of how to do exactly that: on how to set up communication between two modules in Azure IoT Edge. Specifically on configuring ICONICS' IoTWorX device gateway to pull data from sensors, process it, and then push it locally to Azure Stream Analytics (ASA) on IoT Edge for further processing before sending to Azure. Schematically, essentially like this:
While both IoTWorX and ASA are powerful products in their own right, combining the two adds opportunities for even more sophisticated data analytics. But again the main point of this article is not about setting up IoTWorX or ASA on IoT Edge, it is to provide an example of setting up inter-module communications on the edge to give system integrators an additional tool for developing enterprise applications. Read the full whitepaper here.