Azure IoT capabilities offer a number of benefits to customers seeking to manage and operate embedded devices at scale but also to manage, deploy and operate Edge devices deployed to remote endpoints at scale. IoT Edge can be deployed to Windows/Linux devices running ARM/x86 processors of varying sizes up-to and including low-power compute devices like a Raspberry PI.

IoT Edge also offers some unique and powerful capabilities for industrial customers running manufacturing, inspection or control applications inside a plant. These workloads require low-latency connections to align with the tact-time of assembly/manufacturing operations and often times customers are concerned (rightfully so) about relying on WAN connectivity for their plant to operate. This is where IoT Edge can be deployed to offer value to customers needing to run, local in the plant, capabilities to support data collection, processing or to integrate as part of a closed-looped control system where close proximity to devices like Programmable Logic Controllers (PLCs) are running process control functions in a plant. Typically these customers have additional security concerns opening up industrial control system networks that are isolated by design for security reasons directly the Internet or to Internet connected services. IoT Edge can play an important role in this scenario to deploy local to a customer's environment features of Azure such as Azure functions, ML services, data collection/aggregation operations and closed-loop control applications in a secure and low-latency manner while offering the ability to manage and operate these environments at scale from the cloud.

So how can an Edge device connect to these industrial control networks? Usually these networks are specialized to the manufacturer of the PLC such as ControlNet, Profinet, Modbus and many others. For a compute device to easily communicate to these networks manufacturing plants typically employ OPC, or Open Platform Communications as a means to facilitate direct communication to industrial control hardware. This involves software from vendors such as Kepware, Rockwell Automation, Iconics, Siemens and countless others the bridge the gap between the control network and open-APIs that can be used by software to pull/push data to these PLCs running industrial applications/operations.

So how does this extend to the Azure IoT Edge space and how can this help? IoT Edge offers capabilities to run custom 'code' or commercially supported modules on the compute appliance to perform tasks or to be triggered by an external event. Microsoft makes available Industrial IoT modules that can be deployed to IoT Edge devices deployed to plants that allow these devices to easily extract and interact or be triggered by data/state changes on PLCs using OPC! This Hack will involve building a solution on how this can be leveraged to solve many common engineering scenarios in the industrial control space.

IoT Process Control at the Edge Hackathon