Proving the Business Case for the Internet of Things

Microsoft makes Azure IoT Edge generally available

Steve Rogerson
July 12, 2018

Just over a year after launching the preview version of Azure IoT Edge, Microsoft has announced that the IoT cloud offering has been made generally available globally.
The company said that since its introduction there have been many examples of the real-world impact from the factory floor to the farm to run cloud intelligence directly on IoT devices. Devices can act immediately on real-time data, whether it be recognising a crack in a pipe from an aerial view or predicting equipment failure before it happens.
“As we evolve towards a world of ubiquitous computing, the design of the IoT solution spanning hardware, edge and cloud must be consistent and secure to drive real impact,” said Sam George, partner director for Azure IoT. “We are excited to announce Azure IoT Edge is now generally available globally, enabling our growing list of enterprise customers to bring their edge solutions to production.”
Microsoft is also introducing capabilities on Azure IoT Edge to develop and deploy intelligence to the edge. These updates aim to position Azure IoT Edge as a true end-to-end offering for enterprise-grade edge deployments.
With the generally availability, IoT Edge is open sourced and available on GitHub. This continues the firm’s commitment to open source support for Azure IoT Edge and gives developers flexibility and control of their edge products, enabling them to modify the runtime and debug issues.
There is support for the Moby container management system. Moby is the open-source platform Docker is built on, allowing Microsoft to extend the concepts of containerisation, isolation and management from the cloud to devices at the edge. Moby containers work on Docker-based systems, and vice versa. There are no changes required to existing Docker-based modules.
The company is also expanding the Azure Certified for IoT programme to certify core edge functionalities such as device management and security. Users can find already certified edge hardware in the device catalogue. In addition to hardware, developers can find pre-built edge modules available through Azure Marketplace to accelerate edge development.
Azure IoT Edge has deep integration with the device provisioning service for zero-touch provisioning so a device can simply be provisioned in the field with no operator intervention. With the device provisioning service, users can securely provision tens of thousands of devices, bringing scale to edge deployments.
The Azure IoT Edge security manager acts as a well-bounded security core for protecting the IoT edge device and all its components by abstracting the secure silicon hardware. It is the focal point for security hardening and provides device makers the opportunity to harden their devices based on their choice of hardware secure modules.
An automatic device management (ADM) service allows scaled deployment of IoT Edge modules to a fleet of devices based on device meta data. When a device with the right meta data (tags) joins the fleet, ADM brings down the right modules and puts the edge device in the correct state.
There is broad language support for module SDKs. Azure IoT Edge supports languages such as C#, C, Node.js, Python and Java. Developers can simplify module development by coding, testing, debugging and deploying all from VSCode.
CI and CD pipeline with VSTS allows managing of the complete lifecycle of the Azure IoT Edge modules from development, testing, staging and finally deployment. All this is possible in VSTS using tools already familiar to developers.
At its recent Build developer conference, Microsoft announced IoT Edge capabilities including integration with Microsoft AI services and container support for Kubernetes, partnerships and integrations with third-party hardware including DJI drones and Qualcomm vision AI developer kit at the edge.
“Our preview customers are building sophisticated solutions that enable them to gain real-time insights in remote areas or areas with unreliable connectivity using Microsoft’s AI services supported on Azure IoT Edge,” said George.
These include Schneider Electric, which is doing predictive maintenance on equipment that signals it is about to break so it does not become a safety and waste hazard. And a farmer in Washington is using DJI drones and Microsoft’s FarmBeats on Azure IoT Edge to do precision agriculture.
"As a leading manufacturer of industrial equipment, we leverage cloud intelligence to understand how the equipment is performing and when maintenance is needed,” said Alasdair Monk, group product management director at Weir Group. “With Azure IoT Edge, we have the potential to bring this intelligence down to the device level, to predict failures and maintenance in real time, without a 24/7 dependence on connectivity." 
There are three components required for Azure IoT Edge deployment: Azure IoT Edge Runtime, Azure IoT Hub, and edge modules. The Azure IoT Edge runtime is free and will be available as open source code. Users will need an Azure IoT Hub instance for edge device management and deployment if they are not using one for their IoT offering already.
In spring, Microsoft announced that it would invest $5bn in IoT and the intelligent edge over the next four years to accelerate innovation while also simplifying the user experience to build IoT products.