As businesses learn to harness the transformational power of IoT, IoT devices are becoming a mission-critical business asset. Today, IoT solutions use IoT devices to sense things in the real world with processing and decision making happening in the cloud, but as IoT continues to mature there are many use cases where it’s more appropriate to process data or take action directly on the IoT device itself.
Earlier this year at our //Build developer conference, we introduced a revolutionary new product, Azure IoT Edge, to address these needs. Azure IoT Edge enables businesses to run cloud intelligence directly on IoT devices even smaller than a Raspberry Pi or as powerful as they need.
Today at our Connect(); developer conference, we are thrilled to announce the public preview of new Azure IoT Edge capabilities including support for:
- AI Toolkit for Azure IoT Edge
- Azure Machine Learning
- Azure Stream Analytics
- Azure Functions
- Your own code in Containers
- Protocol adaptor as modules – OPC-UA and Modbus
Build AI applications for the Edge
Traditional IoT sensors measure things like temperature, humidity, acceleration, vibration, and more. While this is powering IoT solutions today, IoT solutions in the future need much more sophisticated sensors – for example, sensors that can detect visual defects, identify objects, and find visual anomalies. Azure IoT Edge now includes support for AI to enable these scenarios and more.
Today we’re announcing the AI Toolkit for Azure IoT Edge to jumpstart the process of creating AI applications that run at the edge. Developers can build AI applications with the toolkit with any framework using Azure Machine Learning, and then easily deploy and manage models to the Azure IoT Edge, including a set of pre-built models for common tasks.
You can find the AI Toolkit for Azure IoT Edge on Github.
Azure IoT Edge in detail
Azure IoT Edge can be used in many IoT scenarios. As an example, a complex data pipeline can be created on Azure IoT Edge (running on an edge device) pulling data from IoT devices and running it in a combination of Azure Machine Learning, Azure Stream Analytics, Azure Functions, and any third-party code. This pipeline can be configured and deployed from Azure IoT Hub in the cloud, with the Azure IoT Edge device pulling down the appropriate containers with these services and linking them.
Azure IoT Edge is designed to run on multiple platforms (Windows and many versions of Linux), and hardware architectures (x64 and ARM). To deploy workloads, Azure IoT Edge can use Linux Containers for Docker or Windows Containers for Docker, with an open design to incorporate number of popular container management systems.
Azure IoT Edge also allows developers to write their own code in multiple languages (C#, C and Python for now, more coming in the future) and deploy to Azure IoT Edge. We provide tools to develop, debug and deploy this code in containers in VSCode (for C#). In addition, Azure IoT Hub has the user experiences to not only deploy Edge modules on a single device, but at scale on a fleet of IoT Edge devices. This functionality is available in Azure portal, as well as APIs for businesses to build their own business applications for deployment and configuration management. Azure IoT Edge is available in most Azure regions today, including US West Central, East Asia, North Europe, and West US, and the rest of the regions will be available shortly.
Many customers are already seeing benefit and new opportunities with Azure IoT Edge. Here is what they are saying:
“Azure IoT Edge provided an easy way to package and deploy our Machine Learning applications. Traditionally, machine learning is something that has only run in the cloud, but for many IoT scenarios that isn’t good enough, because you want to run your application as close as possible to any events. Now we have the flexibility to run it in the cloud or at the edge—wherever we need it to be.”
– Matt Boujonnier, Analytics Application Architect for Schneider Electric
“NEC sees great value in Azure Stream Analytics on IoT Edge to increase the responsiveness of IoT solutions, while ensuring data privacy and sovereignty by processing data locally on edge devices. We see great potential to use this service across both our own IoT solutions, and also those of our customers who benefit from NEC’s Azure Plus consultancy."
– Hiroyuki Ochiai, Director, IT platform division, NEC Corporation
“The term ‘intelligence at the edge’ for Sandvik Coromant means doing useful processing of the data as close to the collection point as possible, allowing systems to make some operational decisions there. At Sandvik Coromant, we are streaming data from manufacturing machines, industrial equipment, pipelines and other remote devices connected to the IIoT. By running the data through an analytics algorithm, at the edge inside a corporate network with Azure IoT Edge, we can set parameters on what information is worth sending to a cloud or on-premises data store for later use — and what isn't. Edge analytics makes it possible to react very quickly which as an example can prevent crashes in the machine, this will enable organizations to reduce or avoid unplanned equipment downtime.”
– Magnus Ekbäck, VP Business Development, Sandvik Coromant.
We will be doing a webinar on Azure IoT Edge soon. To get more information, register for the webinar today.
Commitment to security at the Edge
To empower developers building applications for Azure IoT Edge, security is a fundamental requirement for success. Just last month we announced updates to our work with NXP for LS1012 and Microchip for ATSAMA5D2, with both product families built using the ARM processor architecture with TrustZone technology. We will continue to work across the industry to make Azure IoT Edge a secure Intelligent Edge platform that is operating system, processor architecture, and hardware agnostic.
More news for Azure IoT
Today we’re also announcing the general availability for Azure Time Series Insights, a fully managed service for the analytics, storage, and visualization of time series data, to do real-time anomaly detection, data streaming, and analysis that will power apps at the edge. Since April, hundreds of customers have pushed 100’s of billions of events into TSI for use in production environments. Now, any organization that produces massive amounts of IoT data has a scalable, enterprise-grade solution for storing and gleaning insights from their data that can be applied to Azure IoT solutions in the cloud or on the edge.
By 2020, it’s estimated that there will be 30 billion connected devices, according to the IDC research group. The ability to securely provision and harness data from these devices at scale has never been so critical to business transformation. Microsoft is helping customers simplify their IoT journey through comprehensive and integrated services to manage devices and solutions at scale securely. Check out our demo and docs on how to get started with Azure IoT Edge today.
Source: Azure Blog Feed