GitHub Docs: The concise guide to IoT Edge on Azure: https://github.com/UCdavidUC/mty2020
- AI - How tu run AI models on the Edge and stream it to the cloud.
- CV - Deploy your Cognitive Service Computer Vision and get insights from what you see.
- Function - Create an IoT Edge integrated environment with Azure Functions running on the Edge.
- ML - Run your machine learning models on the edge and get predictions and classifications models streaming to the cloud.
- Stream Analytics - Gather only valuable inforamtion from your devices in real time, get your stream onto real time monitoring services in Azure with Power BI.
In this guide it is described how to prepare your Azure and IoT Edge environments for our different labs and scenarios.
For this scenarios I will be using a Raspberry Pi 3 B. Support for Raspberry Pi 4 will be comming soon.
You can also use a Linux or Windows environment. Plan ahead on aqcuiring a device with one CPU architectures: ARM32v7, AMD64 or ARM64 (in Preview, some features may not be available).
Enjoy!
- Create your resource group based on the region you want to provision your resorces.
- Use a meaningful name so you can identify what the resoruce's group contents are about.
- IoT Hub is you device gateway to the cloud.
- You can register your devices here and receive information from them such as: monitoring informacion, data gathered from the environment or the result of on Edge data processing.
- With Azure IoT Device Provisioning Service (IoT DPS) in Azure you can update all of your devices or specific single or grouped devices on zero touch on-deman way.
- Configure your IoT Hub on the available regions depending on the features you want to deploy. *
- IoT Hub may require some time to provision. Please be patient :D and have a coffee in the meantime.
- Once it is provision, you can go to the next step.
- Now it´s time to generate your Azure to the Edge endpoint by creating an IoT Edge device.
- In your IoT Hub tools on your left you can find a section called "Automatic device management", under this section click the "IoT Edge" button.
- In the top bar select "Add and IoT Edge device".
- Fill the information requested in the form and then click the "Save" button at the bottom.
- There you go! You've created an IoT Edge device onto your IoT Hub.
- Once your IoT Edge device is created you can select it from the Azure IoT Edge blade.
- You will see a set of fields containing different keys and connection strings, this information will be used to connect our device to our IoT Hub and create a bidirectional connection between Azure and your device.
- Copy both your Primary Key and connection string into a safe place, we will need it when we configure our IoT Edge device.
- As you can see in the bottom of the IoT Edge management blade there are two different modules:
- $edgeAgent
- $edgeHub
- These must be up and running when we finish the device setup in the following steps.
- First of all, you can provision almost any device that can support to run containers as mobi-engine and docker and devices with different operating systems: Linux or Windows.
- Not all devices are suitable, depending on the CPU architecture they may be supported or not.
- Please checkout the following link to know if your device is supported or if it is suitable for IoT Edge:
- Supported OS: https://docs.microsoft.com/en-us/azure/iot-edge/support
- IoT Edge Certified Devices: https://catalog.azureiotsolutions.com/alldevices?filters={%2218%22:[%221%22]}
- Debian 10 systems, including Raspian Buster, use a version of OpenSSL that IoT Edge doesn't support. Use the following command to install an earlier version before installing IoT Edge:
sudo apt-get install libssl1.0.2
Access to the source demo and guidelines: https://docs.microsoft.com/en-us/azure/iot-edge/tutorial-deploy-custom-vision
Note: IoT DPS service is only available at certain locations, please review the following link to check service availability in Azure: https://azure.microsoft.com/en-us/global-infrastructure/services/?products=iot-hub