Get started in Azure IoT with Just your Phone
Vision for getting started in Azure IoT without hardware
Sometimes you want to jump in and learn how to make IoT a reality in a cloud provider (Like Azure) but not get bogged down in X.509 certificates, Docker (and Moby) containers, and massive data engineering workflows.
Sometimes, you just want a quick working solution to understand how specific resources in Azure (Or other cloud providers) work.
From all this, Microsoft launched IoT Plug and Play.
There are some great resources that provide instructions on how to use IoT Plug and Play with Azure IoT Central (Their SaaS offering), and you can easily go further with Azure IoT Hubs, Streaming Analytics, Synapse, PowerBI, and more.
Purpose
The number and complexity of IoT devices is only growing. IoT Analytics forecasts 22% CAGR growth into 2025. IoT is expanding into nearly every application. Whether you’re growing as a data scientist, ML engineer, software developer, or business stakeholder, IoT is critical to demystify.
Too often, getting into IoT can require setting up Raspberry Pi devices, configuring the appropriate shields in Arduino, or leveraging other devices. These are all relatively low entry and have tons of open source instructions and guides. But sometimes, you may want to quickly explore IoT use cases or convey the simplicity of IoT to your team. This is where the Azure Plug and Play App comes in.
Download the Azure Plug and Play device from your app store to get started right away.
Low-Code / No-Code IoT: Azure IoT Central
You can quickly build a solution with Azure’s SaaS offering — Azure IoT Central. This Azure documentation (https://docs.microsoft.com/en-us/azure/iot-fundamentals/iot-phone-app-how-to) is a great reference as well.
This is a great starter service, as you can connect your device with just a QR code. Telemetry is easy to access and doesn’t require any complicated queries. It’s simple but can still be used for some industrial applications.
Actual-Code IoT: Azure IoT Hubs
You can make a lot of immediate progress and insight with Azure IoT Central and Azure IoT Plug and Play, and you can continue to go further with using PaaS (Platform as a Service) modules in Azure.
Here are some typical steps:
- Provision an Azure IoT Hubs module and create a device with symmetric key authentication; copy the connection string and paste it in the Azure IoT Plug and Play setup configuration to access the data via IoT Hubs
- Experiment with routings (See Message Routings blade) to send batches of messages directly to sources such as Azure Storage
- Experiment with analyzing streaming telemetry by provisioning the Azure Stream Analytics resources (Link to Example Query)
- Send Azure Stream Analytics queries to a variety of sources including PowerBI, Synapse, Databases, and more
Details on Connecting Plug and Play Device to IoT Hubs
Connecting PnP App to IoT Hubs
Create an Azure IoT Hub (You can use the free tier (F1) for this demo) and create a device as shown below:
Copy the primary connection string and send it to your phone:
For reference, this is Symmetric Key authentication. It is less secure than X.509 authentication, but much easier for rapid proof of concept projects.
Confirmation: Going to the the Device Twin, you can confirm the device is “Connected”
Initial Experimentation
An initial foray into PaaS Azure IoT could look like routing batches of IoT data to a storage account (e.g. Blob Storage), using Azure Stream Analytics to send data to PowerBI and SQL, and more.
This architecture leverages the message routing functionality to dump batches of IoT data directly to Azure Storage (e.g. Blob Storage) as well as provide some streaming analytics and visualizations with Azure Stream Analytics. You can use Azure Stream Analytics to stream data directly to PowerBI as well as stream to a database and later visualize in PowerBI. Obviously, the options can vary and grow in complexity as you think about architecting a solution for thousands of devices.
Potential Next Steps
If you’re intrigued and want to learn even more, I’d recommend the following resources:
- Study up for the AZ220 certification to go deeper in Azure IoT knowledge
- Buy a Raspberry Pi and start hacking
- Experiment with demystifying both IoT and Web3 with the Helium Network for IoT devices (I provide a demo here on Medium)
- Experiment with big data streaming analytics tools such as Synapse and Databricks
- Learn more about visualizing data with PowerBI
The ability to demystify new technology and learn by doing is powerful in growing fields such as IoT. As you can see, the path to mastery in such a big field is a long one, but getting started can take less than an hour. Get out there and build!
