如何在Microsoft Azure IoT中创建流数据的自定义仪表板？

Question

I have created an Event Hub in the Azure portal. I am streaming Temperature data into the hub (1 value per minute) and would like to create 2 plots (a Temperature time series and a Temperature histogram of the respective last 480 minutes) in a dashboard. I have created a Stream Analytics job to achieve this. It takes the Event Hub data as input.

Ideally I'd like to have the visualizations created directly from the data stream without any data storage, ie "stream -> dashboard" rather than "stream -> storage -> dashboard". I'm used to creating dashboards in Tableau and I'd love to use a similarly intuitive and maybe even interactive tool.

One option may be to store the streaming data in Azure and to import from Azure to Tableau. Another option might be to output the Stream Analytics job to Microsoft Power BI.

Question is: What other/more convenient options are there to create my dashboard from my streaming data? In particular: Are there any native ways within Azure for creating customizable visualizations/dashboards of streaming data? Does Azure IoT Suite enable creating customizable dashboards of streaming data?

Thanks.

Answer 1

Did you looked into https://powerbi.microsoft.com/en-us/blog/using-power-bi-real-time-dashboards-to-display-iot-sensor-data-a-step-by-step-tutorial/ ?

It shows how to communicate with PowerBI directly through API. If you want to keep routing capabilities and extra functionality which IOT Hub provides - create azure functions subscribed to event-hub triggers.

Within a function once it is triggered you can call PowerBI apis. IOTHub exposes event hubs endpoints, so you can have azure function triggers hooked up which will call Power BI.

Updated: I created a sample for using RaspberryPI, Azure IotHub, Azure Functions and PowerBi. See https://github.com/gtrifonov/azure-iothub-functions-powerbi . You can apply additional azure iothub functionality such as routing,ip blocking and etc if you route messages through hub. Also azure functions give you extra point of adding some processing.