MicrosoftHello Jose,
Thank you for your article. In fact what I'm looking for is to download the data in my computer and to process the data.
Is it possible to achieve this with Graph? Otherwise a simple export of the data from the dashboard will be convenient enough.
Regards,
Jean-Luc
MicrosoftAt the end of these steps, you will have an export of the data as JSON files in the Azure Storage account. You could then download it from there. Note that the process has several steps to set up the required resources, create the pipeline and run it.
Can you connect Microsoft Graph to Azure Event hub, instead of using Azure storage?
MicrosoftMGDC is designed to deliver data periodically (daily, weekly) to an Azure Storage account (or to Microsoft Fabric). If you need the data delivered to an Azure Event hub, you will need additional steps.
Copilot suggested:
To transfer data from Azure Storage (like Blob Storage) to Azure Event Hub, you’ll need to set up a pipeline that reads the data and pushes it as events. Azure doesn’t offer a direct built-in connector for this, but here are several effective approaches:
Use a time-triggered Logic App to pull data from Blob Storage and send it to Event Hub.
This is ideal for simple workflows and low-volume data.
🔗 Learn more: Connect to Event Hub from Logic Apps
Write a custom Azure Function that reacts to blob creation or modification events.
This gives you full control over the transformation and routing logic.
Use Stream Analytics to stream data from Blob Storage to Event Hub.
⚠️ Parquet files aren’t natively supported, so you may need preprocessing.
Set up Event Grid to monitor blob changes and route events to Event Hub.
This is useful for event-driven architectures where you want to react to blob changes.
🔗 Example setup: Azure Event Hub listening to blob changes
How big would be the data stored in Azure storage account? It will be overwritten after every pipeline? Are there other option for export data except Power BI? Splunk?
MicrosoftHi!
You will get one row/object per device in the company.
The size of a row varies, but it should be less than 1KB per row/object.
The data will not overwrite previous jobs (it uses unique IDs for each run), but you could delete old versions if you want.
Hope this helps,
Jose
Thank you.
