Save table from dataset to delta table

[Shrujan1612]

I'm exploring the capabilities of Dataflow Gen2 in Power BI Fabric, and I have a use case where I want to:

1.Read data from an existing Power BI dataset, and

2. Push that data into a Delta table (e.g. in OneLake ).

Is this possible using Dataflow Gen2?

[V-yubandi-msft]

Hi @Shrujan1612 ,

Has your issue been resolved, or do you require any further information? Your feedback is valuable to us. If the solution was effective, please mark it as 'Accepted Solution' to assist other community members experiencing the same issue.

[V-yubandi-msft]

Hi @Shrujan1612 ,

Could you let us know if your issue has been resolved or if you are still experiencing difficulties? Your feedback is valuable to the community and can help others facing similar problems.

Thank You.

[V-yubandi-msft]

Hi @Shrujan1612 ,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank You.

[V-yubandi-msft · Accepted solution]

Hi @Shrujan1612 ,

Thank you for engaging with the Microsoft Fabric Community.

1. Reading data from a Power BI dataset using Dataflow Gen2 is currently not supported. Dataflow Gen2 allows connections to many sources, including Power BI Dataflows (Legacy), but not to Power BI Datasets (the semantic models used in reports). These are separate components, and as of now, datasets cannot be used as a data source in Dataflow Gen2.

Alternative approaches include.

Connecting Dataflow Gen2 directly to the original data source that feeds the Power BI dataset.

Exporting the dataset data manually to Excel/CSV and ingesting that into a Lakehouse and shared data source.

2. Dataflow Gen2 supports writing output to Delta tables in OneLake via Lakehouse, allowing data to be persisted in an open format usable by Spark, SQL, and other Fabric engines. You can schedule refreshes in Dataflow Gen2 to keep the Delta tables in sync with your data source.

If my response resolved your query, kindly mark it as the Accepted Solution to assist others. Additionally, I would be grateful for a 'Kudos' if you found my response helpful.