Performance and size of shared dataflows

[joshua1990]

We work with Pro licenses and have a workspace that has dataflows with all the data. Other workspaces are based on exactly the same dataflows (linked).
What impact does shared dataflows have on the performance of the reports? I mean, what if 10 workspaces are based on the same dataflow?
And how to determine the size of a dataflow? At least not in the service. Do I need to submit all dataflows in PBI Desktop and then publish them?

[v-kkf-msft]

Hi @joshua1990 ,

Does your problem have been solved? If it is solved, please mark a reply which is helpful to you.

If the problem is still not resolved, please provide detailed error information or the expected result you expect. Let me know immediately, looking forward to your reply.

Best Regards,
Winniz

[jeffshieldsdev]

I think there wouldn't be any impact due to number of assets consuming from dataflows. The output of dataflows is saved as CSV files in data lake storage, so consumers are reading from that.

In Premium capacity, these CSVs can get cached in a managed SQL Server store, so there could be an impact there--but just DirectQuery reports and downstream dataflows using these dataflows for specific operations like merges and group by's. This would go against your Premium capacity consumption.

Two options for getting size of dataflows: use a ADLS Gen2 storage for your workspace (requires Premium), you can use dataflows or other tools to query the container then.

Other option is create dataflows from your dataflows and use Table.RowCount() to get a count of rows. You can also use Table.Profile() to get more metadata. There isn't really a good way to get size--pulling into PBI Desktop as you suggested the data will be compressed so the .pbix file size wouldn't be an accurate representation.