Execute Dataflow Tables/Queries in Parallel

[brent_reffner]

The dataflow refresh history allows you to see the execution times by table including start and end time. You can clearly see that tables are executed sequentially. I have not found any option which allows you to execute in parallel as you can in a dataset. Hopefully, this is something the Microsoft Power BI team will address.

Requested on	Dataflow name	Dataflow refresh status	Table name	Refresh status	Start time	End time	Duration
10/13/2023 3:42	Common Dataflow	Completed	Calendar	Completed	10/13/2023 3:42:02	10/13/2023 3:42:33	00:31.1
10/13/2023 3:42	Common Dataflow	Completed	Customer	Completed	10/13/2023 3:42:33	10/13/2023 3:43:04	00:30.5
10/13/2023 3:43	Common Dataflow	Completed	Item	Completed	10/13/2023 3:43:04	10/13/2023 3:43:35	00:30.6

[asocorrocognex]

Do all the tables depend on the same query?  For example, do you have a query Q that is the starting point for all the others?  That would cause sequential executions.

[edhans]

Those are really small. I've got some dataflows with long running times and there is some concurrency.

[brent_reffner]

I gave those as an example. The entire dataflow takes 45 minutes with the longest table taking 11 minutes.  When you say you have some concurrency, what do you mean and was that something you controlled?

[edhans]

Meaning I've seen times during the refresh where multiple tables/entities are refreshed in parallel.

But you have no control. It does what it does, and concurrency might be a factor of the license/capacity, dependencies, etc.