Slow Dataflow editor: Queries that are quite fast in Power BI Desktop take long in the web editor

[trutz]

Hi all,

As others, I noticed that the performance of the Dataflow editor is really slow. Queries that execute quickly in the Desktop take a long time in the web-based editor.

Is there any setting in the Premium/PBI Embedded capacity management, that allows to speed up things here. Looking at the QPU measure in Azure metrics for my Embedded node, I see no usage when just editing the dataflow queries in my Premium-enabled workspace.

So I was asking myself, if the editor takes advantage of my dedicated capacity or is just run on the shared service.

Any details on this?

[GilbertQ]

Hi there

As far as I have seen this is how it works in the experience because it is web based it often does take longer to do something in Power BI Desktop (which runs as an application)

[trutz]

Still it is strange why the power of the dedicated capacity is not put to good use to speed up the edit experience.

[Anonymous]

Hi @trutz,

I also checked the official documents but not found they told about where the M query formula executed.
Perhaps you can try to add the following custom function to trace the process of common data set and computed entity to confirm if premium capacity adds more resource to process these queries:

Timing M Query Execution In Power Query And Power BI (Revisited) 
Regards,

Xiaoxin Sheng