Multiple duplicate datasources

[PVosESKA]

Hi Guys, Long time PowerUser, first time poster on this board.

I recently got the task of optimizing workspaces and apps in our tenant.

In our most important workspaces and apps, there are connections to our Azure Analysis Services model. 

There are multiple developers in our organisation, each working in their PBI desktop app, creating connection to our AAS model, create their creation and upload the .PBIX to the Power BI service.

This has resulted in about 40 duplicate connections to the same datasource, each to connect to the same AAS model (See image below). 

Could there be any performance optimisation in caching data for example, by reducing the number of connections, or is it actually a good idea to split the datasets over these datasources?

Let me know what you think!

Regards,

Peter

[Greg_Deckler]

@PVosESKA Are these live/directquery connections to the tabular cube or import?

[PVosESKA]

Hi @Greg_Deckler ,

These are al live connections to the cube. Multiple workspaces, all with 10's of connections.

Regards,

[Greg_Deckler · Accepted solution]

@PVosESKA Not sure there is anything that can be done about it then. Since they are live connections to the cube, unless I am mistaken, you cannot have define a dataflow that is a live connection so no real way to consolidate.

[PVosESKA]

Hi @Greg_Deckler ,

Someone told me about a way of editing the datasets to use the same datasource, so 40 datasets could be using the same datasource.

I just wondered if there is any benefit of doing this, but judging by your answer, having multiple duplicate datasource connections is a 'normal' thing.

Thank you for taking the time to reply. I will mark your answer as the solution.

Regards,

Peter