Error when connecting to SQL server analysis services database

Is there any resolution to this in Power BI updates since this post. We are also facing the same issue with loading more data from Mutidimensional SSAS cube into Power BI. We need to use DAX to acheive the desired customization in Power BI. Hence need to go for Import Mode. Is increasing MemoryBufferLimit in SSAS a recommended and tested option. We need to do impact analysis if at all needs to change settings for SSAS standard installation.

Thanks, I will look at that.

I don't get that error when using Excel with the same data from Analysis Services though.

John

Hi All,

My client entered a ticket on this and they had her run the same exercise using excel - where we received no error.  HOWEVER:  What we discovered was that we were treating a query to LOAD data from an SSAS Tabular cube the same way we defined that query in 'get data' for a live connection to the cube, expecting the same results.  In live connection,  we selected all the dimensions and one of the fact tables defined within that cube.  What we got from that single 'Gest Data'/ 'Query', was the set of folders representing each entity in that cube with nomenclature as appeared in the cube itself.   However, the result when LOADing data was a single folder with the name of the cube and each data element named in the format of <entityname>.<attributename>.<attributename>.  For whatever reason, the process of generating the cube into a flat representation of that data would finish when pulling via Power Query in Excel, and result in an error when doing so in Power BI.  We further tested the same with the fact table and a few small dimensions (i.e. a subset of dimensions) in Power BI.  Sometimes it would finish, sometimes the error above. 

Because the result was NOT what we intended:  One folder for each cube entity, we had to load each entity within the cube, one entity at a time (i.e created a new query using Get Data), as one would do for each star schema entity in SQL Server.  Because that STILL resulted in oddly named folders and entity names, we had to edit each query to result in the desired folder names and attributes as seen in the model itself.  Without those changes, we ended up with each folder name being the same name as the cube, or appended with a number 1-n-1.  (Example:  CubeName, CubeName1, CubeName2...)  As before, each attribute was named in a very 'user unfriendly' manner:  EntityName.AttributeName.AttributeName. 

End Result: We never were able to determine why it fails when loading into Power BI desktop and does not fail when loading into Excel via Power Query (Now 'Get Data' in Excel 2016).  That said, the results were not as we expected with data returned in a single folder - hence we loaded each entity via a seperate Get Data/Query. 

PS:  I am entering a new suggestion to change the above behavior regarding nomenclature when loading SSAS Cube entities.

Hope this helps!