Query Folding compared to writing SQL statement

[Bokchoy]

Hi all,

When importing data from SQL, i usually load in the data alongside a pre-written SQL statement with most transformation defined. From my understanding of QueryFolding, it does the exact same thing for you step by step with each transformation done on PowerQuery.

Is there any difference in performance given that transformation is identical between both methods? Wouldn’t having a pre-written SQL statement be more efficient on the SQL side and prevent potential table locking?
Also, can i enable Queryfolding after hand when I import the data with an SQL script?

[SpartaBI · Accepted solution]

@Bokchoy 
Pre-written SQL statement with NO power query transformations is the most efficient.
If you use it then you can't add power query steps or you will lose folding.
If you use the GUI of Power Query then you can also achieve Query folding but that's depend on the transformations you do. 
You can check to see if the query is folded by right clicking the last step and view the native query if available. Best thing is to check with SQL profiler what is the queries being sent to the data source to verify if there was a query folding or not.

      

Showcase Report – Contoso By SpartaBI