Reference vs Duplicate tables in Power Query, Power BI

[Jocie]

Hi Everyone,

I have a weird situation in the tables in Power BI.  There is a big parent table that gets loaded into Power BI from SQL Server.  Once it is loaded into Power BI, multiple subsets gets created depending on various criteria.  And all the subsets have to interact with each and with the parent table through various filter selections.  I am able to achieve this through Power Query.

My doubt is, which practice is good ?  Is making a subset out of the parent table through "reference" option or "duplicate" option in Power Query?  I know using reference is a neat way of replicating things.  But my concern in this scenario is the performance, specially if they have to behave independently with a big volume of data and interactions. 

Any help/thoughts/ideas are appreciated.

Thanks in advance.  

Cheers.

[amitchandak · Accepted solution]

@Jocie , If you want to create a reference, do not do a change in the original script.

It should be like

Base Table

Created D1,D2, D3

Created a Fact Using Base Table, D1, D2, D3 using merge as new.

If you want to change something to base, then before any change duplicate else you will have all the steps, when you copy.

I would prefer the first approach. not to disturb the base table