Can someone tell me if merge queries is a performance intensive operation?
At the moment I am not creating any queries to merge buy just conecting tables via the relationship table designer and using visualisation filtering on the report page.
Does this improve performance?
Performance is a complicated thing. There are so many things that can influence the performance. Usually, "merge tables" won't influence the performance. For instance, the performance won't be good whether we merge the tables or don't if the two tables are too big. "Merge tables" is helpful when we have complicated conditions related to many columns, so we can bring them together to make it easy. You can reference here for more information about improving performance. https://blogs.msdn.microsoft.com/charles_sterling/2017/01/31/how-to-tank-your-power-bi-performance/
So what your saying that if I had 2 large tables say > 50K records the best approach would be NOT to merge them, but to bring all records into the PB Desktop andt hen fiter out the records that I don't need using visualisation filtering?
Usually, there is no difference. There are two aspects I can find out. You can take into consider.
Could you please mark the proper answer if it's convenient for you? That will be a help to others.
Ressurecting this thread !
What I really want to know is what is the difference in terms of performance when applying queries to the dataset using the query editor vs filtering on the visualisation. ?