Join us at FabCon Atlanta from March 16 - 20, 2026, for the ultimate Fabric, Power BI, AI and SQL community-led event. Save $200 with code FABCOMM.
Register now!The Power BI Data Visualization World Championships is back! Get ahead of the game and start preparing now! Learn more
Hello together!
I ahve just started using PowerQuery in Fabric in Microsoft Edge to create several Dataflows. I have several questions regarding performance issues, and hope to find answers and tips for improvement here.
1. Are the calculations done locally on my PC, or on a "server" of some sort?
2. If the calculations are done on my local PC, is ther a way to allocate more RAM and increase overall performance?
3. When i perform CPU-expensive steps, i am facing the problem that my Join-keys get confused. I create a new join-operation, choose my columns accordingly, the estimation on the bottom tells me that everything matches (matched tables are very similar) and I press "OK". I then let PowerQuery do it's thing for a minute or two. If i now klick on the little Gear for my join-operation in the Queries, i get to see the following:
PowerQuery renews the numbers from left to right, which is a problem, because the order of my columns isn't the same. It then projects, that it cannot match anything with these keys. As my key contains several columns (more than shown in the image) i then have to redo all of them if i want to add one more attribute to it.
Is there a way to fix it - except the workaround of adding stuff to the code directly?
4. Filter operations sometimes randomly take ages, although any Step before is already done. In my opinion, filtering is "easy", am I correct? 3 Filter operations in a row freezed my Microsoft Edge.
Any help will be greatly appreciated. Thank you all in advance.
Best,
Alex
Merging in Power Query is generally very expensive and should be avoided. You seem to be merging based on multiple columns which is even worse. Consider creating composite keys.
You need to check if and how your transforms fold. If it doesn't then the filter steps are very (VERY) expensive.
Hey, thanks for your reply! I've now read some stuff about folding - unfortunately, I have to use several excel files, so I don't think there's much folding going on.
The composite keys however seemed to improve performance quite a bit!
I have two ways of combining my tables, one uses more Joins, the other one uses more filters - which one would you use and why?
EDIT: Added question
The Power BI Data Visualization World Championships is back! Get ahead of the game and start preparing now!
| User | Count |
|---|---|
| 19 | |
| 10 | |
| 9 | |
| 7 | |
| 7 |