Advance your Data & AI career with 50 days of live learning, dataviz contests, hands-on challenges, study groups & certifications and more!
Get registeredGet Fabric Certified for FREE during Fabric Data Days. Don't miss your chance! Learn more
Hi All
We have a dataset with 100K rows in the following format:
And we have a flow that is getting data from this dataset
Due to the fact that flow cannot get all 100k rows in one go we are getting it in bunches of 25k rows, using autoincremented index, after that we are creating csv file from it - no problems so far.
But we also need to be able to sort  data by any of those columns and then get data in 25k rows bunches, so we cannot use min-max ID for that any more 😞 , and becouse we have a lots of columns, creating index for each column is not an option, all to be done dynamically.
 So the question is - how can we get 100k rows of dynamicaly sorted data from dataset into flow ?
Thanks
Solved! Go to Solution.
You could use the WINDOW function, that allows you to order by whichever column you want and you can provide the offsets you need.
You could use the WINDOW function, that allows you to order by whichever column you want and you can provide the offsets you need.
Finaly works with
WINDOW(0, REL, 10 , ABS 
Thanks for your reply
I am trying it with WINDOW with following query on a 100 rows table
     EVALUATE  
         VAR Window1 =  WINDOW(1, REL, 5 , REL, Data100Rows, MATCHBY([Index]))
    RETURN Window1
But it returns all 100 rows instead of 5 rows
I think it wants to be ABS, not REL
EVALUATE
VAR Window1 =
    WINDOW ( 1, ABS, 5, ABS, Data100Rows, MATCHBY ( [Index] ) )
RETURN
    Window1
😪
If you're getting all the data each time, does it matter what order it is in ? Also, have you considered using a Live Connected xlsx file rather than exporting to CSV ?
100 rows is just for testing
we have a 200k rows dataset which we need to convert into csv with a Flow (powerautomate), and flow can only take 25-30k rows in one go, so I am doing some looping to get all the data. But we also need to sort that data by different columns before getting it to csv 
 
					
				
				
			
		
Advance your Data & AI career with 50 days of live learning, contests, hands-on challenges, study groups & certifications and more!
Check out the October 2025 Power BI update to learn about new features.
 
            | User | Count | 
|---|---|
| 9 | |
| 5 | |
| 4 | |
| 3 | |
| 3 | 
| User | Count | 
|---|---|
| 23 | |
| 12 | |
| 11 | |
| 9 | |
| 8 |