Performance Issue after using GroupBy

[Anonymous]

Hi All,

I used group by to add an index column that restart counting at 1 when the unique id changes, i referred to this https://community.powerbi.com/t5/Desktop/power-query-condtitional-index-column-by-category-and/td-p/... and managed to do it with the following queries but when i try to sort one of the rows, it took more than 30 mins and it load up to 7Gb and still loading and i could not wait for it to finish loading. My source is a 7.65Mb text file with 8 million rows. i have 7 columns and two other table inside the query editor.

Please help me to solve this performance issue. 

 #"Grouped Rows" = Table.Group(#"Reordered Columns", {"Unique"}, {{"Index1", each Table.AddIndexColumn(_,"Index1", 1,1), type table}}),
 #"Expanded Index1" = Table.ExpandTableColumn(#"Grouped Rows", "Index1", {"Column1.1", "Custom", "Steptime", "Index", "Custom.1", "Index1"}, {"Index1.Column1.1", "Index1.Custom", "Index1.Steptime", "Index1.Index", "Index1.Custom.1", "Index1.Index1"})

Any help is greatly appreciated!

Thank You!

[ImkeF]

Hi @Anonymous ,

you will most likely solve this by using a buffer function somewhere before the group-step.

It's a bit of a trial and error to find the right place for it: 

https://www.thebiccountant.com/speedperformance-aspects/

[v-frfei-msft]

Hi @Anonymous ,

To optimize the performance, please refer to this case.

https://stackoverflow.com/questions/7943957/huge-performance-difference-when-using-group-by-vs-distinct

[AnkitBI]

This is unexpected. Can you share complete queries for all from Advanced Editor