Slow Aggregation Join in M Language

[ttinsley]

The problem,

I was joining a student table with trainings table.  Join on empoloyees ID, then expand trainings and look for the FIRST training date for each student.

It was SO SLow.  And though the trianing completion file is only 5 MB, it was showing >1 Gig "downloaded" from that training completion file.

 

[ttinsley · Accepted solution]

I created a new ETL table in M language that does not load into the model.  In that table, I used "Group By" to group to the MIN training date.  Then joined the tables based off this already grouped and summarized table (as opposed to joining the Raw Training Completion table and asking the Expand of that join to do the aggregation for me.)

SO MUCH FASTER