Forum Discussion
Rolling sum performance on based on second date column
I have a cash flow table with the characteristics: cob_date: date that the data is generated for cash_flow_date: the date for which the cash flow is produced for amount: the cash flow amount var...
Yeah... My advice is this: Build a correct star-schema model with proper date tables. If you want to know why, there are YT videos by Alberto Ferrari that explain why that should be. Great chances are that when you start doing things as Best Practices dictate, you won't have any issues even on very big models.
I have also tested remaking the measure by referencing a date table in a relationship with the [cash_flow_date] column. Unfortunately still resulting in bad performance.