Forum Discussion
Direct Query Date Conversion in DAX - Performance Issues
This date table solution didn't seem to work for performance, but that could have been due to other constructions in my report, so I've decided to split the report into functional sub groups, in this way one report becomes many, which isn't ideal, but the populations are about 1/5 what the total report was originally. I sacrificed scope for run time.
It's a Direct Query from an ERP, so something in the order of 700K with the hard filtering limits that I placed in M Query.
Perhaps you don't need to create this column. You can locally import or create a date table that contains the text data type date then connect the two tables using this column.
I'll give this a try for the four most important dates, and see how the performance is effected.
The data cube's current design requires this volume to calculate correctly, so I suspect that the size is too big, in which case there are going to be some unhappy data cube designers when we next meet... We're designing another cube together but that's 3 months off. Fingers crossed that this is within acceptable limits as leadership wants this ASAP...
This date table solution didn't seem to work for performance, but that could have been due to other constructions in my report, so I've decided to split the report into functional sub groups, in this way one report becomes many, which isn't ideal, but the populations are about 1/5 what the total report was originally. I sacrificed scope for run time.