> Not always. The case for having (col1,col2) might be very
> compelling.
True. But in our case, the table has like 8M rows, so and col1 is some kind of job identifier, so it's evenly
distributed.Col2 on the other hand is a customer id, so it has much higher cardinality. Previous DBA missed it during
creation,and it was never loud enough in the logs for me to notice it until recently. Looks like I need to do a
column-orderingaudit. :)
> If only someone else would offer to do it for me....
Don't look at me. My C is rustier than a 50-year-old bike chain. :)
______________________________________________
See http://www.peak6.com/email_disclaimer/ for terms and conditions related to this email