Re: Slow 3 Table Join with v bad row estimate

From: pgsql-performance-owner(at)postgresql(dot)org [mailto:pgsql-performance-owner(at)postgresql(dot)org] On Behalf Of David Osborne
Sent: Tuesday, November 10, 2015 12:32 PM
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: [PERFORM] Slow 3 Table Join with v bad row estimate

Ok - wow.

Adding that index, I get the same estimate of 1 row, but a runtime of ~450ms.
A 23000ms improvement.

http://explain.depesz.com/s/TzF8h

This is great. So as a general rule of thumb, if I see a Join Filter removing an excessive number of rows, I can check if that condition can be added to an index from the same table which is already being scanned.

Thanks for this!

David,
I believe the plan you are posting is the old plan.
Could you please post explain analyze with the index that Tom suggested?

Regards,
Igor Neyman