Re: Slow 3 Table Join with v bad row estimate

[Igor Neyman <ineyman@xxxxxxxxxxxxxx>]

 

 

From: pgsql-performance-owner@xxxxxxxxxxxxxx [mailto:pgsql-performance-owner@xxxxxxxxxxxxxx] On Behalf Of David Osborne
Sent: Tuesday, November 10, 2015 12:32 PM
To: Tom Lane <tgl@xxxxxxxxxxxxx>
Cc: pgsql-performance@xxxxxxxxxxxxxx
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