From: Tom Lane <tgl@sss.pgh.pa.us>
To: Dane Foster <studdugie@gmail.com>
cc: psql-performance <pgsql-performance@postgresql.org>
Subject: Re: Plan weirdness. A sort produces more rows than the node beneath
 it
In-reply-to: 
 <CA+WxinK_PeQBeZ-SWVqheHNjyjtD3QGZ5gZSzVKxHRYgUG-ktw@mail.gmail.com>
References: 
 <CA+WxinKHxRUgDwbUfsKNiq5MXn8kRFayo0KqZzTmxXG93BbK4g@mail.gmail.com>
 <3658472.1691161821@sss.pgh.pa.us>
 <CA+WxinK_PeQBeZ-SWVqheHNjyjtD3QGZ5gZSzVKxHRYgUG-ktw@mail.gmail.com>
Comments: In-reply-to Dane Foster <studdugie@gmail.com>
	message dated "Fri, 04 Aug 2023 11:15:40 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <3661331.1691163117.1@sss.pgh.pa.us>
Date: Fri, 04 Aug 2023 11:31:57 -0400
Message-ID: <3661332.1691163117@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/3661332.1691163117%40sss.pgh.pa.us>
Precedence: bulk

Dane Foster <studdugie@gmail.com> writes:
>> If the sort is the inner input to a merge join, this could reflect
>> mark-and-restore rescanning of the sort's output.  Are there a
>> whole lot of duplicate keys on the merge's other side?

> Yes. The course_id column's values repeat a LOT on the merge side.

Hmm.  The planner should avoid using a merge join if it knows that
to be true.  Maybe analyze'ing that table would prompt it to use
some other join method?

			regards, tom lane