Date: Sat, 11 Nov 2017 11:19:45 -0600
From: Justin Pryzby <pryzby@telsasoft.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-performance@postgresql.org
Subject: Re: overestimate on empty table
Message-ID: <20171111171945.GA2167@telsasoft.com>
References: <20171110204043.GS8563@telsasoft.com>
 <9772.1510348781@sss.pgh.pa.us>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <9772.1510348781@sss.pgh.pa.us>
User-Agent: Mutt/1.5.23 (2014-03-12)
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

On Fri, Nov 10, 2017 at 04:19:41PM -0500, Tom Lane wrote:
> Justin Pryzby <pryzby@telsasoft.com> writes:
> > (or, the opposite of the more common problem)

> > As the queued_alters table is typically empty (and autoanalyzed with
> > relpages=0), I see "why":
> 
> > ./src/backend/optimizer/util/plancat.c
> > |                        if (curpages < 10 &&
> > |                                rel->rd_rel->relpages == 0 &&
> > |                                !rel->rd_rel->relhassubclass &&
> > |                                rel->rd_rel->relkind != RELKIND_INDEX)
> > |                                curpages = 10;
> 
> So I'm sure you read the comment above that, too.

> One idea is to say that relpages = reltuples = 0 is only the state that
> prevails for a freshly-created table, and that VACUUM or ANALYZE should
> always set relpages to at least 1 even if the physical size is zero.

> Dunno if that would confuse people.

What about adding && rel->rd_rel->reltuples==0, and make VACUUM/ANALYZE instead
set only reltuples=1, since that's already done at costsize.c: clamp_row_est()
and therefor no additional confusion?

Justin


-- 
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance