From: Tom Lane <tgl@sss.pgh.pa.us>
To: Laurenz Albe <laurenz.albe@cybertec.at>
cc: Jeff Janes <jeff.janes@gmail.com>, Peter Geoghegan <pg@bowt.ie>,
        Shijia Wei <shijiawei@utexas.edu>,
        Pgsql Performance <pgsql-performance@lists.postgresql.org>
Subject: Re: Consecutive Query Executions with Increasing Execution Time
In-reply-to: <2c41458116a7b80fa8eba117d14fdbd5489c4847.camel@cybertec.at>
References: 
 <CABSfb-6iUvpGCzcgni+tVoiUWTc9aQdUtKbOzFzGqJcypiLa4g@mail.gmail.com>
 <adb87d39c5faa26d85274824ea4dd15ffc44ac40.camel@cybertec.at>
 <CABSfb-55VAcUwvy+PtsBsEk8EMfODtC6p65qVHSYhOqWWf80mA@mail.gmail.com>
 <CAH2-WzneCi_hM5_JGoPvGqKJr_Qd5hXcQzABz9-tQVndw9vwpw@mail.gmail.com>
 <19575.1576529417@sss.pgh.pa.us>
 <e7acf6bcdf5f267d8195377e74bd12005ce5949d.camel@cybertec.at>
 <CAMkU=1w2i=Ph7840GYtf2eWta858Y3Eyb0zRKJd84cyEEctBZg@mail.gmail.com>
 <2c41458116a7b80fa8eba117d14fdbd5489c4847.camel@cybertec.at>
Comments: In-reply-to Laurenz Albe <laurenz.albe@cybertec.at>
	message dated "Wed, 18 Dec 2019 13:17:10 +0100"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <28391.1576676654.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Wed, 18 Dec 2019 08:44:14 -0500
Message-ID: <28392.1576676654@sss.pgh.pa.us>
Precedence: bulk

Laurenz Albe <laurenz.albe@cybertec.at> writes:
> On Tue, 2019-12-17 at 11:11 -0500, Jeff Janes wrote:
>> If it is doing a seq scan (I don't know if it is) they intentionally us=
e a
>> small ring buffer to, so they evict their own recently used blocks, rat=
her
>> than evicting other people's blocks.  So these blocks won't build up in
>> shared_buffers very rapidly just on the basis of repeated seq scans.

> Sure, but according to the execution plans it is doing a Parallel Index =
Only Scan.

Nonetheless, the presented test case consists of repeatedly doing
the same query, in a fresh session each time.  If there's not other
activity then this should reach some sort of steady state.  The
table is apparently fairly large, so I don't find it surprising
that the steady state fails to be 100% cached.

			regards, tom lane