MIME-Version: 1.0
In-Reply-To: 
 <CADFyZw4diKqonCmcnC=ac7VMEcJLu-LaYkH84uFq+E_hQ1Ugzg@mail.gmail.com>
References: 
 <CADFyZw7aGoD0AaStxdyHByR5Qta=M5wx0v=iptKLhPUp+EOKvA@mail.gmail.com>
 <CAHyXU0zhpFtoOHgC8Fvb92rs038ofc0VBP7aKVnv1=rsLCxazQ@mail.gmail.com>
 <8422687a-e7bb-abad-987b-a102573e3979@commandprompt.com>
 <CADFyZw4diKqonCmcnC=ac7VMEcJLu-LaYkH84uFq+E_hQ1Ugzg@mail.gmail.com>
From: bricklen <bricklen@gmail.com>
Date: Wed, 12 Jul 2017 07:11:54 -0700
Message-ID: 
 <CAGrpgQ89GO51Esvg=AE7-k=ypxrQSE_trzG5pAL16N2JhiX+vA@mail.gmail.com>
Subject: Re: Very poor read performance, query independent
To: Charles Nadeau <charles.nadeau@gmail.com>
Cc: postgres performance list <pgsql-performance@postgresql.org>
Content-Type: multipart/alternative; boundary="94eb2c05e5d2f4a64905541f651b"
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

--94eb2c05e5d2f4a64905541f651b
Content-Type: text/plain; charset="UTF-8"

On Wed, Jul 12, 2017 at 12:30 AM, Charles Nadeau <charles.nadeau@gmail.com>
wrote:

>
> I use noop as the scheduler because it is better to let the RAID
> controller re-arrange the IO operation before they reach the disk. Read
> ahead is set to 128:
>
> charles@hpdl380g6:~$ cat /sys/block/sdc/queue/read_ahead_kb
> 128
> charles@hpdl380g6:~$ cat /sys/block/sdc/queue/scheduler
> [noop] deadline cfq
>
>
>
Perhaps pg_test_fsync (
https://www.postgresql.org/docs/9.6/static/pgtestfsync.html) and
pg_test_timing will help shed some light here, or at the very least give
some numbers to compare against.

--94eb2c05e5d2f4a64905541f651b
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote">On W=
ed, Jul 12, 2017 at 12:30 AM, Charles Nadeau <span dir=3D"ltr">&lt;<a href=
=3D"mailto:charles.nadeau@gmail.com" target=3D"_blank">charles.nadeau@gmail=
.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"ma=
rgin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:=
1ex"><div dir=3D"ltr"><br><div>I use noop as the scheduler because it is be=
tter to let the RAID controller re-arrange the IO operation before they rea=
ch the disk. Read ahead is set to 128:</div><blockquote style=3D"margin:0px=
 0px 0px 40px;border-width:medium;border-style:none;border-color:currentcol=
or;padding:0px"><div>charles@hpdl380g6:~$ cat /sys/block/sdc/queue/read_<wb=
r>ahead_kb</div><div>128</div><div>charles@hpdl380g6:~$ cat /sys/block/sdc/=
queue/scheduler</div><div>[noop] deadline cfq=C2=A0</div></blockquote><div>=
<br></div></div></blockquote><div><br></div><div>Perhaps pg_test_fsync (<a =
href=3D"https://www.postgresql.org/docs/9.6/static/pgtestfsync.html">https:=
//www.postgresql.org/docs/9.6/static/pgtestfsync.html</a>) and pg_test_timi=
ng will help shed some light here, or at the very least give some numbers t=
o compare against. <br></div></div><br></div></div>

--94eb2c05e5d2f4a64905541f651b--