From: Peter Schuller <peter.schuller@infidyne.com>
To: pgsql-performance@postgresql.org
Subject: Re: shared_buffers in 8.3 w/ lots of RAM on dedicated PG machine
Date: Mon, 18 Feb 2008 08:39:46 +0100
User-Agent: KMail/1.9.7
Cc: Greg Smith <gsmith@gregsmith.com>
References: <20080215123528.GA16532@hyperion.scode.org>
	<20080215135846.GA17327@hyperion.scode.org>
	<Pine.GSO.4.64.0802150916380.28161@westnet.com>
In-Reply-To: <Pine.GSO.4.64.0802150916380.28161@westnet.com>
MIME-Version: 1.0
Content-Type: multipart/signed; boundary="nextPart3003813.aTUxKR4d5m";
	protocol="application/pgp-signature"; micalg=pgp-sha1
Content-Transfer-Encoding: 7bit
Message-Id: <200802180839.46557.peter.schuller@infidyne.com>

--nextPart3003813.aTUxKR4d5m
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

> PostgreSQL only uses direct I/O for writing to the WAL; everything else
> goes through the regular OS buffer cache unless you force it to do
> otherwise at the OS level (like some Solaris setups do with
> forcedirectio).  This is one reason it still make not make sense to give
> an extremely high percentage of RAM to PostgreSQL even with improvements
> in managing it. =20

Ok - thank you for the input (that goes for everyone).

> Another is that shared_buffers memory has to be=20
> reconciled with disk at every checkpoint, where OS buffers do not.

Hmm. Am I interpreting that correctly in that dirty buffers need to be flus=
hed=20
to disk at checkpoints? That makes perfect sense - but why would that not b=
e=20
the case with OS buffers? My understanding is that the point of the=20
checkpoint is to essentially obsolete old WAL data in order to recycle the=
=20
space, which would require flushing the data in question first (i.e., =20
normally you just fsync the WAL, but when you want to recycle space you nee=
d=20
fsync() for the barrier and are then free to nuke the old WAL).

=2D-=20
/ Peter Schuller

PGP userID: 0xE9758B7D or 'Peter Schuller <peter.schuller@infidyne.com>'
Key retrieval: Send an E-Mail to getpgpkey@scode.org
E-Mail: peter.schuller@infidyne.com Web: http://www.scode.org


--nextPart3003813.aTUxKR4d5m
Content-Type: application/pgp-signature; name=signature.asc 
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.4 (FreeBSD)

iD8DBQBHuTZCDNor2+l1i30RAnUuAKCAZo9Ifl2qjXFAuTGaV3yrXQIPYwCeJZhm
6w5rCIoHO850luHVd26oU78=
=C6kJ
-----END PGP SIGNATURE-----

--nextPart3003813.aTUxKR4d5m--