X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [64.117.224.130]) by svr1.postgresql.org (Postfix) with ESMTP id 298E0D1C511 for ; Mon, 28 Jul 2003 17:44:40 +0000 (GMT) Received: from svr1.postgresql.org ([64.117.224.193]) by localhost (neptune.hub.org [64.117.224.130]) (amavisd-new, port 10024) with ESMTP id 07123-03 for ; Mon, 28 Jul 2003 14:43:43 -0300 (ADT) Received: from lucifer.akyasociados.com.ar (unknown [200.69.203.237]) by svr1.postgresql.org (Postfix) with ESMTP id 256CBD1C4E3 for ; Mon, 28 Jul 2003 14:44:37 -0300 (ADT) Received: from taz.oficina (taz.oficina [192.168.1.174]) (authenticated bits=0) by lucifer.akyasociados.com.ar (8.12.9/8.12.9) with ESMTP id h6SHmp2J000756; Mon, 28 Jul 2003 14:48:51 -0300 (ART) (envelope-from franco@akyasociados.com.ar) Subject: Re: Mapping a database completly into Memory From: Franco Bruno Borghesi To: Josh Berkus Cc: Tom Lane , pgsql-performance@postgresql.org In-Reply-To: <200307280950.36076.josh@agliodbs.com> References: <97842368.20030727104901@artis.uni-oldenburg.de> <1059401815.749.7.camel@taz.oficina> <10185.1059409557@sss.pgh.pa.us> <200307280950.36076.josh@agliodbs.com> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-6xp8HrO23Z50AhoMANm1" Organization: AK y Asociados S.R.L. Message-Id: <1059414481.1957.1.camel@taz.oficina> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.0 Date: 28 Jul 2003 14:48:02 -0300 X-MailScanner-Information: Please contact the ISP for more information X-MailScanner: Found to be clean X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200307/447 X-Sequence-Number: 2740 --=-6xp8HrO23Z50AhoMANm1 Content-Type: multipart/alternative; boundary="=-UrP0EGn2eV4SfGs9VKAw" --=-UrP0EGn2eV4SfGs9VKAw Content-Type: text/plain Content-Transfer-Encoding: quoted-printable But I think it's still a good option.=20 For example, in servers where there are other applications running (a web server, for example) that are constantly accesing the disk and replacing cached postgresql pages in the kernel, having shared buffers could reduce this efect and assure the precense of our pages in memory... I gues :) On Mon, 2003-07-28 at 13:50, Josh Berkus wrote: > Tom, >=20 > > If we had a portable way > > of preventing the kernel from caching the same page, it would make more > > sense to run with large shared_buffers. >=20 > Really? I thought we wanted to move the other way ... that is, if we cou= ld=20 > get over the portability issues, eliminate shared_buffers entirely and re= ly=20 > completely on the OS cache. --=-UrP0EGn2eV4SfGs9VKAw Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable But I think it's still a good option.

For example, in servers where there are other applications running (a web s= erver, for example) that are constantly accesing the disk and replacing cac= hed postgresql pages in the kernel, having shared buffers could reduce this= efect and assure the precense of our pages in memory... I gues :)

On Mon, 2003-07-28 at 13:50, Josh Berkus wrote:
Tom,

> If we had a portable way
> of preventing the kernel from caching the same page, it would make more
> sense to run with large shared_buffers.

Really?  I thought we wanted to move the other way ... that is, if we could=
=20
get over the portability issues, eliminate shared_buffers entirely and rely=
=20
completely on the OS cache.
--=-UrP0EGn2eV4SfGs9VKAw-- --=-6xp8HrO23Z50AhoMANm1 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (FreeBSD) iD8DBQA/JWHR21dVnhLsBV0RAogxAJ9veZe97qtA2bhSMQWrEeRnFK3KbwCffLMy QJch389SPV72kivv8frmQqw= =VkEW -----END PGP SIGNATURE----- --=-6xp8HrO23Z50AhoMANm1--