From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756015Ab1ISMkp (ORCPT ); Mon, 19 Sep 2011 08:40:45 -0400 Received: from tamriel.snowman.net ([72.66.115.51]:55754 "EHLO tamriel.snowman.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755756Ab1ISMkn (ORCPT ); Mon, 19 Sep 2011 08:40:43 -0400 X-Greylist: delayed 582 seconds by postgrey-1.27 at vger.kernel.org; Mon, 19 Sep 2011 08:40:43 EDT Date: Mon, 19 Sep 2011 08:31:00 -0400 From: Stephen Frost To: Benjamin LaHaise Cc: Andres Freund , Matthew Wilcox , Andi Kleen , viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, robertmhaas@gmail.com, pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Improve lseek scalability v3 Message-ID: <20110919123100.GJ12765@tamriel.snowman.net> References: <1316128013-21980-1-git-send-email-andi@firstfloor.org> <201109161616.50004.andres@anarazel.de> <20110916153620.GA9913@parisc-linux.org> <201109161927.34472.andres@anarazel.de> <20110916200817.GD28519@kvack.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="GmIAiBEz2jSsbeLA" Content-Disposition: inline In-Reply-To: <20110916200817.GD28519@kvack.org> X-Editor: Vim http://www.vim.org/ X-Info: http://www.snowman.net X-Operating-System: Linux/2.6.32-5-amd64 (x86_64) X-Uptime: 08:30:20 up 228 days, 21:00, 13 users, load average: 0.28, 0.14, 0.06 User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --GmIAiBEz2jSsbeLA Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable * Benjamin LaHaise (bcrl@kvack.org) wrote: > For such tables, can't Postgres track the size of the file internally? I= 'm=20 > assuming it's keeping file descriptors open on the tables it manages, in= =20 > which case when it writes to a file to extend it, the internally stored s= ize=20 > could be updated. Not making a syscall at all would scale far better tha= n=20 > even a modified lseek() will perform. We'd have to have it in shared memory and have a lock around it, it wouldn't be cheap at all. Thanks, =09 Stephen --GmIAiBEz2jSsbeLA Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) iEYEARECAAYFAk53NgQACgkQrzgMPqB3kigQkACbBedEBkNbhU0Eg/LgMO7xPcdZ VRgAn20rNS0pVWDTsFdladAr9UzPxo8Z =K4cW -----END PGP SIGNATURE----- --GmIAiBEz2jSsbeLA--