From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id q7C9EXNa091900 for ; Sun, 12 Aug 2012 04:14:33 -0500 Received: from mailsrv14.zmi.at (mailsrv14.zmi.at [212.69.164.54]) by cuda.sgi.com with ESMTP id udVuUe0WDnnIAwsn (version=TLSv1 cipher=AES256-SHA bits=256 verify=NO) for ; Sun, 12 Aug 2012 02:14:30 -0700 (PDT) Received: from mailsrv.i.zmi.at (mailgate.i.zmi.at [10.72.17.3]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (Client CN "mailsrv2.i.zmi.at", Issuer "power4u.zmi.at" (not verified)) by mailsrv14.zmi.at (Postfix) with ESMTPS id E99B31800185 for ; Sun, 12 Aug 2012 11:14:27 +0200 (CEST) Received: from saturn.localnet (saturn.i.zmi.at [10.72.27.2]) by mailsrv.i.zmi.at (Postfix) with ESMTP id B14B0C8C449 for ; Sun, 12 Aug 2012 11:14:36 +0200 (CEST) From: Michael Monnerie Subject: howto keep xfs directory searches fast for a long time Date: Sun, 12 Aug 2012 11:14:22 +0200 Message-ID: <6344220.LKveJofnHA@saturn> MIME-Version: 1.0 List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============8584240910641599949==" Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: xfs-oss --===============8584240910641599949== Content-Type: multipart/signed; boundary="nextPart1406184.WAYbJTnMsq"; micalg="pgp-sha1"; protocol="application/pgp-signature" Content-Transfer-Encoding: quoted-printable --nextPart1406184.WAYbJTnMsq Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" I need a VMware VM that has 8TB storage. As I can at max create a 2TB=20= disk, I need to add 4 disks, and use lvm to concat these. All is on top= =20 of a RAID5 or RAID6 store. The workload will be storage of mostly large media files (5TB mkv Video= =20 + 1TB mp3), plus backup of normal documents (1TB .odt,.doc,.pdf etc).=20= The server should be able to find files quickly, transfer speed is not=20= important. There won't be many deletes to media files, mostly uploads=20= and searching for files. Only when it grows full, old files will be=20 removed. But normal documents will be rsynced (used as backup=20 destination) regularly. I will set vm.vfs_cache_pressure =3D 10, this helps at least keeping=20= inodes cached when they were read once. - What is the best setup to get high speed on directory searches? Find,= =20 ls, du, etc. should be quick.=20 - Should I use inode64 or not? - If that's an 8 disk RAID-6, should I mkfs.xfs with 6*4 AGs? Or what=20= would be a good start, or wouldn't it matter at all? And as it'll be mostly big media files, should I use sunit/swidth set t= o=20 64KB/6*64KB, does that make sense? I'm asking because I had such a VM setup once, and while it was fairly=20= quick in the beginning, over time it felt much slower on traversing=20 directories, very seek bound. That xfs was only 80% filled, so shouldn'= t=20 have had a fragmentation problem. And I know nothing to fix that apart=20= from backup/restore, so maybe there's something to prevent that? --=20 mit freundlichen Gr=C3=BCssen, Michael Monnerie, Ing. BSc it-management Internet Services: Prot=C3=A9ger http://proteger.at [gesprochen: Prot-e-schee] Tel: +43 660 / 415 6531 --nextPart1406184.WAYbJTnMsq Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.18 (GNU/Linux) iEYEABECAAYFAlAnc/MACgkQzhSR9xwSCbTSGACeNOZQbhwltGIBGBV9RpZrxKLr zowAn1c6ykB4P+cBakkV0Hot3rNETdOJ =fW71 -----END PGP SIGNATURE----- --nextPart1406184.WAYbJTnMsq-- --===============8584240910641599949== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs --===============8584240910641599949==--