From owner-freebsd-fs@FreeBSD.ORG Thu Nov 4 18:23:48 2010 Return-Path: Delivered-To: freebsd-fs@FreeBSD.ORG Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 645B01065673 for ; Thu, 4 Nov 2010 18:23:48 +0000 (UTC) (envelope-from olli@lurza.secnetix.de) Received: from lurza.secnetix.de (lurza.secnetix.de [IPv6:2a01:170:102f::2]) by mx1.freebsd.org (Postfix) with ESMTP id D51F08FC0A for ; Thu, 4 Nov 2010 18:23:47 +0000 (UTC) Received: from lurza.secnetix.de (localhost [127.0.0.1]) by lurza.secnetix.de (8.14.3/8.14.3) with ESMTP id oA4INT2A097845; Thu, 4 Nov 2010 19:23:44 +0100 (CET) (envelope-from oliver.fromme@secnetix.de) Received: (from olli@localhost) by lurza.secnetix.de (8.14.3/8.14.3/Submit) id oA4INTIL097844; Thu, 4 Nov 2010 19:23:29 +0100 (CET) (envelope-from olli) Date: Thu, 4 Nov 2010 19:23:29 +0100 (CET) Message-Id: <201011041823.oA4INTIL097844@lurza.secnetix.de> From: Oliver Fromme To: freebsd-fs@FreeBSD.ORG, freebsd@jdc.parodius.com In-Reply-To: <20101104143542.GA628@icarus.home.lan> X-Newsgroups: list.freebsd-fs User-Agent: tin/1.8.3-20070201 ("Scotasay") (UNIX) (FreeBSD/6.4-PRERELEASE-20080904 (i386)) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.3.5 (lurza.secnetix.de [127.0.0.1]); Thu, 04 Nov 2010 19:23:45 +0100 (CET) Cc: Subject: Re: Using an SSD "disk" for / X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: freebsd-fs@FreeBSD.ORG, freebsd@jdc.parodius.com List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 04 Nov 2010 18:23:48 -0000 Jeremy Chadwick wrote: > On Thu, Nov 04, 2010 at 03:29:22PM +0100, Oliver Fromme wrote: > > What kind of tuning do you apply to those file systems? > > > > I'm asking because I can't really reproduce those results. > > On a machine with plenty of RAM I've created memory disks > > for obj and src, which should at least be as fast (probably > > even faster) than an SSD. Buildworld was somewhat faster, > > compared to standard (well-tuned) UFS+SU filesystems on a > > fast SATA HDD, but not _that_ much faster. In fact the > > difference was small enough that I stopped using memory > > disks and returned to using UFS+SU+noatime (and a few other > > tuning options) on a HDD. > > > > So, if you experience a _dramatic_ speed-up when using SSDs > > for buildworld, I would really very much like to know why > > this is the case, and what kind of tuning you performed. > > Literally: absolutely no tuning. I treat the SSD the exact same as I > would a standard mechanical HDD: install FreeBSD on it, and that's that. > > I imagine the performance difference greatly depends on what SSD you're > actually using. Again, the Intel drives have the highest IOPS out there > (at least in the consumer-grade market), and I can't justify the cost of > an SLC-based drive, so I stick to MLC. > > Possibly you and I have different interpretation of what "dramatic" > means? :-) My opinion is that cutting 6-7 full minutes off a > buildworld is pretty dramatic. Well, here are some hard numbers: With /usr/src and /usr/obj both on the same SATA-300 HDD, a buildworld takes 10 minutes and 13 seconds. With /usr/src and /usr/obj on separate memory disks, a buildworld takes 10 minutes and 4 seconds. This is stable/8 with empty /etc/src.conf, i.e. a complete buildworld. The machine is an AMD Phenom II X6 (hexacore). I think 10:13 vs. 10:04 is not "dramatic". I don't think there's a way I can cut 6-7 minutes from that with current (and affordable) hardware. Best regards Oliver -- Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M. Handelsregister: Registergericht Muenchen, HRA 74606, Geschäftsfuehrung: secnetix Verwaltungsgesellsch. mbH, Handelsregister: Registergericht Mün- chen, HRB 125758, Geschäftsführer: Maik Bachmann, Olaf Erb, Ralf Gebhart FreeBSD-Dienstleistungen, -Produkte und mehr: http://www.secnetix.de/bsd "C is quirky, flawed, and an enormous success." -- Dennis M. Ritchie.