From owner-freebsd-chat Sat Jan 26 8:47:38 2002 Delivered-To: freebsd-chat@freebsd.org Received: from riker.skynet.be (riker.skynet.be [195.238.3.132]) by hub.freebsd.org (Postfix) with ESMTP id 61CE637B400 for ; Sat, 26 Jan 2002 08:47:34 -0800 (PST) Received: from [10.0.1.14] (ip-27.shub-internet.org [194.78.144.27] (may be forged)) by riker.skynet.be (8.11.6/8.11.6/Skynet-OUT-2.16) with ESMTP id g0QGlA515776; Sat, 26 Jan 2002 17:47:10 +0100 (MET) (envelope-from ) Mime-Version: 1.0 X-Sender: bs663385@pop.skynet.be Message-Id: In-Reply-To: <20020125212742.C75216@over-yonder.net> References: <20020123124025.A60889@HAL9000.wox.org> <3C4F5BEE.294FDCF5@mindspring.com> <20020123223104.SM01952@there> <15440.35155.637495.417404@guru.mired.org> <15440.53202.747536.126815@guru.mired.org> <15441.17382.77737.291074@guru.mired.org> <20020125212742.C75216@over-yonder.net> Date: Sat, 26 Jan 2002 17:35:52 +0100 To: "Matthew D. Fuller" , Brad Knowles From: Brad Knowles Subject: Re: Bad disk partitioning policies (was: "Re: FreeBSD Intaller (was "Re: ... RedHat ...")") Cc: Mike Meyer , chip , freebsd-chat@freebsd.org Content-Type: text/plain; charset="us-ascii" ; format="flowed" Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org At 9:27 PM -0600 2002/01/25, Matthew D. Fuller wrote: >> Well, since user-level programs can't write into the reserved >> disk space allocation (used to be ~10%, but I don't think that this >> really makes much sense with modern 100GB+ high-capacity disks, > > Why doesn't it make sense? > > The space allocation isn't intended to 'keep user-level program from > filling up the disk', While that is not the primary goal, that is one very useful side-effect. > it's intended to allow the fragmentation-avoidance > to work. That is the primary goal, yes. > Size doesn't matter; percentage does. I've heard somewhere > (from Terry, I think) that 15% is the 'optimal' setting for this, and > 10% was a compromise that wasn't too far below optimal, but gave that 5% > of extra available space. 8% is the current default in newfs(8). I disagree. Size does matter. The fragmentation-avoidance algorithms should still work at the sector/block/cylinder level, but the total disk space available is now many, many, many, many orders of magnitude larger than when these algorithms were first created. On modern high-capacity disks, 1% should be way more than you could ever need, in terms of what is required by the fragmentation-avoidance algorithms. Now, there may be other reasons why you might want to allocate more than 1% to this reserved disk space, including the reasons I've previously mentioned. -- Brad Knowles, H4sICIFgXzsCA2RtYS1zaWcAPVHLbsMwDDvXX0H0kkvbfxiwVw8FCmzAzqqj1F4dy7CdBfn7 Kc6wmyGRFEnvvxiWQoCvqI7RSWTcfGXQNqCUAnfIU+AT8OZ/GCNjRVlH0bKpguJkxiITZqes MxwpSucyDJzXxQEUe/ihgXqJXUXwD9ajB6NHonLmNrUSK9nacHQnH097szO74xFXqtlbT3il wMsBz5cnfCR5cEmci0Rj9u/jqBbPeES1I4PeFBXPUIT1XDSOuutFXylzrQvGyboWstCoQZyP dxX4dLx0eauFe1x9puhoi0Ao1omEJo+BZ6XLVNaVpWiKekxN0VK2VMpmAy+Bk7ZV4SO+p1L/ uErNRS/qH2iFU+iNOtbcmVt9N16lfF7tLv9FXNj8AiyNcOi1AQAA To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message