From owner-freebsd-current Mon Nov 11 09:06:17 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id JAA26014 for current-outgoing; Mon, 11 Nov 1996 09:06:17 -0800 (PST) Received: from smily.3skel.com (3skel.com [206.138.212.40]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id JAA26004 for ; Mon, 11 Nov 1996 09:06:11 -0800 (PST) Received: from fnur.3skel.com (root@fnur.3skel.com [192.168.0.8]) by smily.3skel.com (8.8.2/8.8.2) with ESMTP id MAA06305; Mon, 11 Nov 1996 12:04:40 -0500 (EST) Received: (from danj@localhost) by fnur.3skel.com (8.8.2/8.8.2) id MAA12463; Mon, 11 Nov 1996 12:04:35 -0500 (EST) Date: Mon, 11 Nov 1996 12:04:35 -0500 (EST) From: Dan Janowski Message-Id: <199611111704.MAA12463@fnur.3skel.com> To: bsdcur@shadows.aeon.net, jgreco@brasil.moneng.mei.com Subject: Re: ufs is too slow? Cc: freebsd-current@freebsd.org Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > my friend told me that the ufs filesystem is too slow for a high end > > nntp server usage... > > > > the average 7 articles per second is supposed to be too much, and that > > using some log type filesystem, like xfs (in sgi) would be better... > > > > he was saying that ufs cant create those 7 files in the second. > > > > is that so? > > That was, however, on a machine heavily optimized towards the task. > We were flyyyyyying.... > > FFS is in general a great FS... but it is optimized for the general case. > > It is particularly poor at handling large directories or lots of writes, > and news is excellent at pounding on these qualities. > > (see my other note this morning, or I can cc: you a copy). me, please. > ... JG At one point I had inquired about lfs (log file system), in part because of my experience with xfs (SGI's). Although lfs is not xfs, they are both better performers than ufs/ffs (which are both REALLY old, I think ufs dates from the 50's and ffs from the 70's). I have seen that OpenBSD is doing something with lfs, but I am not sure what. It would be worth while to get lfs running for sure; if you ever wondered, a lot of that disk bandwith goes to filesystem overhead. I once exchanged some e-mail with someone at BSDI and with Margo Seltzer, who was a principle for lfs. The apparent primary reason why lfs does not run here is that lfs does some wierd stuff with the ATT buffer code that is missing in 4.4-lite. I was not able to get a synopsis of what or how to get around it, but it didn't sound like lfs was broken, it's just missing some wheels. Maybe we can all talk about it a little and figure out how hard it would be to get going. If we were running a kick-ass big/fast file system, FreeBSD would capture some more of the Int(er|tra)+Net market. In addition to which, the infamous 'make world' time would surely benefit. Dan -- danj@3skel.com Triskelion Systems, Inc. New York