From owner-freebsd-hackers Mon Dec 10 18:30:10 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from finch-post-11.mail.demon.net (finch-post-11.mail.demon.net [194.217.242.39]) by hub.freebsd.org (Postfix) with ESMTP id 3F41A37B417 for ; Mon, 10 Dec 2001 18:30:05 -0800 (PST) Received: from ubik.demon.co.uk ([194.222.125.229]) by finch-post-11.mail.demon.net with esmtp (Exim 2.12 #1) id 16DcgF-000BZc-0B for freebsd-hackers@freebsd.org; Tue, 11 Dec 2001 02:30:04 +0000 Message-ID: <7UAeNSA2SWF8Iwe5@ubik.demon.co.uk> Date: Tue, 11 Dec 2001 01:43:18 +0000 To: freebsd-hackers@freebsd.org From: Tony Subject: Re: [SUGGESTION] - JFS for FreeBSD References: <20011210220153.50612.qmail@web21102.mail.yahoo.com> <20011210161410.L92148@elvis.mu.org> <002601c181cb$8c6a5e90$1200a8c0@gsicomp.on.ca> <20011210174711.A3208@mail.slc.edu> <20011210220153.50612.qmail@web21102.mail.yahoo.com> <20011210161410.L92148@elvis.mu.org> <002601c181cb$8c6a5e90$1200a8c0@gsicomp.on.ca> <20011210220153.50612.qmail@web21102.mail.yahoo.com> <20011210174711.A3208@mail.slc.edu> <20011211105617.K63585@monorchid.lemis.com> <20011210195725.A4697@mail.slc.edu> In-Reply-To: <20011210195725.A4697@mail.slc.edu> MIME-Version: 1.0 X-Mailer: Turnpike Integrated Version 5.01 U Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG In article <20011210195725.A4697@mail.slc.edu>, Anthony Schneider writes >> > perhaps there could be an upgrade to offer >> > options SOFTERUPDATES >> > as an equal-but-different alternative to jfs? >> >> And what would that do? > >My thoughts were that if the two were similar in effect that it might be >a relatively easy project to escalate towards achieving the same effects >in one as the other. I understand that this is not necessarily the case. The philosophies in designing the file systems were quite different. At the abstract these are efficiency versus reliability. The philosophy behind a file system not only guides the structure of code it also shapes the structures on disk. A transaction based or journal file system will, for example, appear to grow a database file by whole valid records. If the change is interrupted and the system is restarted the file will either be on the state it was in before the change or in the state afterwards. The usual efficiency oriented file systems can be interrupted at intermediate points: extra disk space is marked as allocated, disk space is assigned to file, file write has happened but the file length has not been updated (or vice versa). Most current users will probably not like the speed penalties of a journal file system, and stick to the faster FS. On the other hand a solid journal FS may encourage more take up for back end databases, for e-commerce, data warehousing, etc... Tony -- "One needs literature in one's life, because without it one deteriorates." - Nelson Mandela To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message