From owner-freebsd-fs Mon Jun 10 17:25:50 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id RAA28471 for fs-outgoing; Mon, 10 Jun 1996 17:25:50 -0700 (PDT) Received: from palmer.demon.co.uk (palmer.demon.co.uk [158.152.50.150]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id RAA28432 for ; Mon, 10 Jun 1996 17:25:33 -0700 (PDT) Received: from palmer.demon.co.uk (localhost [127.0.0.1]) by palmer.demon.co.uk (sendmail/PALMER-1) with ESMTP id BAA23823 for ; Tue, 11 Jun 1996 01:24:57 +0100 (BST) Prev-Resent: Tue, 11 Jun 1996 01:24:56 +0100 Prev-Resent: "fs@freebsd.org " To: "Hr.Ladavac" cc: fs@FreeBSD.ORG, james@jraynard.demon.co.uk From: "Gary Palmer" Subject: Re: Breaking ffs - speed enhancement? In-reply-to: Your message of "Mon, 10 Jun 1996 16:12:07 +0200." <199606101412.AA062675927@ws2301.gud.siemens.co.at> Date: Mon, 10 Jun 1996 19:38:06 +0100 Message-ID: <22695.834431886@palmer.demon.co.uk> Resent-To: fs@FreeBSD.ORG Resent-Date: Tue, 11 Jun 1996 01:24:57 +0100 Resent-Message-ID: <23821.834452697@palmer.demon.co.uk> Resent-From: Gary Palmer Sender: owner-fs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk "Hr.Ladavac" wrote in message ID <199606101412.AA062675927@ws2301.gud.siemens.co.at>: > Well, all I've seen is what I've seen in the manpage and sources. I did > not actually test it, but I think one needs the ability to add more than > /p/ip and /p/unix (whatever they are named.) I'd like to use it as a > clone of Atria MVFS and ClearAudit. Naturally, my portal daemon will have > to do that, somehow. MVFS? Sorry, not familiar with that. My /etc/portal.conf has: tcplisten/ tcplisten tcplisten/ tcp/ tcp tcp/ fs/ file fs/ pipe/ pipe foo/ exec ./bar bar baz the first 3 are known to work (at least, they USED to when I was testing/developing portals), the last 2 don't. (or at least the pipes doesn't work, haven't been able to contrive a test for the exec case). I'm not even 100% sure what the author was trying to do with the ``pipe'' implimentation. The IDEAL case would allow you to plug in your own modules to extend the filesystem, but to do that without opening up MASSIVE security holes would be ``interesting'' to say the least. As it is the ``exec'' case may be left unimplimented (if it isn't already) for that very reason. > Since I did not test it, do the portals work, and from which OS version > onwards? Portals have worked to a greater or lesser extent since (at least) 2.0.5-RELEASE, probably also (to a lesser extent) in 2.0-RELEASE. Nothing has changed in the portals system for a while now as I haven't had the time of late (although I should try to get back to it sometime, or help someone else who has the time) > Shall we move there, then? Done Gary -- Gary Palmer FreeBSD Core Team Member FreeBSD: Turning PC's into workstations. See http://www.FreeBSD.ORG/ for info From owner-freebsd-fs Mon Jun 10 20:10:56 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA15434 for fs-outgoing; Mon, 10 Jun 1996 20:10:56 -0700 (PDT) Received: (from jmb@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA15418; Mon, 10 Jun 1996 20:10:53 -0700 (PDT) Date: Mon, 10 Jun 1996 20:10:53 -0700 (PDT) From: "Jonathan M. Bresler" Message-Id: <199606110310.UAA15418@freefall.freebsd.org> To: fs, hackers Subject: LFS papers available Sender: owner-fs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk the LFS papers that i listed earlier are available in freefall.freebsd.org:/home/jmb/Papers. i wont be able to put them up for ftp till i get my regular net access back (damn terminal server leaks and locked... needs to be rebooted.....grrrr) if someone would be willing to put them up for ftp i can even mail them to you ;) From owner-freebsd-fs Mon Jun 10 22:30:42 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id WAA25092 for fs-outgoing; Mon, 10 Jun 1996 22:30:42 -0700 (PDT) Received: from eddie.cis.uoguelph.ca (eddie.cis.uoguelph.ca [131.104.48.34]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id WAA25074; Mon, 10 Jun 1996 22:30:37 -0700 (PDT) Received: (from josh@localhost) by eddie.cis.uoguelph.ca (8.7.5/8.7.3) id BAA02766; Tue, 11 Jun 1996 01:27:39 -0400 (EDT) From: Josh Tiefenbach Message-Id: <199606110527.BAA02766@eddie.cis.uoguelph.ca> Subject: Re: LFS papers available To: jmb@freefall.freebsd.org (Jonathan M. Bresler) Date: Tue, 11 Jun 1996 01:27:39 -0400 (EDT) Cc: fs@freefall.freebsd.org, hackers@freefall.freebsd.org In-Reply-To: <199606110310.UAA15418@freefall.freebsd.org> from "Jonathan M. Bresler" at "Jun 10, 96 08:10:53 pm" X-Mailer: ELM [version 2.4ME+ PL15 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-fs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > the LFS papers that i listed earlier are available in > freefall.freebsd.org:/home/jmb/Papers. Those papers are now avail for ftp at: ftp://ftp.scis.org/pub/lfs/ More to follow, as I dig them up josh -- Josh Tiefenbach | "No, I'm not going to explain it. If you President, | can't figure it out, you didnt want to Society for Computing | know anyways..." -- Larry Wall and Information Science. | Email: josh@scis.org Web: http://www.scis.org/~josh/ From owner-freebsd-fs Tue Jun 11 08:43:50 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA08541 for fs-outgoing; Tue, 11 Jun 1996 08:43:50 -0700 (PDT) Received: from eddie.cis.uoguelph.ca (eddie.cis.uoguelph.ca [131.104.48.34]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id IAA08533; Tue, 11 Jun 1996 08:43:45 -0700 (PDT) Received: (from josh@localhost) by eddie.cis.uoguelph.ca (8.7.5/8.7.3) id LAA03533; Tue, 11 Jun 1996 11:40:56 -0400 (EDT) From: Josh Tiefenbach Message-Id: <199606111540.LAA03533@eddie.cis.uoguelph.ca> Subject: Re: Breaking ffs - speed enhancement? To: gpalmer@freebsd.org (Gary Palmer) Date: Tue, 11 Jun 1996 11:40:56 -0400 (EDT) Cc: fs@freebsd.org In-Reply-To: <22695.834431886@palmer.demon.co.uk> from Gary Palmer at "Jun 10, 96 07:38:06 pm" X-Mailer: ELM [version 2.4ME+ PL15 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > MVFS? Sorry, not familiar with that. MultiVersion FS. Used in Atria's Clearcase (SCM) tool to implement their method of version control. It allows neat things like specify which version of the file you want. eg: /usr/src/foo.c (The version of foo.c selected based on some config spec) /usr/src/foo.c@@ (The actual element) /usr/src/foo.c@@/TESTING (the version of foo.c labeled TESTING) Turning back to LFS, I'd like to throw my hat into the ring (as soon as I finish reading the papesr of course :) josh -- Josh Tiefenbach | "No, I'm not going to explain it. If you President, | can't figure it out, you didnt want to Society for Computing | know anyways..." -- Larry Wall and Information Science. | Email: josh@scis.org Web: http://www.scis.org/~josh/ From owner-freebsd-fs Tue Jun 11 10:20:56 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id KAA13245 for fs-outgoing; Tue, 11 Jun 1996 10:20:56 -0700 (PDT) Received: from eac.iafrica.com (slipper101145.iafrica.com [196.7.101.145]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id KAA13214 for ; Tue, 11 Jun 1996 10:20:36 -0700 (PDT) Received: (from rnordier@localhost) by eac.iafrica.com (8.6.12/8.6.12) id TAA01192; Tue, 11 Jun 1996 19:10:04 +0200 From: Robert Nordier Message-Id: <199606111710.TAA01192@eac.iafrica.com> Subject: Re: bit 7 in filenames To: khetan@iafrica.com (Khetan Gajjar) Date: Tue, 11 Jun 1996 19:10:03 +0200 (SAT) Cc: fs@freebsd.org In-Reply-To: from "Khetan Gajjar" at Jun 11, 96 00:15:07 am X-Mailer: ELM [version 2.4 PL24 ME8a] Content-Type: text Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk (Moved from -hackers to -fs) Khetan Gajjar wrote: > > On Mon, 10 Jun 1996, Robert Nordier wrote: > > >No, the vfatfs is probably about 80 percent complete, but still in > >pieces, and not expected in -current for a while. > > Ok. > > >As the msdosfs code has changed comparatively little, the msdosfs > >fixes will probably apply to most recent versions of FreeBSD. > >These patches should be out in a few weeks and will address the > >problem of corrupting BSD partitions. > > What about dos partitions ? ;-) I don't think I've any records relating to the msdosfs corrupting its own partition (but I'm still working through all the msdosfs-related stuff I've collected over several months). Anyway, the aim of the patches will be to remedy those reproducible serious problems that won't require fundamental design changes. -- Robert Nordier From owner-freebsd-fs Tue Jun 11 20:33:13 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA22196 for fs-outgoing; Tue, 11 Jun 1996 20:33:13 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id UAA22191; Tue, 11 Jun 1996 20:33:08 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.5/8.6.9) id UAA04275; Tue, 11 Jun 1996 20:33:05 -0700 (PDT) Date: Tue, 11 Jun 1996 20:33:05 -0700 (PDT) Message-Id: <199606120333.UAA04275@silvia.HIP.Berkeley.EDU> To: fs@freebsd.org, hackers@freebsd.org Reply-to: fs@freebsd.org In-reply-to: <199606111540.LAA03533@eddie.cis.uoguelph.ca> (message from Josh Tiefenbach on Tue, 11 Jun 1996 11:40:56 -0400 (EDT)) Subject: Re: Breaking ffs - speed enhancement? From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Speaking of filesystems, does anyone know about works done on very large (disk) filesystems? By "large", I'm talking in the range of 100GB - 5TB. I know that FFS will have some problems once we go over about 200GB (multi-hour fsck being one). But all we can find are studies on tape-based systems (how to migrate stuff to/from disk, etc.)... Thanks Satoshi From owner-freebsd-fs Tue Jun 11 21:27:15 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id VAA25466 for fs-outgoing; Tue, 11 Jun 1996 21:27:15 -0700 (PDT) Received: from neurocog.lrdc.pitt.edu (hahn@neurocog.lrdc.pitt.edu [136.142.95.44]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id VAA25458 for ; Tue, 11 Jun 1996 21:27:13 -0700 (PDT) Message-Id: <199606120427.VAA25458@freefall.freebsd.org> Received: by neurocog.lrdc.pitt.edu (1.37.109.10G/16.2) id AA208713631; Wed, 12 Jun 1996 00:27:11 -0400 From: Mark Hahn Subject: Re: Breaking ffs - speed enhancement? To: fs@freebsd.org Date: Wed, 12 Jun 1996 00:27:11 -0400 (EDT) In-Reply-To: <199606120333.UAA04275@silvia.HIP.Berkeley.EDU> from "Satoshi Asami" at Jun 11, 96 08:33:05 pm X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Speaking of filesystems, does anyone know about works done on very > large (disk) filesystems? By "large", I'm talking in the range of > 100GB - 5TB. large fs's are usually populated by larger-than-normal files. > > I know that FFS will have some problems once we go over about 200GB > (multi-hour fsck being one). But all we can find are studies on > tape-based systems (how to migrate stuff to/from disk, etc.)... log-structuring is the obvious response: simply never have to do fsck, or at least nothing proportional to the total size, rather than proportional to the amount of activity... regards, mark hahn. -- operator may differ from spokesperson. hahn@neurocog.lrdc.pitt.edu http://neurocog.lrdc.pitt.edu/~hahn/ From owner-freebsd-fs Wed Jun 12 02:31:32 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id CAA11696 for fs-outgoing; Wed, 12 Jun 1996 02:31:32 -0700 (PDT) Received: from proxy.siemens.at (proxy.siemens.at [192.138.228.19]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id CAA11676; Wed, 12 Jun 1996 02:31:10 -0700 (PDT) Received: from sol1.gud.siemens.co.at (sol-f.gud.siemens-austria) by proxy.siemens.at with SMTP id AA02060 (5.67a/IDA-1.5); Wed, 12 Jun 1996 11:30:22 +0200 Received: from ws2301.gud.siemens.co.at by sol1.gud.siemens.co.at with smtp (Smail3.1.28.1 #7 for ) id m0uTmFm-00020DC; Wed, 12 Jun 96 11:30 MET DST Received: by ws2301.gud.siemens.co.at (1.37.109.16/1.37) id AA231331802; Wed, 12 Jun 1996 11:30:02 +0200 From: "Hr.Ladavac" Message-Id: <199606120930.AA231331802@ws2301.gud.siemens.co.at> Subject: Re: Breaking ffs - speed enhancement? To: gpalmer@freebsd.org (Gary Palmer) Date: Wed, 12 Jun 1996 11:30:01 +0200 (MESZ) Cc: freebsd-fs@freebsd.org In-Reply-To: <22695.834431886@palmer.demon.co.uk> from "Gary Palmer" at Jun 10, 96 07:38:06 pm X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In his e-mail Gary Palmer wrote: > > Portals have worked to a greater or lesser extent since (at least) > 2.0.5-RELEASE, probably also (to a lesser extent) in > 2.0-RELEASE. Nothing has changed in the portals system for a while now > as I haven't had the time of late (although I should try to get back > to it sometime, or help someone else who has the time) > > Gary I don't know if I have understood the docs correctly, but it woild seem that portalfs passes only the opened fd's to the client. I don't know whether it supports stat(2) and friends needed for directory lookups. If yes, great. If not, ... Because, portalfs is an ideal place to put the SMB filesystem in. That way smb-portald can take care of security problems: it knows what a uid does its client have, and can log him in the net. That way not everyone gets to be logged into the net (which would be the case if smb-fs were mounted any other way.) What do you think? /Marino From owner-freebsd-fs Wed Jun 12 13:34:18 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id NAA00628 for fs-outgoing; Wed, 12 Jun 1996 13:34:18 -0700 (PDT) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id NAA00595; Wed, 12 Jun 1996 13:34:10 -0700 (PDT) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA06704; Wed, 12 Jun 1996 13:33:29 -0700 From: Terry Lambert Message-Id: <199606122033.NAA06704@phaeton.artisoft.com> Subject: Re: Breaking ffs - speed enhancement? To: fs@FreeBSD.ORG Date: Wed, 12 Jun 1996 13:33:29 -0700 (MST) Cc: hackers@FreeBSD.ORG In-Reply-To: <199606120333.UAA04275@silvia.HIP.Berkeley.EDU> from "Satoshi Asami" at Jun 11, 96 08:33:05 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-fs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Speaking of filesystems, does anyone know about works done on very > large (disk) filesystems? By "large", I'm talking in the range of > 100GB - 5TB. > > I know that FFS will have some problems once we go over about 200GB > (multi-hour fsck being one). But all we can find are studies on > tape-based systems (how to migrate stuff to/from disk, etc.)... Contact Matt Day about soft updates, which are running under Windows 95 in the cubilcle next to mine (mday@elbereth.org). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-fs Fri Jun 14 11:33:32 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id LAA05730 for fs-outgoing; Fri, 14 Jun 1996 11:33:32 -0700 (PDT) Received: from mail.barrnet.net (mail.barrnet.net [131.119.246.7]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id LAA05725 for ; Fri, 14 Jun 1996 11:33:30 -0700 (PDT) Received: from p133.i-node.at (p133.i-node.at [194.152.162.81]) by mail.barrnet.net (8.7.5/MAIL-RELAY-LEN) with SMTP id LAA22149 for ; Fri, 14 Jun 1996 11:33:24 -0700 (PDT) Received: from schubert (inode2.i-node.at [194.152.162.85]) by p133.i-node.at (8.6.12/8.6.9) with SMTP id UAA26910; Fri, 14 Jun 1996 20:33:09 +0100 Message-ID: <31C1AFAE.30B1@i-node.at> Date: Fri, 14 Jun 1996 20:30:06 +0200 From: versuspc Organization: Versus PC Systeme X-Mailer: Mozilla 2.02 (Win95; I) MIME-Version: 1.0 To: versuspc@i-node.at CC: freebsd-fs@freebsd.org Subject: Extent like FFS Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi all, after reading Larry McVoy´s paper, one cannot help but wonder if the write clustering has been incorporated in the FFS as well. At least according to the paper it should not be too much work, and the on-disk representation of the FFS does not change at all. They present up to 100% improvements in performance with only a slight degradation (~10%) on random-access updates. /Marino From owner-freebsd-fs Fri Jun 14 11:56:40 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id LAA07060 for fs-outgoing; Fri, 14 Jun 1996 11:56:40 -0700 (PDT) Received: from mail.barrnet.net (mail.barrnet.net [131.119.246.7]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id LAA07055 for ; Fri, 14 Jun 1996 11:56:39 -0700 (PDT) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by mail.barrnet.net (8.7.5/MAIL-RELAY-LEN) with SMTP id LAA22503 for ; Fri, 14 Jun 1996 11:56:37 -0700 (PDT) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id MAA22361; Fri, 14 Jun 1996 12:52:05 -0600 Date: Fri, 14 Jun 1996 12:52:05 -0600 From: Nate Williams Message-Id: <199606141852.MAA22361@rocky.sri.MT.net> To: versuspc Cc: freebsd-fs@freebsd.org Subject: Re: Extent like FFS In-Reply-To: <31C1AFAE.30B1@i-node.at> References: <31C1AFAE.30B1@i-node.at> Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > after reading Larry McVoy´s paper, one cannot help but wonder > if the write clustering has been incorporated in the FFS as well. It was incorporated into 4.4BSD, which means it's in FreeBSD which is based on 4.4BSD. Nate From owner-freebsd-fs Fri Jun 14 13:18:06 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id NAA13571 for fs-outgoing; Fri, 14 Jun 1996 13:18:06 -0700 (PDT) Received: from ns.via.net (ns.via.net [140.174.204.1]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id NAA13566 for ; Fri, 14 Jun 1996 13:18:02 -0700 (PDT) Received: (from joe@localhost) by ns.via.net (8.6.9/8.6.9) id NAA20592; Fri, 14 Jun 1996 13:16:08 -0700 Date: Fri, 14 Jun 1996 13:16:08 -0700 From: Joe McGuckin Message-Id: <199606142016.NAA20592@ns.via.net> To: nate@sri.MT.net Subject: Re: Extent like FFS Cc: freebsd-fs@freebsd.org Sender: owner-fs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Why not incorporate some ideas from Sprite? One feature that seemed to speed things up in general was the fact that no metadata gets written out for a newly created file until it has aged 30 seconds. It seems to me this could be a win in a couple of ways: - Small, tmp files exist only in the buffer cache & never get written to disk if they don't live more that 30 seconds. - There out to be a way to queue all the newly created files together then write them out together at once after they've lived 'X' amount of time. Where can I find an indepth explanation of how the FFS file system code works? -joe From owner-freebsd-fs Fri Jun 14 13:56:18 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id NAA16713 for fs-outgoing; Fri, 14 Jun 1996 13:56:18 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id NAA16688 for ; Fri, 14 Jun 1996 13:56:13 -0700 (PDT) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by who.cdrom.com (8.6.12/8.6.11) with ESMTP id NAA10527 for ; Fri, 14 Jun 1996 13:20:43 -0700 Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id OAA22670; Fri, 14 Jun 1996 14:20:35 -0600 Date: Fri, 14 Jun 1996 14:20:35 -0600 From: Nate Williams Message-Id: <199606142020.OAA22670@rocky.sri.MT.net> To: Joe McGuckin Cc: nate@sri.MT.net, freebsd-fs@FreeBSD.ORG Subject: Re: Extent like FFS In-Reply-To: <199606142016.NAA20592@ns.via.net> References: <199606142016.NAA20592@ns.via.net> Sender: owner-fs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Why not incorporate some ideas from Sprite? One feature that seemed to > speed things up in general was the fact that no metadata gets written out > for a newly created file until it has aged 30 seconds. I think this is a lose as far as POSIX goes, but the FS gurus can thrash this out. > It seems to me this could be a win in a couple of ways: > > - Small, tmp files exist only in the buffer cache & never get > written to disk if they don't live more that 30 seconds. If you are worried about files in /tmp, use MFS. :) > - There out to be a way to queue all the newly created files together > then write them out together at once after they've lived 'X' amount > of time. And when your system power goes away in the middle of writing out these files and they happen to be pretty important what happens? > Where can I find an indepth explanation of how the FFS file system code works? There are papers in the system, plus the 4.4BSD book apparently goes into detail on it. Nate