From owner-freebsd-current Tue Nov 12 12:25:11 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA25047 for current-outgoing; Tue, 12 Nov 1996 12:25:11 -0800 (PST) Received: from freefall.freebsd.org (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id MAA25041; Tue, 12 Nov 1996 12:25:09 -0800 (PST) Message-Id: <199611122025.MAA25041@freefall.freebsd.org> To: Mark Mayo cc: Terry Lambert , roberto@keltia.freenix.fr, freebsd-current@FreeBSD.org Subject: Re: ufs is too slow? In-reply-to: Your message of "Tue, 12 Nov 1996 14:12:35 EST." Date: Tue, 12 Nov 1996 12:25:08 -0800 From: "Justin T. Gibbs" Sender: owner-current@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >I just checked, and all they mention in the press_release is that they >will be using the Veritas Volume Manger - I don't kow if this >_necessarily_ implies use of the Veritas File sSystem. The volume manager can work on other FS types too, but according to my friend that works at Veritas, VxFS will be ported to NT. I'd have to check again with him to see if they've struck a deal with MS similar to what they have with other vendors (Sun and HP) where you pay X dollars extra and you get VxFS in the box. >And when you look >at it, all the veritas file system does above and beyond a NTFS is the >ability to 'Live" manage volumes. Please correct me if I'm wrong. They >even talk about the performance increases through the use of 'extents" - >the same terminology is used to decribe NTFS. To me, they really don't >seem that different, and they mention an upgrade procut to be released in >mid 1997 for existing NT 4.0 servers -- all this leads me to believe that >they might be implementing the volume manager on exisiting NTFS >filesystems. The main reason MS is looking to Veritas is that NTFS just doesn't perform when striped or mirrored. The NOW project at Cal did I/O comparison studies on x86 platforms using Solaris, FreeBSD, Linux and NT and found that while FreeBSD and Solaris could achieve upwards of 30MB/s through a striped file system, NT topped out at around 9MB/s. This was NT3.51 - they may have improved things for 4.0. VxFS is quite a bit different then NTFS even though they use similar, extent based, algorithms. VxFS is actually very closely related to XFS if you replace VxFS's "binary buddy maps" with the B+Trees of XFS. (XFS was written by a former Veritas employee that jumped ship to SGI). >All that said, do you think it's worth my time to try and implement NTFS >support under FreBSD? It is most certainly valuable. Any FS type we can support that extends the ability of FreeBSD to interoperate in mixed environments is a big win. You want to write an HPFS filesystem too? 8-) >-Mark > >--------------------------------------------------- >| Mark Mayo mark@quickweb.com | >| RingZero Comp. vinyl.quickweb.com/mark | >--------------------------------------------------- >"To iterate is human, to recurse divine." > - L. Peter Deutsch > > >> know 4.0 dropped HPFS, and *no* upgrade path was given... >> >> >> Terry Lambert >> terry@lambert.org >> --- >> Any opinions in this posting are my own and not those of my present >> or previous employers. >> > -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations ===========================================