OT: Journaling FS Comparison

OT: Journaling FS Comparison

[Ian Chilton]

Hello,

If this question is already covered somewhere by a web page or article,
then please drop me the URL, but I have not seen such a thing, so was wondering
if someof the developers of the journaling filesystems could comment on the
progress of the systems.

I'd also be interested in knowing what journaling fs the developers around here
use in general...

With there been 4 of them (ext3, reiserfs, XFS and JFS),
it's not an easy choice for anyone.

ext3 stands out because of it's compatibility with ext2 - this makes
it easy to 'upgrade' from ext2 to ext3 without loosing/moving data.
Also it would be much easier to move a drive into another machine
without worrying about the kernel having reiserfs etc compiled in.

However, I have heard ext3 is slower (obviously because it has extra
writes) and sometimes has instibilities.

I also heard that ReiserFS is the fastest out of the bunch, but all
data is lost on converstion, and obviously rescuing and moving disks is
harder. But, it is in the main kernel tree..

Anyway, maybe someone could comment on the different options across the
filesystems, like speed, stable?, nfs?, raid?, convert from ext2?


Also, while I am posting off topic, as part of going to a journaling
fs, I was considering moving my nfs/netboot server from 2.2.19 to 2.4.x
- how's the nfs code working on 2.4, compared to 2.2, especially speed
wise?


Thanks in Advance!


Bye for Now,

Ian


                                  \|||/ 
                                  (o o)
 /-----------------------------ooO-(_)-Ooo----------------------------\
 |  Ian Chilton                    E-Mail: ian@ichilton.co.uk         |
 |  IRC Nick: GadgetMan            Backup: ichilton@www.linux.org.uk  |
 |  ICQ: 16007717 / 104665842      Web   : http://www.ichilton.co.uk  |
 |--------------------------------------------------------------------|
 |       For people who like peace and quiet: a phoneless cord        |
 \--------------------------------------------------------------------/

Re: OT: Journaling FS Comparison

[Steven Cole]

On Sunday 22 July 2001 09:21, Ian Chilton wrote:
> Hello,
>
> If this question is already covered somewhere by a web page or article,
> then please drop me the URL, but I have not seen such a thing, so was

This URL may be of interest: http://aurora.zemris.fer.hr/filesystems/

Steven

Re: OT: Journaling FS Comparison

[Hans Reiser]

Ian Chilton wrote:
 
> I also heard that ReiserFS is the fastest out of the bunch, but all
> data is lost on converstion, and obviously rescuing and moving disks is
> harder. But, it is in the main kernel tree..
tar works....:-)  and it has the advantage that you don't have to worry about a bug in the
conversion program, which was always the thing I feared enough to keep us from writing such a
conversion program.

The last ReiserFS patch for NFS in Linux 2.4 seems to have resulted in no more complaints regarding
nfs and reiserfs used in combination since it went in.  It went in quite recently though.

Hans

Re: OT: Journaling FS Comparison

[Constantin Loizides]

Hi, 

maybe you want also to take a look at
http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/

I am constantly updating this page!

Constantin

Re: OT: Journaling FS Comparison

[Tigran Aivazian]

On Sun, 22 Jul 2001, Ian Chilton wrote:
> With there been 4 of them (ext3, reiserfs, XFS and JFS),
> it's not an easy choice for anyone.

at the time when I did the comparison using SPEC SFS to benchmark, the
choice was not hard at all -- absolute and obvious winner was reiserfs.
That is, amongst the freely available ones. (this was not too long ago, a
mere 2 months or so).

However, if you are willing to pay for your filesystem, our vxfs beats all
of the above at _very_ (very) high loads (loads unreachable by any other
filesystem so far ;) in both performance and stability. (well, it beats
them in most situations at low loads as well but that is not interesting)

It should be available to our beta-customers via www.veritas.com
somewhere...

Regards,
Tigran

Re: OT: Journaling FS Comparison

[Hans Reiser]

Tigran Aivazian wrote:
> 
> On Sun, 22 Jul 2001, Ian Chilton wrote:
> > With there been 4 of them (ext3, reiserfs, XFS and JFS),
> > it's not an easy choice for anyone.
> 
> at the time when I did the comparison using SPEC SFS to benchmark, the
> choice was not hard at all -- absolute and obvious winner was reiserfs.
> That is, amongst the freely available ones. (this was not too long ago, a
> mere 2 months or so).
> 
> However, if you are willing to pay for your filesystem, our vxfs beats all
> of the above at _very_ (very) high loads (loads unreachable by any other
> filesystem so far ;) in both performance and stability. (well, it beats
> them in most situations at low loads as well but that is not interesting)
> 
> It should be available to our beta-customers via www.veritas.com
> somewhere...
> 
> Regards,
> Tigran
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

SPEC SFS is a proprietary and expensive benchmark which precludes us from optimizing for it, which
is a pity, I suspect we'd learn something from analyzing its results.

How much does vxfs cost these days?

Hans

Re: OT: Journaling FS Comparison

[Tigran Aivazian]

On Mon, 23 Jul 2001, Hans Reiser wrote:
> How much does vxfs cost these days?

Sorry, I do not know, simply because it is not like a "pound of apples"
which one can buy in a supermarket, i.e. there are complex things like
"bundles" and other entities which may contain vxfs (as well as other
VERITAS products) which one may have to buy to get it.

I retained cc:linux-kernel so that people don't think there is something
secretive going on, but the discussion of proprietary products' pricing is
most likely offtopic... :)

Regards,
Tigran

Re: OT: Journaling FS Comparison

[Pavel Machek]

Hi!

> ext3 stands out because of it's compatibility with ext2 - this makes
> it easy to 'upgrade' from ext2 to ext3 without loosing/moving data.
> Also it would be much easier to move a drive into another machine
> without worrying about the kernel having reiserfs etc compiled in.

Plus, ext2 has *very* good fsck, tested by years.

Last time I ran reiserfsck, it found some errors, but warned me *against*
trying to fix them. I have SuSE7.2 system on reiserfs, and fsck complains
slighlty. System seems to work well. How do I fix it up? Is reiserfsck
really so dangerous as it cllaims?
-- 
Philips Velo 1: 1"x4"x8", 300gram, 60, 12MB, 40bogomips, linux, mutt,
details at http://atrey.karlin.mff.cuni.cz/~pavel/velo/index.html.

Re: OT: Journaling FS Comparison

[Hans Reiser]

Martin Knoblauch wrote:
> 
> > Re: OT: Journaling FS Comparison
> >
> > From: Hans Reiser (reiser@namesys.com)
> > Date: Sun Jul 22 2001 - 13:59:42 EST
> >
> >
> > The last ReiserFS patch for NFS in Linux 2.4 seems to have resulted in no more complaints regarding
> > nfs and reiserfs used in combination since it went in. It went in quite recently though.
> >
> > Hans
> 
>  Actually: what is the status if integration of the various ReiserFS
> patches in the mainstream or AC kernels. e.g. the "unmount" patch does
> not seem to be incorporated in 2.4.[5-7].
> 
> Thanks
> Martin
> --
> ------------------------------------------------------------------
> Martin Knoblauch         |    email:  Martin.Knoblauch@TeraPort.de
> TeraPort GmbH            |    Phone:  +49-89-510857-309
> C+ITS                    |    Fax:    +49-89-510857-111
> http://www.teraport.de   |    Mobile: +49-170-4904759
Its functional substitute is in 2.4.6

Re: OT: Journaling FS Comparison

[Martin Knoblauch]

> Re: OT: Journaling FS Comparison
> 
> From: Hans Reiser (reiser@namesys.com)
> Date: Sun Jul 22 2001 - 13:59:42 EST
> 
> 
> The last ReiserFS patch for NFS in Linux 2.4 seems to have resulted in no more complaints regarding
> nfs and reiserfs used in combination since it went in. It went in quite recently though.
> 
> Hans

 Actually: what is the status if integration of the various ReiserFS
patches in the mainstream or AC kernels. e.g. the "unmount" patch does
not seem to be incorporated in 2.4.[5-7].

Thanks
Martin
-- 
------------------------------------------------------------------
Martin Knoblauch         |    email:  Martin.Knoblauch@TeraPort.de
TeraPort GmbH            |    Phone:  +49-89-510857-309
C+ITS                    |    Fax:    +49-89-510857-111
http://www.teraport.de   |    Mobile: +49-170-4904759