l i n u x - u s e r s - g r o u p - o f - d a v i s
Next Meeting:
July 7: Social gathering
Next Installfest:
Latest News:
Jun. 14: June LUGOD meeting cancelled
Page last updated:
2006 Mar 29 22:16

The following is an archive of a post made to our 'vox-tech mailing list' by one of its subscribers.

Report this post as spam:

(Enter your email address)
Re: [vox-tech] Reiserfs v. Ext3
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [vox-tech] Reiserfs v. Ext3

On Mon, Mar 27, 2006 at 07:50:45AM -0800, Bob Scofield wrote:
> Within the next 2 or 3 weeks I plan to do a Debian install in my SuSE 
> partition.  Is there any reason why I would choose Reiserfs over Ext3, or 
> vice versa?

Er, that would depend on what you want in your filesystem.

There used to be clear differences, things like journaled vs not, large
dir performance.  Alas, the Linux filesystems I know of are getting more
similar over time.

Now the differences are more obscure, things like packing efficiency
for small files, memory overhead, and performance under ever more
particular situations.  Even in those situations performance often
depends on specific variables/flags used to create the filesystem.

BTW, I would look for the impartial benchmark comparisons, not the ones
that advocates of any side scream the loudest about.

> I've used both in the past, and have never had problems with either (at least 
> that I know about).

If reliability is your #1 concern I would use whatever is most popular
on your Linux distribution.  While load testing fileservers I didn't find
much difference in the filesystem except for XFS which had substantially
better bandwidth to a 15 disk array I was testing.

It does concern me that Reiserfs is not in the main line kernel, so
I would imagine that it receives a bit less review and testing when
compared to the filesystems that are in the kernels.

Additionally when people say "We rewrote almost everything" that makes me
want to back away for awhile.  There are so many variables that can take
awhile to track down.  SMP, ACPI, variable clock speeds, NUMA, x86-64,
IA32, bigmem, a zillion different controller chips, kernel versions,
raid controllers, linux software RAID, EVMS, etc.  Not to mention the
numerous different failure modes of different hardware, etc.  Things that
should never happy seem to have a way of happening.

Then again probably 99% of the linux kernels in the world are from a
linux distribution, as opposed to a self built kernel from kernel.org.

In any case it's one of my goals not to be the first one to find some
filesystem killing bug in my specific cpu, motherboard, controller,
driver, and disk combination.  

Pioneers are the ones that get arrows in their back, and my goal
is not to be one.

In short, I know too late, unless you have specific needs and know a
particular file system is good at that workload, pick the one that is
default for your distribution.

Bill Broadley
Computational Science and Engineering
UC Davis
vox-tech mailing list

LUGOD Group on LinkedIn
Sign up for LUGOD event announcements
Your email address:
LUGOD Group on Facebook
'Like' LUGOD on Facebook:

Hosting provided by:
Sunset Systems
Sunset Systems offers preconfigured Linux systems, remote system administration and custom software development.

LUGOD: Linux Users' Group of Davis
PO Box 2082, Davis, CA 95617
Contact Us

LUGOD is a 501(c)7 non-profit organization
based in Davis, California
and serving the Sacramento area.
"Linux" is a trademark of Linus Torvalds.

Sponsored in part by:
EDGE Tech Corp.
For donating some give-aways for our meetings.