Re: [Dump-users] Re: Dump deprecation FAQ

[Stelian P. <st...@po...>]

On Sun, Sep 14, 2003 at 10:13:59PM +0100, Dragan Krnic wrote:

> It is a valid point, Antonios. The old ext2/3 is
> in many respects inferior to reiserfs. 
[...]

Comparisions between reiserfs and ext3 performance are off topic on
this list.

Dragan, I've already told you in the past that your tests were not
very 'professionnal'. You compare file system creation time (which
is irrelevant since you only do it once), file system size (and you
don't say what options you are using), you don't distinguish between
the 3 ext3 mount modes, etc.

I pointed you towards the ext2/3 or linux kernel mailing lists where
people did real tests against those two filesystems, and the results
were more or less equivalent. If you want to discuss your personal
performance measurements, please do it there.

> As conclusion, I think it is time to move on. 
> The good old ext[23] is pretty stable, which means
> it won't develop in any significant way.

ext3 is still in active development, both feature-wise and
performance-wise. 

> On the
> other hand reiserfs is evolving in very many 
> interesting ways, promissing even better performance
> and more features than any other fs in use. For
> example, there will be a reiser-tapefs plug-in, 
[...]

Interesting. That's IMHO the advantage of reiserfs: they are exploring
different algorithms, new ideas, etc. But unfortunately they often tend
to forget stability... 

I never had a single problem with ext3, but the reiserfs history is
filled with security/migration/loss of data/hash conflicts problems.
Yes, I am biased but it has happened to me in the real life.

Stelian.
-- 
Stelian Pop <st...@po...>