[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [sct@redhat.com: EXT2 - EXT3 - Reiserfs]

Hash: SHA1

On Thursday 14 June 2001 15:58, Stephen C. Tweedie babbled:
> ext3 can do both metadata-only and full data journaling.  You can
> enable data journaling for the entire filesystem or on a per-file
> basis (ext3 uses that functionality internally to journal writes to
> the quota files to keep them consistent, for example.)

 I  learn something new yet again. How would one go about setting the 
journaling mode. On my system, it's currently 'ordered data' mode.

> > However, it is
> > more than sufficient more home use, and most servers (except file/app
> > servers).
> I'm not quite sure what you think would make ext3 unsuitable for
> fileservers!

that's bad grammer on my part. It is definately NOT unsuitable for 
fileservers. What I had intended to get across was that the b-tree algorithm 
in reiserfs might be *better* suited to a very active, heavy I/O fileserver. 
ext3 is certainly usable in that situation, I just wanted to convey that 
reiser might be better. my bad.

> reiserfs only journals metadata at the moment.

again, something new. I was under the impression that is was presently 
capable of data-journalling. again, my bad.

> > reiserfs has a lot of different performance features due to its
> > b-tree heritage.
> Yep.  It's also worth noting that there are a number of advanced
> performance features available as patches to ext2, including indexed
> directories for very fast operation on huge directories, and
> tail-packing for high efficiency on small files.  There are also ACL
> patches available for ext2, and all of these are hoped to be merged
> into ext3 officially at some point.  We also have prototypes working
> for additional ext3-specific functionality such as using a separate
> disk for the journal.

I tried to allude to this fact by stating that the new ext2 features (like 
Daniel's (right?) indexed directories which I eagerly await official 
inclusion) would be 100% plugin-able. Perhaps I should have been more clear. 

> Cheers,
>  Stephen

thanks for jumping in so quickly Stephen! And keep up the excellent hacking 
on the code and the hand-holding help on the ext3 list!
- -- 
Douglas J. Hunley (Linux User #174778)
http://hunley.homeip.net/	http://linux.nf/	

printk("VFS: Busy inodes after unmount. "
        "Self-destruct in 5 seconds.  Have a nice day...\n");
        2.3.99-pre8 /usr/src/linux/fs/super.c
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]