[CentOS] mkfs.ext3 on a 9TB volume

Peter Arremann loony at loonybin.org
Sun Sep 11 16:20:51 UTC 2005

On Sunday 11 September 2005 01:26, Francois Caen wrote:
> On 9/10/05, Bryan J. Smith <b.j.smith at ieee.org> wrote:
> > I don't have a list because I personally never make an Ext3 filesystem
> > greater than 1TB, period.  And I try to keep them below 100GBs if I can
> > help it.
> Why?
> Perfomance? Stability???
ext2 and therefore ext3 was designed with a few GB disks in mind... XFS and 
JFS were both designed with multi TB disks in mind... Therefore it _should_ 
work better if you use one of those more modern fs... There are benchmarks 
and all for that - i.e. the performance of reiserfs has been disected over 
and over again when it comes to handling small files. 

Anyway, when running oracle on a cooked space I've not yet seen performance 
differences. If you just have a huge mirror (i.e. all the linux distributions 
and more) I have not noticed performance either... Sure, there are places 
where xfs and the like will work better but I haven't seen an environment 
like that yet. 

All that said, I too would recommend going with Reiser or XFS. I once had a 
ext3 filesystem that had one damaged sector in the journal... of course it 
fell back to ext2 behavior and the fs check took all weekend :-) XFS will 
handle a defective journal much better - so the chance that you ever 
encounter a situation where you have to do a full fs check is much lower.


More information about the CentOS mailing list