On Fri, Jul 13, 2001 at 09:20:59PM +1000, Mike Gigante wrote:
>
> XFS is pretty tuneable - so you could play with inode size, log size
> number of allocation groups etc to best suit your usage.
>
> Such tuning doesn't seem possible with ReiserFS ...
Reiserfs doesn't have inodes or allocation groups, so naturally you
can't change them. :-)
However, it does use a pretty modular system for some datastructures -
eg the stat-data and hashes. This means you can add a different hash og
a different kind of stat-data (e.g. maybe you don't need timestamps for
your files) rellatively easy. (yes, it requires changing kernel code,
but is still rellatively easy).
I believe you now can change the logsize though (maybe you still need a
patch).
> The mongo benchmarks use relatively small files. XFS scales very well to
> large files/directories, but that benchmark sticks to small scale stuff.
> It *seems* to show ReiserFS in the best possible light and the XFS results
> don't seem to reflect anything like real world experience (except that it
> correctly highlights the relatively slow delete performance of XFS!)
I partly disagree. mongo tests filesystem performace - benchmarks that
test writing really big files is mostly testing your hardware. The
filesystem still matters, but not so much as on small files. I think
the difference between ext2/xfs/reiserfs is less than 10%. (ext3 with
datalogging may be a different story though).
--
Ragnar Kjorstad
Big Storage
|