ext3 performance with hardware RAID5

Andreas Dilger adilger at clusterfs.com
Wed Jul 14 21:00:54 UTC 2004

On Jul 14, 2004  15:33 -0500, Steve Hanson wrote:
> Andreas Dilger wrote:
> >You didn't specify the size of the external journal device.  In our
> >testing we found that having a larger journal (internal) made a large
> >difference in performance.  I believe that the external journal will
> >be the full size of the device (could be wrong of course) so it is
> >probably worthwhile to ensure that the external journal is the same
> >size as the internal journal (can be specified with "-J size=<mb>" for
> >an internal journal, and "mke2fs -s 4096 -O journal_dev /dev/sdb6 <blocks>"
> >for the external journal.
> Well, actually, he did say - if you read the whole text of 
> the email the mke2fs commands are included in the mail -
> # external data log
> mke2fs -v -s 4096 -O journal_dev /dev/sdb6
> mkfs.ext3 -v -j -J device=/dev/sdb6 -O sparse_super -O dir_index -R stride=8

Actually, he didn't.  Nowhere does it report the size of /dev/sdb6.
The "-b 4096" option just tells you the blocksize.

The internal journal will be 32MB, but if /dev/sdb6 is 400MB or whatever
then it will give a large advantage to the external journal setup.  It won't
really test whether the external journal is the source of the speedup or if
the large journal is the source.

It might also be interesting to see what the perf looks like with a ramdisk
as the external journal (e.g. simulate an NVRAM device).  I imagine it won't
be much better than just the external journal, since you are doing mostly
sequential IO to the external journal anyways.

Cheers, Andreas
Andreas Dilger
http://members.shaw.ca/adilger/             http://members.shaw.ca/golinux/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/ext3-users/attachments/20040714/bf21b7c8/attachment.sig>

More information about the Ext3-users mailing list