[Gluster-users] GlusterFS 3.0.2 small file read performance benchmark

miloska at gmail.com miloska at gmail.com
Sat Feb 27 18:00:17 UTC 2010


> Some final words:
>
> - don't add performance translators blindly (!)
> - always test with a similar workload you will use in production
> - never go and copy+paste a volume spec, then moan about bad performance
> - don't rely on "glusterfs-volgen", it gives you just a starting point!
> - less translators == less overhead
> - read documentation for all options of all translators and get an idea:
> http://www.gluster.com/community/documentation/index.php/Translators
> (some stuff is still undocumented, but this is open source... so have a
> look)
>

Thanks to share your results.

During my tests I found somewhat similar, although I didn't went into
details like you with GlusterFS, I was more interested in tunning my
RAID controller and ext3 filesystem.

Also note - has been reported on the list before - that with the
default, glusterfs-volgen created configuration 'ls -lR' does NOT
re-sync the content between AFR nodes (at least not with CentOS 5.4
x86_64). 'ls -lR'  works without the translators.

Seems for me it's safer and faster to use a stripped down config.

"Your mileage may vary"



More information about the Gluster-users mailing list