GlusterFS tcp_nodelay patch update
June 29th, 2009 by Erik Ljungstrom
As mentioned in my previous post, I wrote a patch for GlusterFS to increase its performance when operating on many smaller files. Someone told me the other day that this functionality has been pushed to the git repository. Would have been good to have heard about this sooner…
So all of you who emailed me positive feedback and asked to make it a tuneable in the translator config (thanks!) – please check out the above link to the git repository.
On another note, it seems as if they’re breaking away from having the protocol version bound to the release version, good progress in my opinion!
Posted in misc
October 6th, 2009 at 10:11 am
Nice one!
December 4th, 2009 at 8:01 pm
Hi,
if I may ask an uninformed question…
do you know if gluster at some point will offer a more kernel-internal implementation without FUSE?
I might be wrong but my feeling says there would be certain performance benefits if it ran from the kernel, right next to the network/IB drivers ;)
December 22nd, 2009 at 8:55 pm
Hi,
Unfortunately I don’t know. FUSE is quite alright performance wise, but of course there’s always drawbackcs. However, for a lot of workloads, you hit other bottlenecks before kernel/fuse becomes one.