[Gluster-devel] performance

Kevan Benson kbenson at a-1networks.com
Tue Nov 6 18:32:12 UTC 2007


1) Stock fuse, or glusterfs patched fuse?  See 
http://ftp.zresearch.com/pub/gluster/glusterfs/fuse/.  The Glusterfs 
team has some changes to some default values in fuse to make it perform 
better for common glusterfs scenarios, as well as a fix for locking, so 
you are better off using the glusterfs supplied fuse if you want better 
performance and or locking.

2) The read-ahead and write-behind translators are there to boost 
performance for certain scenarios if you know the types of access your 
mount will be doing much of the time.

3) The real speed benefits arise when you are able to span reads across 
multiple servers, increasing response and transfer rate.  This is where 
the real benefits are (as well as redundancy), which NFS can't really 
compete with (unless you're using Solaris).

4) That's a real close benchmark.  Are you sure the medium over which 
you are transferring the data isn't maxed?  IB or TCP/IP?  100Mbit or 
1000Mbit (and server grade or workstation grade cards if gigabit).

Chris Johnson wrote:
> On Tue, 6 Nov 2007, Anand Avati wrote:
> 
>      Fuse is fuse-2.7.0-1.  Can you give me an example of a volume
> specification file please?
> 
>> Chris,
>> posting your volume specification files will help us tune it futher. 
>> Also,
>> what fuse version (kernel module) are you using?

-- 

-Kevan Benson
-A-1 Networks





More information about the Gluster-devel mailing list