[Users] Creation of preallocated disk with Gluster replication

[gregoire.leroy at retenodus.net]

> Do you happen to notice what is consuming CPU?

When I check with top, glusterfs and glusterfsd are the only process who 
use a significant amount of CPU. Load average is between 5 and 6, and I 
don't have any started VM.

> Since the same cluster
> does both virtualization and storage, a GigE network might get
> saturated very quickly. Is it possible to separate out the management
> and data/gluster traffic in this setup?

Unfortunately, it's not possible. I only have two hosts, both for 
virtualization and storage. Why does glusterfs use so much CPU ?

About the network traffic, the creation of a pre-allocated disk is not 
something which absolutely needs to be fast : if it takes 5 minutes, 
well... it takes 5 minutes. I guess that if there's a way to limit the 
bandwidth available for glusterfs, it'll still have an impact on 
production because glusterfs and VM will be in competition for 
bandwidth, right ? Is there a way to limit the bandwidth only for the 
creation of a pre-allocated disk ?

Thank you,
Regards,
Grégoire Leroy