|
I finally did find some stats on teaming
http://rhelblog.redhat.com/2014/06/23/team-driver/
On 6/19/2017 10:42 AM, WK wrote:
OK, at least its not an *issue* with Gluster. I didn't expect
any but you never know.
I have been amused at the 'lack' of discussion on Teaming
performance found on Google searches.
There are lots of 'here it is and here is how to set it up'
articles/posts, but no 'ooh-wee-wow it is awesome' comments.
It seems that for most people Bonding has worked it kinks out
and thus serves its purpose perfectly fine, so no reason to
change to Teaming.
That or they had Mahdi's experience with Teaming in its present
state and bonding just works better for now.
For now, I'll just use tried and true ALB bonding on this project.
After I have some free time I may play with it some more on a test
bed we have here.
Among the 'advantages' claimed for Teaming is the lower
performance overhead (though "low vs very-low" may not be
significant)
https://github.com/jpirko/libteam/wiki/Bonding-vs.-Team-features
Thanks all for the feedback.
I'll let everyone know if my testing happens and if it is a
significant difference (good or bad).
-bill
On 6/19/2017 6:27 AM, David Gossage
wrote:
I haven't done any testing of performance
differences, but on my oVirt/rhev i use standard bonding as
that's that it supports. On the stand along gluster nodes I
use teaming for bonding.
Teaming may be slightly easier to manage, but not by much
if you are already used to bond setups. I haven't noticed
any bugs or issues using teaming.
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users
|
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users
