Hi Nathan,
please find the inlined comments.
On Mon, Feb 23, 2009 at 7:30 PM, Nathan Stratton <nathan at
robotics.net>wrote:
>
> Lets say you had 4 servers and you wanted to setup replicate and
> distribute. What methoid would be better:
>
> server sdb1
> xen0 brick0
> xen1 mirror0
> xen2 brick1
> xen3 mirror1
>
> replicate block0 - brick0 mirror0
> replicate block1 - brick1 mirror1
> distribute unify - block0 block1
>
> or
>
> server sdb1 sdb2
> xen0 brick0 mirror3
> xen1 brick1 mirror0
> xen2 brick2 mirror1
> xen3 brick3 mirror2
>
> replicate block0 - brick0 mirror0
> replicate block1 - brick1 mirror1
> replicate block2 - brick2 mirror2
> replicate block3 - brick3 mirror3
> distribute unify - block 1 block2 block3 block4
This configuration is better since the brick and mirror are on two different
disks (sdb1 and sdb2 respectively) and data can be still restored if one of
the disk fails.
>
>
> <>
>>
> Nathan Stratton CTO, BlinkMind, Inc.
> nathan at robotics.net nathan at blinkmind.com
> http://www.robotics.net http://www.blinkmind.com
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
>
--
Raghavendra G
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://supercolony.gluster.org/pipermail/gluster-users/attachments/20090311/4148a04b/attachment.html>