[Gluster-devel] Unify and AFR?

[Amar S. Tumballi]

If I understand you correctly, you need 2 copies of every file in filesystem
(which has 10disks), right?
Then yes, unify over afr is able to give you a solution.

Unify (afr1 (brick1,brick2),afr2(brick3,brick4),afr3(brick5,brick6),
afr4(brick7,brick8),afr5(brick9,brick10))

This should work fine for you.

Regards,
Amar

On Thu, Apr 3, 2008 at 8:59 AM, Adam Stiles <adam at stilesoft.com> wrote:

> I've been reading the wiki and pouring over the mailing list archives
> but haven't found the answer to some GlusterFS questions.
>
> I need a distributed filesystem that presents a unified mount point
> and supports some arbitrary number of bricks (sounds like unify). We
> may start with ten storage bricks and may scale to 50 or so.
> Additionally, we need to store two replicas of each file (sounds like
> AFR). Does GlusterFS support the use of AFR and unify when there are
> more bricks than replicas? Any pointers would be greatly appreciated.
> Next step for me will be to setup a demo cluster using this setup,
> probably using Amazon EC2. Thanks,
>
> Adam
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>



-- 
Amar Tumballi
Gluster/GlusterFS Hacker
[bulde on #gluster/irc.gnu.org]
http://www.zresearch.com - Commoditizing Supercomputing and Superstorage!