[Gluster-users] Multiple disks per server functionality

Stas Oskin stas.oskin at gmail.com
Sat Apr 4 19:14:26 UTC 2009


Actually, after posting it I had the brilliant idea of searching the
archives a bit further.

Here is an advice I found from Krishna:

"In case you have 4 machines each with 4 drives.

you can replicate each drive of server1 with corresponding drive of
server2 using AFR.
you can replicate each drive of server3 with corresponding drive of
server4 using AFR.

So you have 8 AFRs.

You can use distribute (aka DHT) to aggregate the 8 AFRs."

So basically the idea was to divide all the servers to pairs, then AFR each
drive in the pair, and then unify/dht all the AFR's in a single volume.

Is this the only approach, or there are other ones?

Because the main downside of this approach seems to me, as the need to
update all the config files on all the participating servers and clients.


2009/4/4 Stas Oskin <stas.oskin at gmail.com>

> Hi.
> Any idea, what is the best configuration to connect multiple disks per
> servers?
> I mean, I have multiple servers with 4 disks per server, and I look to
> unify all this space to a single volume, with replication of 2.
> What would be the best?
> Unify all the disks per single server to a single volume, then AFR with
> another server, then unify all once again?
> Or
> AFR per-disk basis with another server, then unify all AFR's into one
> volume?
> I don't want to use RAID or volume, as I want each server to withstand loss
> of almost all of it's disks (except the system one of course).
> Regards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20090404/1f775080/attachment.html>

More information about the Gluster-users mailing list