James -
The setup you've described is pretty standard, if we assume that you
are going to mount each array at /mnt/array{1-8}, your volume will be
called vol1, and your servers are named server{1-4} your gluster volume
create command would be -
Without replicas -
#gluster volume create vol1 transport tcp server1:/mnt/array1
server2:/mnt/array1 server3:/mnt/array1 server4:/mnt/array1
server1:/mnt/array2 server2:/mnt/array2 server3:/mnt/array2
server4:/mnt/array2 server1:/mnt/array3 server2:/mnt/array3
server3:/mnt/array3 server4:/mnt/array3 server1:/mnt/array4
server2:/mnt/array4 server3:/mnt/array4 server4:/mnt/array4
server1:/mnt/array5 server2:/mnt/array5 server3:/mnt/array5
server4:/mnt/array5 server1:/mnt/array6 server2:/mnt/array6
server3:/mnt/array6 server4:/mnt/array6 server1:/mnt/array7
server2:/mnt/array7 server3:/mnt/array7 server4:/mnt/array7
server1:/mnt/array8 server2:/mnt/array8 server3:/mnt/array8
server4:/mnt/array8
This would get you a single 512TB NFS mount.
With replicas(2) -
#gluster volume create vol1 replica 2 transport tcp server1:/mnt/array1
server2:/mnt/array1 server3:/mnt/array1 server4:/mnt/array1
server1:/mnt/array2 server2:/mnt/array2 server3:/mnt/array2
server4:/mnt/array2 server1:/mnt/array3 server2:/mnt/array3
server3:/mnt/array3 server4:/mnt/array3 server1:/mnt/array4
server2:/mnt/array4 server3:/mnt/array4 server4:/mnt/array4
server1:/mnt/array5 server2:/mnt/array5 server3:/mnt/array5
server4:/mnt/array5 server1:/mnt/array6 server2:/mnt/array6
server3:/mnt/array6 server4:/mnt/array6 server1:/mnt/array7
server2:/mnt/array7 server3:/mnt/array7 server4:/mnt/array7
server1:/mnt/array8 server2:/mnt/array8 server3:/mnt/array8
server4:/mnt/array8
This would get you a single 256TB HA NFS mount.
Gluster specifically doesn't care about LUN/brick size, the ability to
create smaller LUNs without affecting the presentation of that space is
a positive side effect of using Gluster. Smaller LUN's are useful in
several ways, faster fsck's on the LUN if that is ever required, there
is a minor performance hit to running bricks of different sizes in the
same volume, small LUNs make that easier.
Thanks,
Craig
-->
Craig Carl
Senior Systems Engineer
Gluster
On 12/01/2010 08:29 AM, Burnash, James wrote:
> Hello.
>
> So, here's my problem.
>
> I have 4 storage servers that will be configured as replicate + distribute, each of which has two external storage arrays, each with their own controller. Those external arrays will be used to store archived large (10GB) files that will only be read-only after their initial copy to the glusterfs storage.
>
> Currently, the external arrays are the items of interest. What I'd like to do is this:
>
> - Create multiple hardware RAID 5 arrays on each storage server, which would present to the OS as approx 8 16TB physical drives.
> - Create an ext3 file system on each of those devices (I'm using CentOS 5.5. so ext4 is still not really an option for me)
> - Mount those multiple file systems to the storage server, and then aggregate them all under gluster to export under a single namespace to NFS and the Gluster client.
>
> How do I aggregate those multiple file systems without involving LVM in some way.
>
> I've read that Glusterfs likes "small" bricks, though I haven't really been able to track down why. Any pointers to good technical info on this subject would also be greatly appreciated.
>
> Thanks,
>
> James Burnash, Unix Engineering
>
>
> DISCLAIMER:
> This e-mail, and any attachments thereto, is intended only for use by the addressee(s) named herein and may contain legally privileged and/or confidential information. If you are not the intended recipient of this e-mail, you are hereby notified that any dissemination, distribution or copying of this e-mail, and any attachments thereto, is strictly prohibited. If you have received this in error, please immediately notify me and permanently delete the original and any copy of any e-mail and any printout thereof. E-mail transmission cannot be guaranteed to be secure or error-free. The sender therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of e-mail transmission.
> NOTICE REGARDING PRIVACY AND CONFIDENTIALITY Knight Capital Group may, at its discretion, monitor and review the content of all e-mail communications. http://www.knight.com
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
