[Gluster-users] New GlusterFS Config with 6 x Dell R720xd's and 12x3TB storage

Mike Hanby mhanby at uab.edu
Fri Nov 30 19:21:54 UTC 2012


We have the following hardware that we are going to use for a GlusterFS cluster.
6 x Dell R720xd's (16 cores, 96G)
PERC H710 Raid controller
12 x 3TB SAS drives
2 x 270G SAS OS drives configured in RAID 1

I've installed each of these with CentOS 6.3 x86_64 on the built in RAID 1 storage.

I now need to decide how to configure the 12 x 3TB disks in each server, followed by partitioning / formatting them in the OS.

The PERC H710 supports RAID 0,1,5,6,10,50,60. Ideally we'd like to get good performance, maximize storage capacity and still have parity :-)

Any suggestions on how to configure the backend storage?
* RAID level
* Number of virtual disks (supports 1 per RAID set)
* Stripe Element Size: 64, 128, 256, 512KB, 1MB
* Read Policy: Read Ahead, Adaptive Read Ahead, No Read Ahead
* Write Policy: Write Back, Write Through, Force Write Back
* Disk Cache Policy: Enabled, Disabled

According to the PERC manual, FastPath (provides IO performance for block sizes smaller than the virtual disk's stripe size) is available when the virtual disk is configured with Write Through and No Read Ahead cache policies. This is only available for RAID 0,1,5,6

We plan to test multiple configurations, but figured I'd check to with the expertise here before we started.

Thanks for any suggestions,

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20121130/e36b00ad/attachment.html>

More information about the Gluster-users mailing list