I have a question about the limits of cluster peers in 8.1CM. Let's assume the following:
1. Brocade VDX ethernet fabric in two data facilities, LA and NYC
2. Ethernet fabric extended from NYC to LA via 1Gbps ethernet private line
3. 2 FAS32xx dual controller w/FCoE CNA's running 8.1CM - a pair per facility
Can I create a single namespace file system with the ethernet fabric extended (compressed and encrypted via Brocade VCS Fabric Extension) to the other two FAS32xx nodes? My thinking is that if I can stretch layer 2 from NYC to LA and have a single global namespace. This way I can have multiple "active-active" datacenters serving structured content anywhere.
Is this possible over such a distance? Is a WAN accelerator recommended for "semi-sync" replication? I think the distance might be the gating factor here, but correct me if I'm wrong.
The goal is a single namespace w/ONTAP 8.1 ala GlusterFS and Isilon - but with an FC target for databases and NFS for everything else.
No, this solution is currently not supported. May I suggest contacting your local NetApp SE and cluster-mode expert for a more detailed discussion? There may be other ways to provide the functional outcome you are seeking.