I am trying to put together a DR Document for a SharePoint infrastructure and I need to talk about the new AlwaysOn and Availability groups within SQL Server 2012. Has anyone used these yet with RC versions?
What I want to propose is that I will have a virtualised farm of SharePoint servers which will be 'live migrated' to another datacentre in the event of a disaster. On the DB side I was planning to suggest setting up availability groups with the Primary DB being synchonously mirrored to a Secondary DB in the main datacentre and also asynch mirroring to the DR datacentre.
I understand that all the servers in the Availability group must be part of a Windows cluster. The latency between the two datacentres would probably be around 100-200ms so can this support Windows clustering? I must admit I'm not really an infrastructure guy so it's all tough going to get my head around this!
If my proposal of using geo-stretched availability groups for 1 primary, 1 synchronous secondary and one asynch secondary is feasible then I will go with that.
In the SharePoint settings the asynch secondary in the DR site will be listed as the failover DB instance so that when the VMs would be migrated, then won't find the primary SP DBs but the failover will then kick in. All make sense?
What I want to propose is that I will have a virtualised farm of SharePoint servers which will be 'live migrated' to another datacentre in the event of a disaster. On the DB side I was planning to suggest setting up availability groups with the Primary DB being synchonously mirrored to a Secondary DB in the main datacentre and also asynch mirroring to the DR datacentre.
I understand that all the servers in the Availability group must be part of a Windows cluster. The latency between the two datacentres would probably be around 100-200ms so can this support Windows clustering? I must admit I'm not really an infrastructure guy so it's all tough going to get my head around this!
If my proposal of using geo-stretched availability groups for 1 primary, 1 synchronous secondary and one asynch secondary is feasible then I will go with that.
In the SharePoint settings the asynch secondary in the DR site will be listed as the failover DB instance so that when the VMs would be migrated, then won't find the primary SP DBs but the failover will then kick in. All make sense?