We need to provide highly available, disaster recovery solution for SharePoint 2013 deployment. For this purpose, we've to build environments for two data-centres, one would be primary DC1 and another would be secondary (disaster recovery) DC2 site. In case of any disaster/shut-down, all requests to DC1 would be routed to DC2 by DNS-record update.
Both Primary (DC1) & Secondary DC2 site would have the following environments configuration:
2 Web Servers (SharePoint 2013 Web Front End Server running on Windows Server 2008 R2)
2 Application Servers (SharePoint 2013 Application Server/Services/Central Admin running on Windows Server 2008 R2)
2 Database Servers (Hosting SharePoint 2013 Config/Content DBs on SQL Server 2008/2012)
Our requirements are as below:
We need automatic fail-over between these data-centres, as soon as DC1 is unavailable, DC2 would be active.
Do we really need the shared storage like SAN/RAID for SQL Server database in the same data-centre? What would be the criteria for using shared storage?
Should we use Fail-Over Clustering/ SQL Server 2012 AlwaysOn feature for High Availability in the same data-centre and between data-centres?
How should we sync these DC1 & DC2 environment i.e. SharePoint and/or SQL Server databases level (Fail-Over Clustering and/or Log Shipping)