I would like to hear your recommendations and reflections on the following situation/need.
A company needs a solution that allows a maximum of 1-2 hours downtime if major hardware problems occur.
The goal is to install a solution where we run for ex Hyper-V on a physical server, and 3-4 guest VMs with windows 2008 R2 (using the enterprise license). This physical server with internal drives run LOB applications, AD, File/Print services and cannot be down more than 1-2 hours max. This means we cannot only rely on hw vendor 24x7 warranty services.
Instead we need to build a more redundant system with virtual machines secured somehow.
One idea we are investigating is using two physical servers (one primary server) with Hyper-V and VM guests. The primary server would replicate to the secondary server using for example double-take or neverfail (we have no experiences from these and what modules to use).
- Is this a good/best solution to the need above? are there other alternatives using either hw or sw. Other alternatives, perhaps cheaper than DT and Neverfail?
On-top of the replicaton of VM guests we would like to also do virtual machine backups so that if we in worst case get two physical hw server failures dont loose the vm guests. So a integration of VM guest backup to a separate NAS would be great to accomplish.
Any ideas on solutions, hw or sw? comments on how to best achieve HA and disaster recovery for windows server systems.
appreciate any tips.