By the end of this year, we will have 4 small data centers at 4 sites, each with 2 VM hosts.
Each site and DC has a unique set of subnet and vlans.
Of the 2 VM hosts at each site, they would provide DR to each other and not DR between the sites.
Our current DR plan is that if Site A goes down and we lose access to datacenter A, then site A is down and we do not attempt to migrate VMs from Site A to Site B.
We expect that migrating guest from Site A to Site B, each with unique subnets would be too complex with too many dependencies.
We would just accept that Site A is down. If Site A is down, then something big is happening anyway.