I set up a physically isolated Static IP network at my main client over a decade ago, with various DirectSoft/Do-more/C-more nodes. It's nice and simple and I manage the master IP address table. But now I do all my work for them remotely, and the only way to access the PLCs is with VNC Viewer to an on-site engineer's computer that has an extra adapter plugged into our automation ring.
So, for this and other reasons, we are now beginning to migrate our PLC and HMI connections onto the main company network, no longer managed by myself or the automation group. In this particular location this is the only way that we will be allowed to implement remote/VPN access to individual control nodes.
There are scores of DS and DM ethernet nodes involved. The IT folks will work with us, but the network is way too complex for them to be able to keep all the engineer's PC's and the PLC's and HMI's on one subnet. There now are segmented VLANs and routable connections, and we can still connect just fine, but we lose our handy NetEdit browsing to find a PLC's address.
IT agrees to set fixed IP addresses based on unique switch port number or MAC address as we wish, but they would prefer being able to reassign subnets as necessary. If we convert to defining PLC connections by DNS name, there will be far fewer issues as we progress.
As we all continue to migrate toward more remote connectivity on highly-managed company networks, I would expect to see more of this type of thing. I thought I'd ask because I figured it's not difficult to implement. It does involve a subtle change to the connection setup dialog, though, that can potentially throw a lot of users for a loop.