PRTG large installations need XL1 or even EM. Complex network monitorig should make available features that make things simple.
HA cluster has 5000 sensors limitation which will increase server count. This should be overcome. Another problem here is that any customisation on the master note does not sync with the HA cluster node. PRTG cluster formation process should scan for any customisation done on the master and shift those to the HA failover cluster. Doing this manually in large and highly diverse network is going to be messy.
Request this feature to be added to the roadmap.