We currently monitor the VC and 45 hosts(we want to expand this to 3 x VC and 80+ hosts), so adding each one again would take up a lot of additional basic monitor licences. It would also prove tricky to have to check whether a monitor was down from a single location or both in the event of an outage.
The use case is we run VMware split into multiple separate clusters across 2 sites. If the on-prem poller fails we would lose all VMware monitoring. Also if we experienced a site related issue we would get false positives for the 2nd site as the poller would not be able to communicate that the VMware resources were up.
OnPremise Poller High Availability sounds interesting but lI do not see why we shouldn't be able to monitor a single resource ie ESXi host from multiple poller in the same way we would an internal website. That way we would cover ourselves for both situations described above.
I'm also very keen for the Hardware alerting to be implemented for VMware, I see that you can monitor the hardware but there is no way of us being alerted which is one thing holding us back from moving all of our VMware into Site24x7.