Has anyone mastered the art of scaling the Forwarding Appliance for the VMWare app? Say I want to put 100 hosts on one appliance and have sufficient CPU\MEM\DISK resources. Has anyone worked through the requirements and accomplished this reliably in production? I'm looking for some sort of chart which gives a rough idea of the resources required to accommodate x number of ESX hosts.
I think you'll be better off splitting up the hosts between multiple FAs. Scale out, not up.
It adds the benefit of easier scheduling of the FA VMs on your hypervisor.
The documentation suggests 20-30 hosts per 4 vCPU FA. I suspect the limit is more centered around the number of VMs on those hosts and you want to keep it under 300 VMs per FA.
I think you'll be better off splitting up the hosts between multiple FAs. Scale out, not up.
It adds the benefit of easier scheduling of the FA VMs on your hypervisor.
The documentation suggests 20-30 hosts per 4 vCPU FA. I suspect the limit is more centered around the number of VMs on those hosts and you want to keep it under 300 VMs per FA.