So as stated in the title fault tolerance is very similar to HA. If a host goes down it ensures the VMs remain up, however how it does this is different. FT is enabled at a per-VM basis and in order for fault tolerance to even work you need to enable FT logging on at least one VMK for every host in the cluster. How it works is when you enable fault tolerance on a VM it creates a “shadow copy” of the VM and runs that on a separate host, that will continue to sit in the background waiting for a failure. This host takes up memory and storage even though it is not really doing anything, which is why it is not very common to see FT in production. When the VM fails the secondary VM immediately spins up, so essentially no downtime whatsoever. One thing I will note though is that encryption with FT logging is “opportunistic” meaning if both hosts support it it will encrypt otherwise it sends in the clear, you can get into the VMs settings and check under VM options for FT logging and have encryption required for security purposes. Better safe than sorry!