When vSphere High Availability (HA) has been enabled on a vSphere cluster, the virtual machines running on the cluster are protected from a host hardware failure or virtual machine guest operating system crash.
In the event that a host suffers a hardware failure, or if ESXi crashes, the virtual machines are restarted on the surviving hosts in the cluster. Resources must be reserved in the cluster to guarantee that the necessary resources are available to restart the virtual machines.