How to ensure high availability of pods in Openshift

Issue

• We would like to map such a guarantee to pods in order apps may be sure that if a host dies the pods are distributed in a way that at least a pod (or the app defined number) survive without any glitches; I mean that do not want the service interruption that would occur while kube restarts the pods.is tagging openshift nodes (like we did for regions) in a way that zones are mapped to antiaffinity rules on VMWARE the only way to do that ?Do we have to change the scheduler policy of kube