Availability zones are logical separations of compute resources, representing groups of hypervisors that a user will be able to select when requesting to launch an instance. If no availability zones have been created, the default called nova will be used. When an instance is launched in most default cases, the scheduler determines which host will run it within that zone. As a user of the OpenStack cloud, you can specify which availability zone to use if more than one is available. This can help you create more resilient applications. By allowing an instance to be spun up in two separate places, we are protecting ourselves against the failure of a complete zone.