Status page with GPU resource availability per zone

We sometimes are using large GPU accelerated instances (H200, Gaudi 3 etc.) in IBM cloud and have run into the problem that when we provision a new instance there are no resources of the correct profile available in our preferred zone.

It takes about 10-15min before we get a failure to deploy the instance message. This is highly annoying but understandable given that the GPUs are a limited resource. The only way around this issue that we have found is to just test another zone and pray that there are resources available there.

We would like to have a status indication, in what zone different GPU profiles are available for launch. It could be a separate page or integrated into the provisioning form for new instances.