This is exactly what many of the really big enterprises do. At their scale the cloud becomes laughable. But as a backup, it's the clear way to go. Primary on prem with automatic failover to the cloud.
I worked with a large customer to help build this for them. OpenShift running on-prem, but they had some failover equipment that would ansible them an OpenShift cluster on aws. Depending on the nature of the failure it did take a little time to fail over, but 15 to 30 mins of downtime in the event of a catastrophic failure is often worth it to save hundreds of millions per year.
I worked with a large customer to help build this for them. OpenShift running on-prem, but they had some failover equipment that would ansible them an OpenShift cluster on aws. Depending on the nature of the failure it did take a little time to fail over, but 15 to 30 mins of downtime in the event of a catastrophic failure is often worth it to save hundreds of millions per year.