A financial application must continue operating if an entire OCI region becomes unavailable. Which design addresses this failure scope?

1. Create additional compartments in the existing region
2. Place all instances in different fault domains of one availability domain
3. Increase every instance's memory allocation
4. Deploy a recovery environment in another region and replicate required data