For a mission-critical application requiring constant global availability, which deployment strategy should be adopted?

For a mission-critical application that demands constant global availability, the best strategy involves utilizing multiple fault domains within any availability domain alongside multiple regions. This approach ensures that the application is not only robust against localized failures but also benefits from geographic redundancy.

When using multiple fault domains, you are essentially distributing your workload across different physical hardware within the same availability domain, mitigating the risk of a single point of failure. By deploying across multiple regions, you gain protection against larger-scale outages that can occur due to regional disasters or significant events affecting an entire area.

This strategy enhances your application's resilience, allowing it to maintain its operations even in scenarios where one region experiences issues. Furthermore, having a backup in another region ensures that user access remains uninterrupted, providing the constant availability needed for mission-critical applications. This global stretching improves disaster recovery capabilities and aligns with best practices for high availability in cloud architectures.

The other choices, while they offer some level of availability or redundancy, do not provide the same level of protection from global outages or catastrophic events as the selected strategy does. Options that limit deployment to a single region are inherently at risk if that region encounters failures, thereby diminishing the overall availability that the application would require.