Which statement about multi-region, multi-AZ architecture is true?

Which statement about multi-region, multi-AZ architecture is true?

• A. It limits resources to a single region to minimize cross-region latency.
• B. It requires manual failover to switch regions during outages.
• C. It prevents data replication across regions.
• D. It deploys identical resources across regions and zones to improve resilience, reduce latency, and enable local failover.

Answer: (D)

Multi-region, multi-AZ architecture centers on placing identical resources across multiple regions and multiple availability zones so the system remains available even if a region or a zone fails, while also bringing users closer to a functioning copy of the service. This setup enables resilience because a failure in one region or zone does not bring down the entire application; traffic can be served from other regions or zones. It reduces latency because users are routed to the nearest healthy region, improving response times. It also supports local failover, meaning a region or zone can take over traffic quickly without manual intervention in many configurations. The other statements clash with this approach: limiting resources to a single region contradicts the very idea of multi-region deployment; requiring manual failover goes against the design goal of automated or streamlined recovery in distributed architectures; and preventing data replication across regions defeats the purpose of cross-region resilience and DR planning.

Multi-region, multi-AZ architecture centers on placing identical resources across multiple regions and multiple availability zones so the system remains available even if a region or a zone fails, while also bringing users closer to a functioning copy of the service. This setup enables resilience because a failure in one region or zone does not bring down the entire application; traffic can be served from other regions or zones. It reduces latency because users are routed to the nearest healthy region, improving response times. It also supports local failover, meaning a region or zone can take over traffic quickly without manual intervention in many configurations.

The other statements clash with this approach: limiting resources to a single region contradicts the very idea of multi-region deployment; requiring manual failover goes against the design goal of automated or streamlined recovery in distributed architectures; and preventing data replication across regions defeats the purpose of cross-region resilience and DR planning.