A legacy application uses a highly relational MySQL database that requires cross- region failover for disaster recovery with minimal RTO/RPO. The…

Cloud Spanner with a multi-region configuration is uniquely designed to meet the stringent requirements of minimal RTO/RPO and low operational overhead for a relational workload. As a fully managed, globally distributed relational database, it provides synchronous replication across regions. This architecture ensures strong consistency and offers automatic, transparent failover in the event of a regional outage. This results in a Recovery Point Objective (RPO) of zero (no data loss) and a Recovery Time Objective (RTO) of seconds, directly addressing the core disaster recovery needs while minimizing manual intervention.

A. Cloud SQL cross-region replication requires manual promotion of a replica during failover, which increases RTO, RPO (due to replication lag), and operational overhead.

B. Firestore is a NoSQL document database and is fundamentally incompatible with a "highly relational" application without a major, costly re-architecture.

D. Running MySQL on Compute Engine represents the highest possible operational overhead, as the team must manage VMs, OS, patching, replication, and failover scripting.

1. Cloud Spanner Documentation, "Instances": "Multi-region configurations distribute the database across multiple regions... If an entire region becomes unavailable, your database remains available... Spanner provides the industry-leading 99.999% availability for multi-region instances, which translates to less than 5.26 minutes of downtime per year... with zero RPO and low RTO." (Google Cloud Documentation, Cloud Spanner, Concepts, Section: "Multi-region configurations").

2. Cloud SQL Documentation, "Disaster recovery in Cloud SQL": "To configure Cloud SQL for disaster recovery, you can use a cross-region read replica... In the event of a regional failure, you can promote the read replica in your DR region... The promotion process is manual." (Google Cloud Documentation, Cloud SQL for MySQL, High Availability, Section: "Disaster recovery in Cloud SQL"). This confirms the manual process and higher RTO for option A.

3. Cloud Spanner Documentation, "Spanner: A Global, Relationally-Consistent, Scalable, and Highly Available Database": "Spanner is Google's scalable, multi-version, globally-distributed, and synchronously-replicated database... Spanner is the first system to distribute data at global scale and support externally-consistent distributed transactions." (This is a foundational paper often referenced in academic contexts and linked from Google's official documentation, highlighting its relational and consistent nature).

4. Firestore Documentation, "Data model": "Cloud Firestore is a NoSQL, document-oriented database. Unlike a SQL database, there are no tables or rows. Instead, you store data in documents, which are organized into collections." (Google Cloud Documentation, Firestore, Guides, Section: "Data model"). This confirms the mismatch for a highly relational workload.