For a GKE cluster with Cloud SQL storing persistent data, what is the most suitable failover strategy to meet disaster recovery objectives with RTO and RPO within minutes?

*To ensure a robust disaster recovery plan with RTO and RPO objectives within minutes for a GKE cluster with Cloud SQL, the best strategy is to employ a multi-zone GKE cluster along with Cloud SQL configured for regional replication. This setup provides high availability and quick recovery capabilities for both the GKE cluster and Cloud SQL. Multi-zone clusters offer resilience against zone failures, while regional replication in Cloud SQL ensures data persistence and availability even in the event of regional disruptions. Frequent backups or a multi-cluster setup may either be impractical or add unnecessary complexity and cost. Utilizing Cloud Endpoints for API call duplication or relying solely on monitoring metrics for failover does not comprehensively address the quick recovery and data consistency requirements essential for mission-critical workloads.*