What is the best Cloud SQL setup for low RTO/RPO during a regional outage?

PDE By islever Editorial Last updated

Short answer

C Create a highly available Cloud SQL instance in region A. Create a highly available read replica in region B. Scale up read workloads by creating cascading read replicas in multiple regions. Promote the read replica in region B when region A is down.

Question

You are deploying a MySQL database workload onto Cloud SQL. The database must be able to scale up to support several readers from various geographic regions. The database must be highly available and meet low RTO and RPO requirements, even in the event of a regional outage. You need to ensure that interruptions to the readers are minimal during a database failover. What should you do?
  1. Create a highly available Cloud SQL instance in region Create a highly available read replica in region B. Scale up read workloads by creating cascading read replicas in multiple regions. Backup the Cloud SQL instances to a multi-regional Cloud Storage bucket. Restore the Cloud SQL backup to a new instance in another region when Region A is down.
  2. Create a highly available Cloud SQL instance in region A. Scale up read workloads by creating read replicas in multiple regions. Promote one of the read replicas when region A is down.
  3. Create a highly available Cloud SQL instance in region A. Create a highly available read replica in region B. Scale up read workloads by creating cascading read replicas in multiple regions. Promote the read replica in region B when region A is down. Correct Answer
  4. Create a highly available Cloud SQL instance in region A. Scale up read workloads by creating read replicas in the same region. Failover to the standby Cloud SQL instance when the primary instance fails.

Key Takeaways

The exam tests whether you know that promoting a cross-region HA read replica provides faster failover than restoring a backup, and that same-region replicas cannot survive a regional outage.

For a Cloud SQL MySQL workload, deploy an HA instance in Region A, an HA read replica in Region B, and use cascading replicas; promote the Region B replica on regional outage to meet low RTO/RPO. Community consensus favors this option over backup/restore and same-region replication.

Common mistake: Selecting D, because it mentions high availability, but it only covers zone-level failover and does not protect against a regional outage.

Expert Analysis

Why the Answer Is Correct

Option C is the most complete solution. It creates a highly available Cloud SQL instance in the primary region and a highly available read replica in a second region. By using cascading read replicas, it also scales read workloads across multiple geographic areas. When Region A experiences an outage, promoting the HA read replica in Region B provides a fast failover with minimal interruption, and the HA configuration of that replica helps keep it resilient.

The promotion of a replica is a much faster operation than restoring a backup, which is why A is not acceptable. As comment [1] and [2] note, restoring from backup takes too long and would not meet low RTO/RPO requirements. C combines geographic distribution with a quick failover path.

Why the Other Options Are Wrong

Option A relies on restoring a multi-regional Cloud Storage backup to a new instance. Restores are time-consuming and involve data transfer, making it impossible to meet low RTO/RPO in a disaster scenario. The backup approach is also more complex and does not provide seamless reader failover.

Option B creates read replicas in multiple regions and promotes one, but it does not make the secondary replica highly available. A single-zone read replica can become a single point of failure, as noted by the lack of HA in the replica tier. This can increase downtime for readers during failover.

Option D only places read replicas in the same region as the primary. This handles a zonal failure by failing over to the standby, but it does nothing for a regional outage. It also fails to meet the requirement of supporting readers from various geographic regions.

Community Comment Notes

Comment [1] and [2] explicitly reject backup/restore because it is too slow, which supports C over A. Comment [4] argues for B, but it overlooks the need for the promoted replica to be highly available in a secondary region, which C provides. Comment [3] questions whether HA read replicas exist, but Cloud SQL supports HA on read replicas, and the official replication documentation covers this setup.

Official Reference

These links open the official vendor documentation in a new tab.

Exam Strategy

In disaster-recovery scenarios, always weigh RTO and RPO: promoting a read replica is much faster than restoring from backup. Before choosing an answer, verify that the replica you will promote is in a different region and has high availability enabled so that the failover is both quick and resilient.

Community Votes

C
100%

100% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 6 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 6

  1. rohan.sahi 👍 10 Answered C
    Option C: Because HA read replica in multiple regions. NotA: Coz restore from back up is time taking NotB: No HA in Multiple regions read replica Not D: Only one region mentioned.
  2. raaad 👍 5 Answered C
    - Combines high availability with geographic distribution of read workloads. - Promoting a highly available read replica can provide a quick failover solution, potentially meeting low RTO and RPO requirements. ===== Why not A: Restoring from backup to a new instance in another region during a regional outage might not meet low RTO and RPO requirements due to the time it takes to perform a restore.
  3. Matt_108 👍 1 Answered B
    To me, it's B. it provides: High availability: The highly available Cloud SQL instance in region A will ensure that the database remains accessible even if one of the zones in the region becomes unavailable. Scalability: The read replicas in multiple regions will enable you to scale up the read capacity of the database to support the demands of readers from various geographic regions. Minimal interruptions: When region A is down, one of the read replicas in another region will be promoted to become the new primary instance. This will ensure that there is no interruption to the readers.
  4. tibuenoc 👍 1 Answered B
    https://cloud.google.com/sql/docs/mysql/replication This option involves having read replicas in multiple regions, allowing you to promote one of them in the event of a failure in region A. While there may still be a brief interruption during the failover, it is likely to be less than the time required for the synchronization of cascading read replicas.
  5. scaenruy 👍 1 Answered A
    A. Create a highly available Cloud SQL instance in region Create a highly available read replica in region B. Scale up read workloads by creating cascading read replicas in multiple regions. Backup the Cloud SQL instances to a multi-regional Cloud Storage bucket. Restore the Cloud SQL backup to a new instance in another region when Region A is down.
Show 1 more comment
  1. mi_yulai 👍 1
    Why C? Is it possible to have HA enable in different regions? How the synchronization in disk will wokr for HA?

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

How to establish consistent BigQuery spend?

S-grade

Triggering Airflow DAGs Reactively in Private Cloud Composer

S-grade

How to resolve Cloud Composer 2 worker pod evictions and task failures?

S-grade

Which Google Cloud database for on-prem PostgreSQL with analytics?

S-grade

← Back to PDE Study Guide