You want to migrate an on-premises mission-critical PostgreSQL database to Cloud SQL. The database must be able to withstand a zonal failure with less than five minutes of downtime and still not lose any transactions. You want to follow Google-recommended practices for the migration. What should you do?

Question

Accepted Answer

D. Enable high availability (HA) for the database to make it regional. Cloud SQL HA uses a regional configuration with a primary instance in one zone and a standby instance in another zone within the same region, connected via synchronous replication. In the event of a zonal failure, Cloud SQL automatically promotes the standby with no manual intervention, completing failover in well under 5 minutes with no committed transactions lost - satisfying both the RTO and RPO requirements.

Answer

A. Take nightly snapshots of the primary database instance, and restore them in a secondary zone. Nightly snapshots introduce an RPO of up to 24 hours because transactions occurring after the last snapshot would be lost; restoring a snapshot also takes far longer than 5 minutes.

Answer

B. Build a change data capture (CDC) pipeline to read transactions from the primary instance, and Building a custom CDC pipeline is not a Google-recommended practice for zonal HA in Cloud SQL and adds significant operational complexity compared to the built-in HA feature.

Answer

C. Create a read replica in another region, and promote the read replica if a failure occurs. A read replica in another region uses asynchronous replication, meaning some committed transactions may not yet be replicated at the time of failure, causing data loss that violates the zero-transaction-loss requirement.

You want to migrate an on-premises mission-critical PostgreSQL database to Cloud SQL. The database must be able to withstand a zonal failure with less than five minutes of downtime and still not…

Question

Options

How the community answered

Why each option

Topics

Community Discussion