What's changed: Created Professional Cloud Database Engineer Chapter 4 (Domain 4 "Deploy HA": provisioning and testing = HA provisioning/IaC automation/failover drills/Cloud Monitoring; multi-regional replication and read replicas = regional resilience/read-write separation/continuous replication-lag monitoring/Spanner native multi-region).
4.2Multi-regional replication and read replicas
Understand setting up multi-regional database replication, deploying and scaling read replicas, separating reads from writes, and continuously monitoring the overall highly available setup.
To balance availability and scale, use cross-region replication and read replicas appropriately, designing the write and read paths.
4.2.1Multi-regional replication and read replicas
For regional resilience, replicate data to another region via multi-regional replication (Spanner is natively multi-region). To handle read load, deploy/scale read replicas and design read/write separation (writes to the primary, reads to replicas). Read replicas can aid availability too, but differ in purpose from HA auto-failover. Map "scale reads = read replicas" and "regional resilience = multi-regional replication."
4.2.2Continuously monitoring the setup
Continuously monitor multi-region/read-replica setups: watch each replica’s replication lag, read-replica load balancing, and failover readiness with Cloud Monitoring. High lag raises the risk of data loss on failover and reading stale values from read replicas. Map "continuously monitor replica lag and load."
Continue reading — free sign-up
You're reading the free preview. Sign up free to read this section in full, plus every chapter (including 4+) and all questions.