Snowflake · ADA-C02
Data Replication and Disaster Recovery
Replication operates at the database level, creating replica databases in target accounts with synchronized copies of all tables, views, stages, file formats, sequences, functions, and procedures. Continuous replication ensures changes propagate automatically. Supports same-cloud replication across regions and cross-cloud replication between different cloud platforms.
Replication operates at the database level, creating replica databases in target accounts with synchronized copies of all tables, views, stages, file formats, sequences, functions, and procedures. Continuous replication ensures changes propagate automatically. Supports same-cloud replication across regions and cross-cloud replication between different cloud platforms.
Disaster recovery leverages replication combined with failover and failback capabilities. Client redirect enables automatic redirection of client connections during failover events. Failover groups package databases, shares, warehouses, users, roles, and resource monitors into a logical unit that replicates together and fails over atomically.
Key Concepts
- Recovery Time Objective (RTO): how fast you can restore operations
- Recovery Point Objective (RPO): how much data loss is acceptable
- Planned failover exercises to test procedures and measure actual RTO/RPO
- Failback procedures to restore operations to primary region
- Replication lag monitoring to ensure replicas stay current