Multi-Az And Data Durability Design — AWS SysOps Administrator (SOA-C03)
High Availability and Data Durability Are Separate Guarantees
Requirement: an application that survives an AZ failure without data loss. Competing designs: Multi-AZ RDS vs. single-AZ RDS with cross-region snapshots. The deciding constraint is recovery type. Multi-AZ provides synchronous replication with automatic failover and minimal data loss. Cross-region snapshots provide geographic durability but require manual restore and accept data loss proportional to snapshot frequency. Match the protection mechanism to the failure scenario, not just to the presence of redundancy.
What This Pattern Tests
The exam tests AZ-level resilience as the baseline design pattern. RDS Multi-AZ deploys a synchronous standby in another AZ — automatic failover in 60-120 seconds, no data loss. Read Replicas use asynchronous replication — no automatic failover, potential data lag. EBS volumes are AZ-scoped — an AZ failure loses access to the volume. S3 stores objects across at least 3 AZs automatically. The trap is confusing Multi-AZ (high availability, synchronous) with Read Replicas (read scaling, asynchronous) or assuming EBS survives an AZ failure (it does not without snapshots).
Decision Axis
Availability target determines AZ design: 99.9% may need single-AZ with backups, 99.99% needs Multi-AZ, 99.999% needs Multi-Region.
Associated Traps
More Top Traps on This Exam
Decision Rules
Whether to enable RDS Multi-AZ (synchronous standby, automatic promotion, near-zero RPO) or promote an RDS Read Replica (asynchronous replication, manual promotion step required) as the database HA mechanism when the constraint demands both automatic failover and near-zero data loss.
Whether to convert the RDS instance to Multi-AZ (synchronous replication, automatic promotion, RTO within minutes) versus adding an RDS Read Replica (asynchronous, manual promotion required), given that the automatic-failover and sub-five-minute RTO constraints together disqualify the Read Replica path.
Whether to convert the RDS instance to Multi-AZ deployment (synchronous standby, automatic promotion, near-zero RPO) versus adding an RDS Read Replica (asynchronous, manual promotion required), given an explicit RPO-of-zero and sub-2-minute RTO constraint.
Domain Coverage
Difficulty Breakdown