Reduce downtime if a single-AZ RDS instance fails.
Use a synchronized standby and keep the requirement separate from normal read scaling.
Which design is most appropriate?
RDS Multi-AZ maintains a synchronized standby in another Availability Zone and can fail over automatically when the primary is unavailable.
Detailed explanation
Read replicas primarily scale reads and are not the same as Multi-AZ automatic failover.
Read replicas primarily scale reads and are not the same as Multi-AZ automatic failover.
A single-AZ failure can affect both instances and the process is not automatic.
A single-AZ failure can affect both instances and the process is not automatic.
Multi-AZ synchronously replicates to a standby in another AZ and supports failover.
Multi-AZ synchronously replicates to a standby in another AZ and supports failover.
Restoration takes time and does not provide short automatic failover.
Restoration takes time and does not provide short automatic failover.
Check official documentation
Compare the following points with the official documentation.
Amazon RDS Multi-AZの配置とフェイルオーバーの公式説明を確認する。
What to confirm
高可用性と読み取りスケールの要件をMulti-AZとリードレプリカへ分けられる。
Key points
- RDS Multi-AZ
- Automatic failover
- Synchronous standby
- High availability
Notes
- Environment: AWS公式RDSドキュメントの確認
- Confirm current AWS architecture guidance and service behavior in the official documentation.
Foundation review
Failure domain
A different AZ reduces exposure to a failure confined to one AZ.
Standby role
The standby protects availability; it is not the normal read-scaling endpoint.