A regional outage requires recovery in another Region.
The target RTO is a few minutes, but running full production capacity continuously is too expensive.
Which design is most appropriate?
Run a smaller warm-standby environment in another Region and scale it when a disaster occurs.
Detailed explanation
DNS settings alone do not provide an application or data recovery target.
DNS settings alone do not provide an application or data recovery target.
A running reduced environment can be scaled during an outage and generally reaches service sooner than a pilot light.
A running reduced environment can be scaled during an outage and generally reaches service sooner than a pilot light.
This can minimize recovery time but violates the constraint to avoid full steady-state cost.
This can minimize recovery time but violates the constraint to avoid full steady-state cost.
A cold recovery requires construction and startup time, which is unfavorable for a short RTO.
A cold recovery requires construction and startup time, which is unfavorable for a short RTO.
Check official documentation
Compare the following points with the official documentation.
AWS disaster recovery strategies warm standby and pilot lightの公式説明を確認する。
What to confirm
バックアップ、パイロットライト、ウォームスタンバイ、アクティブ・アクティブをRTOとコストで比較できる。
Key points
- Warm standby
- Pilot light
- RTO
Notes
- Environment: AWS公式SOA-C03試験ガイドとAWS公式ドキュメントの確認
- Confirm current AWS service behavior in the official documentation.
Foundation review
Readiness level
Warm standby is more immediately ready than pilot light but costs more while idle.
Capacity expansion
Scaling the reduced environment after failover trades some recovery time for lower steady-state cost.