A global technology company operates a mission-critical application with a multi-tier architecture, including web servers, application servers, and a database. The application is hosted in a single AWS region, `us-east-1`. The company needs to implement a disaster recovery (DR) strategy that minimizes recovery time objective (RTO) to less than 15 minutes and recovery point objective (RPO) to less than 5 minutes, in case of a regional disaster. The DR solution must be cost-effective during normal operations. Which DR strategy is most appropriate?
- AWarm Standby in a secondary region with continuously synchronized data and pre-provisioned EC2 instances.
- BMulti-site Active/Active architecture across two regions with global routing.
- CPilot Light in a secondary region with a continuously replicated database and pre-provisioned minimal compute resources.
- DBackup and Restore with hourly backups to S3 and cross-region replication.
Show answer & explanationAnswer & explanation
Correct answer: A. Warm Standby in a secondary region with continuously synchronized data and pre-provisioned EC2 instances.
A Warm Standby strategy involves maintaining a scaled-down but fully functional duplicate of your production environment in a secondary region. With continuously synchronized data (e.g., using database replication) and pre-provisioned, but scaled-down, EC2 instances, this strategy can achieve RTOs of minutes and RPOs of minutes, while being more cost-effective than a multi-site active/active setup during normal operations.
Why the other options are wrong
- B. Multi-site Active/Active provides the lowest RTO/RPO (seconds/near zero) but is the most expensive strategy as it involves running a full production environment in multiple regions simultaneously, which contradicts the 'cost-effective during normal operations' requirement.
- C. Pilot Light maintains a minimal environment (e.g., just the database) in the DR region. While cost-effective, its RTO is generally higher than Warm Standby because significant compute resources need to be launched and scaled up during a disaster, which would exceed the 15-minute RTO.
- D. Backup and Restore typically has a higher RTO (hours to days) because it involves provisioning infrastructure and restoring data from backups, which is too slow for the required RTO.
Warm Standby DR
A disaster recovery strategy where a scaled-down but functional copy of the production environment is maintained in a secondary region, with continuously synchronized data.
- Achieves RTO in minutes and RPO in minutes.
- More cost-effective than Multi-site Active/Active.
- Requires some pre-provisioned compute resources in the DR region.
Memory trick: DR strategies are like fire drills: Hot is fully ready, Warm is partially ready, Pilot Light needs more setup, Backup is from scratch.