SAA-C03
Deep Dive
The exam presents scenarios with specific RTO/RPO requirements and asks you to choose the correct DR strategy. You need to understand the trade-off between cost and recovery time for each strategy.
Recovery Time Objective
The maximum acceptable time to restore service after a disaster.
"The application must be operational within 5 minutes or less after a failure."
Recovery Point Objective
The maximum amount of data you can afford to lose, expressed as a time window since the last backup/replica.
"We cannot lose more than 1 hour of transactions."
Golden rule
The lower the RTO and RPO, the higher the cost. There's an inversely proportional relationship. The exam will give you RTO/RPO requirements and you must choose the cheapest strategy that meets them.
| Strategy | RTO | RPO | Cost | How it works |
|---|---|---|---|---|
| Backup & Restore | Hours | Hours | $ | Backups to S3/Glacier. Restore from scratch during a disaster. |
| Pilot Light | 10-30 min | Minutes | $$ | Core (replicated DB) always on. Scale up on disaster. |
| Warm Standby | 1-5 min | Seconds | $$$ | Full infrastructure at reduced scale in DR. Scale up on disaster. |
| Multi-Site Active/Active | Seconds | ≈ 0 | $$$$ | Two active regions. Route 53 redirects if one fails. |
RDS Cross-Region Read Replica
Strategy: Pilot Light / Warm Standby
Creates a Read Replica in another region. During a disaster: promote the replica to Primary (minutes). RPO = replication lag time (seconds to minutes).
Aurora Global Database
Strategy: Warm Standby / Multi-Site
Cross-region replication with <1 second latency. Promotion in <1 minute. Up to 5 secondary regions. The most advanced option for Aurora.
DynamoDB Global Tables
Strategy: Multi-Site Active/Active
Active tables in multiple regions. Writes replicated automatically. RPO ≈ 0. Reads and writes from any region.
S3 Cross-Region Replication
Strategy: All tiers
Automatically replicates S3 objects to another region. For assets, backups and data lakes. Combine with CloudFront to serve from the nearest region.
Multi-AZ = High Availability (HA)
Multi-Region = Disaster Recovery (DR)
Exam trap
Multi-AZ is NOT DR across regions. If the exam mentions "an entire region goes down" or "regional disaster", the solution is Multi-Region. If it mentions "an AZ fails" or "high availability", the solution is Multi-AZ.
The 4 strategies with RTO, RPO, relative cost, and key AWS services.
Did you get this topic?
Test what you just learned
A financial services company needs an RTO of 5 minutes and an RPO of 30 seconds for its payment processing system. Which DR strategy is most appropriate considering cost?
Put what you learned to the test with original SAA-C03 exam questions.
Start SAA-C03 mock examUnlimited mock exams with the Pro plan
Log in to track your progress.