ExamPark
Course menuSOA-C03

Exams / SOA-C03 / Study guide / Disaster recovery strategies

Disaster recovery strategies: SOA-C03 study notes and practice questions

What to know about Disaster recovery strategies for the SOA-C03 exam, from the Reliability and Business Continuity domain (22% of the exam guide), then 1 free question to test it. Part of the SOA-C03 study guide.

Last verified against current AWS documentation ·How we write our questions

What to know for the exam

DR questions give recovery targets and ask which strategy or service meets them at the lowest cost.

Strategies

Strategy In the recovery Region RPO / RTO Cost
Backup and restore Copies of backups only Hours / hours Lowest
Pilot light Data replicating continuously; servers off or not yet created Minutes / tens of minutes Low
Warm standby A smaller, fully working copy Seconds to minutes / minutes Higher
Multi-site active/active Full production serving traffic Near zero / near zero Highest

Services

  • AWS Backup cross-Region and cross-account copy rules.
  • AWS Elastic Disaster Recovery continuously replicates servers into a low-cost staging area and launches recovery instances in minutes; good for pilot light from on premises or another Region. Supports non-disruptive drills.
  • Aurora Global Database: secondary Regions typically under a second behind; switchover (planned, no data loss) or failover (unplanned) to a secondary.
  • DynamoDB global tables: active-active in several Regions.
  • S3 Cross-Region Replication (with Replication Time Control for a 15-minute SLA).
  • RDS cross-Region read replicas and cross-Region automated backups.
  • CloudFormation / StackSets to create the recovery stack quickly; AMIs copied to the recovery Region.
  • Route 53 failover routing with health checks, or Global Accelerator, to move users.
  • Amazon Application Recovery Controller (ARC) for readiness checks and routing controls that shift traffic reliably during an event.

Operating DR

  • Test regularly. An untested plan is a guess.
  • Automate failover steps with runbooks.
  • Watch replication lag metrics so you know your real RPO.

Common traps

  • Active/active for a 24-hour RTO. Too expensive.
  • Backup and restore for a 5-minute RPO. Needs continuous replication.
  • Forgetting dependencies (secrets, AMIs, DNS, IAM) in the recovery Region.

Try the 1 practice question

Practice Disaster recovery strategiesTake the free practice exam

Question 1

Harder ·Link to this question (question 1)

A solar installer runs its customer portal on EC2 instances and an Amazon Aurora PostgreSQL database in one Region.

Its new disaster recovery requirements are an RPO of under one minute and an RTO of under one hour if the Region fails.

The infrastructure is defined in CloudFormation templates. The finance team wants the LOWEST ongoing cost that meets both targets.

Which strategy should a CloudOps engineer implement?

Choose one answer.

  1. A. Back up the database and EC2 volumes with AWS Backup, copy recovery points to the second Region every night, and restore them after a disaster.
  2. B. Run a full-size copy of the portal in the second Region behind Route 53 latency routing, with both Regions serving users all the time.
  3. C. Add the second Region to an Aurora global database, keep application servers off there, and launch them from the templates during a failover.
  4. D. Run a scaled-down but working copy of the portal in the second Region that receives no traffic, and scale it up during a failover.
Show the answer and explanation

Answer: C

Explanation

Disaster recovery strategies trade cost against recovery speed. Pilot light keeps the data layer replicating continuously and the rest of the stack switched off, ready to launch from infrastructure as code, which suits a short RPO and an RTO of tens of minutes. Backup and restore is too slow, and warm standby and active/active cost more than the targets require.

An Aurora global database across two Regions In the primary Region, the application writes to the primary Aurora cluster. Storage-level replication copies the data to a read-only secondary cluster in another Region, usually within about a second. In a failover the secondary is promoted and the application runs there. Primary Region Secondary Region Application writes here Application off or scaled down Primary cluster reads and writes Secondary cluster promote on failover storage replication, usually about 1 second
The second Region holds a live copy of the data, while application servers launch there only during a failover.

Study

Want more on Disaster recovery strategies? The 5 full-length practice sets have 5 more Disaster recovery strategies questions, 325 in all, none of them in the free exam. One payment of $14.99, no subscription. See the practice sets

Study Disaster recovery strategies

More Reliability and Business Continuity topics