VocalHyre
Sign up

AWS, Azure & GCP · Cloud across providers

How do you design backup and disaster recovery?

Hardbackupdisaster recoveryrpoUpdated Oct 2026
Quiz this track

Short answer

Define RPO and RTO first, then choose backup, pilot light, warm standby or active-active. Test restores regularly, isolate backups from the primary account, and automate failover where the business needs it.

Backup and disaster recovery answer different questions. Backup protects against data loss and corruption; DR restores service after a large failure. Both start with business requirements: how much data can you lose, and how long can the service be down?

Strategies and trade-offs

  • Backup and restore: cheapest, slowest recovery, suitable for non-critical systems
  • Pilot light: core data and minimal services kept ready, scaled up during recovery
  • Warm standby: a smaller but functional copy running in another region
  • Active-active: full capacity in multiple regions, highest cost and complexity
  • RPO is tolerated data loss; RTO is tolerated downtime
bash
aws backup start-backup-job   --backup-vault-name prod-vault   --resource-arn arn:aws:rds:ap-south-1:123456789012:db:orders   --iam-role-arn arn:aws:iam::123456789012:role/backup
Start an on-demand backup through a central backup service.

Protect the backups themselves. Use a separate account or vault, immutable retention where available, encryption and cross-region copies. A compromised production account should not be able to delete the recovery path.

A backup you have never restored is a hypothesis. Schedule restore drills, measure the actual RTO, and record what broke. Replication alone does not protect against accidental deletion or ransomware.

Follow-ups you should expect

Interviewers use these to check whether the first answer was understanding or memorisation.

  1. 1How do RPO and RTO drive strategy choice?
  2. 2Why should backups live in a separate account?
  3. 3What does a restore drill need to prove?
  4. 4How does DR differ from high availability?

Say it in the interview

I start from RPO and RTO, because they determine whether backup and restore is enough or whether I need pilot light, warm standby or active-active. I isolate backups in a separate account, make them immutable where possible, and copy them cross-region. Most importantly, I run restore drills and measure the real recovery time, because an untested backup is only an assumption.

Use this as a model, not a script. Saying it in your own words is what makes it survive follow-up pressure.

Related questions

MediumHow do you build a highly available system in the cloud?HardHow do you design a multi-region architecture?EasyWhat is object storage and when should you use it?

Reading is step one

Now say it out loud.

Understanding an answer and delivering it under pressure are different skills. Rehearse this topic in a live AWS, Azure & GCP mock interview and get scored feedback on depth, clarity and confidence.

Start a free AWS, Azure & GCP interviewBrowse all questions