Day 19 - Disaster Recovery on AWS

Date: 2025-10-02 (Thursday)
Status: “Done”

Lecture Notes

Disaster Recovery (DR) on AWS

Disaster Recovery is about restoring IT services after major incidents (outages, disasters, hardware failures, cyberattacks).

  • RTO (Recovery Time Objective): How quickly to restore service.
  • RPO (Recovery Point Objective): How much data loss (time window) is acceptable.

DR Strategies (ordered by complexity & cost)

Backup & Restore

  • Maintain backups only (EBS/RDS snapshots, S3/Glacier).
  • Restore to new infrastructure during incidents.
  • RTO: hours–days. RPO: depends on backup frequency. Cost: lowest.

Pilot Light

  • Minimal core services always running on AWS.
  • Scale out to full production during DR.
  • RTO: hours. RPO: minutes. Cost: moderate.

Warm Standby

  • Full system running at reduced scale on AWS.
  • Scale up on failover.
  • RTO: minutes–hours. RPO: seconds–minutes. Cost: higher.

Multi-Site (Active/Active or Active/Passive)

  • Production running across on-prem and AWS, or multi-Region AWS.
  • Traffic can be shifted instantly (Route 53, Global Accelerator).
  • RTO/RPO: near zero. Cost: highest.

DR Strategy Comparison:

Strategy RTO RPO Cost Complexity
Backup & Restore Hours-Days Hours $ Low
Pilot Light Hours Minutes $$ Medium
Warm Standby Minutes Seconds $$$ Medium-High
Multi-Site Seconds Near-zero $$$$ High

DR Best Practices

Planning

  • Define RTO and RPO requirements
  • Document recovery procedures
  • Identify critical systems and dependencies
  • Establish communication plans

Implementation

  • Automate recovery processes
  • Use multiple AZs and Regions
  • Implement data replication
  • Regular backup testing

Testing

  • Conduct DR drills regularly
  • Test recovery procedures
  • Measure actual RTO/RPO
  • Update documentation

Hands-On Labs

Lab 14 – AWS VM Import/Export (Part 2)

  1. Import Virtual Machine to AWS → 14-02.3
  2. Deploy Instance from AMI → 14-02.4
  3. Set Up S3 Bucket ACL → 14-03.1
  4. Export Virtual Machine from Instance → 14-03.2
  5. Resource Cleanup on AWS → 14-05