What's changed: Deepened SOA-C02 Chapter 2 to Associate depth (tables, scenarios, FAQ, traps; localized figures)
2.3Backup and Disaster Recovery (RPO/RTO)
Understand EBS/RDS snapshots, centralized management with AWS Backup (schedules, cross-region/account copy), and choosing a DR strategy via RPO/RTO.
To guard against data loss and disasters, take backups on a plan and choose a DR strategy matching your recovery objectives (RPO/RTO).
2.3.1Backups and recovery objectives
- Snapshots: point-in-time backups of EBS/RDS, stored managed in S3.
- AWS Backup: schedules backups centrally across services, with cross-region/account copy.
- RPO: acceptable data-loss window. RTO: acceptable time to recover. These drive the DR strategy.
Common on SOA: RPO = how much data loss is acceptable, RTO = how fast you must recover, central cross-service backup = AWS Backup, replicate to another region for DR = cross-region copy.
Decide backup/DR in order: "objectives (RPO/RTO) → strategy → cost." RPO is "how far back you can recover = acceptable data loss"; RTO is "how fast you recover." Means include snapshots (EBS/RDS point-in-time, incremental, managed in S3) and AWS Backup for many services at once (backup plans for schedule/retention, cross-region/cross-account copy, and tamper-proof Backup Vault Lock). For region-wide failures, DR strategies span four tiers: backup & restore (cheap, slow) → pilot light (a minimal core on standby) → warm standby (a scaled-down stack always on) → multi-site/active-active (fast, costly). Shorter RPO/RTO means higher tier and higher cost. S3 cross-region replication and Route 53 failover routing (health-check to a standby region) are typical building blocks. Selecting targets by tag automatically includes new resources in the backup plan.
| DR strategy | RTO/RPO | Cost |
|---|---|---|
| Backup & restore | Long | Low |
| Pilot light | Medium | Low–medium |
| Warm standby | Short | Medium–high |
| Multi-site | Shortest | High |
Scenario: govern backups across all services and replicate to another region. Centralize schedule/retention via an AWS Backup plan and select targets by tag (new resources auto-included). Use cross-region copy for region-wide failures and Backup Vault Lock to prevent deletion. For a short RTO, choose warm standby + Route 53 failover.
Q. Acceptable-data-loss metric? RPO. Q. Recovery-time metric? RTO. Q. Central cross-service backup + cross-region? AWS Backup. Q. Cheapest DR? Backup & restore. Q. Fastest-recovery DR? Warm standby/multi-site.
Watch the mix-ups: (1) RPO = data-loss window, RTO = recovery time—frequently swapped. (2) EBS snapshots are managed in S3 (not visible in your S3 bucket). (3) Shorter RPO/RTO costs more—"always multi-site" is wrong (match the requirement).
Cost trades off against RPO/RTO: looser requirements suit backup & restore; stricter ones suit pilot light through multi-site.
2.3.2Section summary
- Snapshots / AWS Backup (central, cross-region)
- RPO (loss window) / RTO (recovery time) drive the DR strategy
Sign in to track progress — Log in.
Quick check
(just a quick review)Q1. You want centralized backups across services with schedules and cross-region copy. What do you use?
Q2. Which metric expresses "how much data loss is acceptable"?
Q3. Where are EBS point-in-time backups stored?