Skip to content

PATTERN Cited by 1 source

Backup and restore tier

Shape

DR tier where nothing runs in the recovery environment during steady state — only periodic backups are stored there. On disaster, provisioning + restoration + reconfiguration run from zero. The lowest tier of the DR ladder.

RPO / RTO profile

  • RPO: interval between backups — typically hours to days.
  • RTO: backup-restore + full-stack provisioning + configuration translation — typically hours to days.
  • Steady-state cost: near-zero beyond backup storage itself.

Canonical AWS realization

  1. Data backed up via AWS Backup into vaults, optionally cross-Region + cross-account into a clean-room account.
  2. Automation via Amazon EventBridge + AWS Lambda for restore orchestration — the canonical "Backup and Restore with Rapid Recovery" pattern (see prior AWS blog linked from the 2026-03-31 post).
  3. Infrastructure-as-code templates re-provision compute + networking + IAM in the recovery environment.
  4. Configuration translation — endpoints / credentials / identifiers mapped from old to new (see concepts/dr-config-translation).

When this tier is sufficient

  • RPO hours-to-days is acceptable (many batch-oriented systems, compliance-only data retention, some dev/test environments).
  • Cost sensitivity dominates — the recovery environment's storage cost is the only steady-state bill.
  • Infrequent, bounded recovery — the DR drill cadence can be annual or less.

When to move up the ladder

  • RPO needs to be sub-hour → move to pilot light with continuous replication.
  • RTO needs to be sub-hour → move to pilot light (IaC instantiation during incident) or warm standby (stack already running).
  • The workload is revenue-generating online traffic → almost always a higher tier.

Seen in

Last updated · 200 distilled / 1,178 read