Disaster recovery (DR) comprises all plans, techniques and processes for restoring IT systems after a severe outage — after ransomware, data center failure, natural events or large-scale error. The core is the DR plan: documented recovery sequences, owners, communication paths and the technical standby environment.
DR is the technical counterpart to organisational business continuity management: BCM keeps the business running, DR brings IT back. Both are measured by two metrics — RTO (how fast) and RPO (how much data loss is tolerable).
DR strategies: from backup-restore to hot site
The strategies differ in speed and cost: backup and restore is the baseline (hours to days), pilot light keeps a minimal core environment permanently ready, warm standby a scaled-down running copy, hot site or multi-site operation takes over almost seamlessly. The more critical the process, the hotter the strategy — and the more expensive.
Cloud DR has shifted the economics: standby capacity no longer needs to be permanently maintained but can be provisioned in an emergency — disaster recovery as a service (DRaaS) turns this into a managed service. The prerequisite remains a network design that carries the switchover: DNS failover, replicated networks, reachable access for the workforce.
What makes a DR plan viable
- Prioritised recovery order by business criticality — including dependencies (AD, DNS, network first).
- Tested procedures: DR exercises at least annually, updated after every major change.
- Access in the emergency: out-of-band access in case VPN, IdP or firewall are affected themselves.
- Plan the ransomware scenario separately: restore into a clean environment, do not destroy forensics.