Test whether a backup, pilot light, warm standby or active-active design actually meets your RTO and RPO, and what it costs.
Each recovery tier is modelled as a fraction of production capacity kept warm, which is what actually drives the standby bill, plus the replication and backup meters that run continuously. The achievable RTO for a backup-and-restore design is dominated by physics rather than price: moving 40 TB at 2 Gbps takes over 40 hours no matter how the runbook is written. Most DR plans fail their own RTO because nobody converted the dataset size and restore bandwidth into hours, and the gap only appears during a real incident. Comparing the annual DR cost with the expected outage loss shows whether the next tier up is worth buying; these are planning estimates, not an insurance calculation.
Cloud Disaster Recovery
DR cost = production cost × standby fraction × region premium + replicated GB × transfer rate + backup GB × backup rate; restore hours = backup GB × 8 ÷ (Gbps × 3,600).
DR cost = production cost × standby fraction × region premium + replicated GB × transfer rate + backup GB × backup rate; restore hours = backup GB × 8 ÷ (Gbps × 3,600). Each recovery tier is modelled as a fraction of production capacity kept warm, which is what actually drives the standby bill, plus the replication and backup meters that run continuously. The achievable RTO for a backup-and-restore design is dominated by physics rather than price: moving 40 TB at 2 Gbps takes over 40 hours no matter how the runbook is written.
Most DR plans fail their own RTO because nobody converted the dataset size and restore bandwidth into hours, and the gap only appears during a real incident. Comparing the annual DR cost with the expected outage loss shows whether the next tier up is worth buying; these are planning estimates, not an insurance calculation.
This calculator takes 13 inputs: Recovery strategy, Production environment cost, Price premium in the recovery region, Data replicated between regions, Inter-region transfer rate, Backup data held, Backup storage rate, Backup interval, Restore throughput, RTO target, RPO target, Cost of an hour of outage, Expected invocations per year. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.
The pilot light keeps the core infrastructure, databases and network already built and replicating, so recovery is a scale-up rather than a rebuild. Only a fraction of the data ever needs restoring, which is why the restore term is discounted rather than removed.