Free planner · No account required
Capacity and failover planner
Check whether measured capacity can absorb forecast demand when instances are unavailable.
Your inputs stay in this browser tab. Saving to this device is optional.
Draft name and device saving
Current draft matches the last save or initial state.
Complete your inputs
Calculated result
Awaiting complete inputsEnter your values to see the result, or choose Try an example to explore a completed worksheet.
How this result is calculated
Forecast demand = current peak × (1 + growth percentage). Usable capacity keeps your chosen reserve unused: instances × tested sustainable capacity × (1 − reserve percentage). The failure scenario removes the specified instances before checking demand. Required fleet size rounds forecast demand divided by usable per-instance capacity up, then adds the instances you need to survive losing.
Keep your work
Resume later
Save an editable JSON draft. Use Open draft to bring it back into this tool.
Share a report
Reports include your inputs, assumptions and any unresolved checks.
Read the complete report
# Capacity and failover planner Tool defaults; replace with your settings before operational use. ## Review issues - Peak demand: enter a number from 0 to 1000000000000. - Tested capacity per instance: enter a number from 0.001 to 1000000000000. ## Inputs Service and capacity evidence: Not entered Current peak requests per second: Not entered Forecast growth (%): 20 Healthy instances: 3 Tested sustainable requests/sec per instance: Not entered Instances unavailable in the scenario: 1 Capacity to keep unused (%): 20 ## Method Forecast demand = current peak × (1 + growth percentage). Usable capacity keeps your chosen reserve unused: instances × tested sustainable capacity × (1 − reserve percentage). The failure scenario removes the specified instances before checking demand. Required fleet size rounds forecast demand divided by usable per-instance capacity up, then adds the instances you need to survive losing. This model assumes equal instances, linear scaling and balanced traffic. It does not capture shared databases, hot partitions, autoscaling delay or entire-zone placement. Sustainable capacity must come from a representative test; a CPU percentage or vendor maximum is not interchangeable with measured throughput. https://aiopssre.com/capacity-headroom-calculator/
How it works
Forecast demand = current peak × (1 + growth percentage). Usable capacity keeps your chosen reserve unused: instances × tested sustainable capacity × (1 − reserve percentage). The failure scenario removes the specified instances before checking demand. Required fleet size rounds forecast demand divided by usable per-instance capacity up, then adds the instances you need to survive losing.
Follow the example
The example forecasts 1,080 requests per second. Three instances provide 1,440 requests/sec after the 20% reserve. Losing one leaves only 960, so the degraded scenario has a 120 requests/sec shortfall even though the normal fleet has spare capacity. Four instances meet the chosen reserve after one failure.
What to account for
This model assumes equal instances, linear scaling and balanced traffic. It does not capture shared databases, hot partitions, autoscaling delay or entire-zone placement. Sustainable capacity must come from a representative test; a CPU percentage or vendor maximum is not interchangeable with measured throughput.