Server Count For Load Calculator
Work out server count for load instantly with clear inputs, formula shown and shareable results.
About Server Count For Load Calculator
Capacity alone needs peak divided by per-server capacity at the target utilisation. Surviving the loss of one availability zone requires the remaining zones to carry the whole load, which multiplies the count by z/(z-1): 50 percent overhead across two zones, but only 33 percent across three and 25 percent across four. That is the main capacity argument for three zones over two.
Formula
Server count
base = ceil(peak / (per-server capacity x utilisation)); with zone loss = ceil(base x zones / (zones - 1)); per zone = ceil(total / zones)
Worked example
With the default values below, Server Count For Load Calculator returns Servers to provision of 18. Change any input above and every figure here is recalculated in your browser.
Inputs used
- Peak request rateminimum 0.01 req/s
- 4,200 req/s
- Capacity per serverminimum 0.01 req/s
- 600 req/s
- Target utilisationaccepts 1–100 %
- 65 %
- Availability zonesaccepts 1–10
- 3
Results
- Servers to provision
- 18
- Servers for capacity alone
- 11
- Servers per availability zone
- 6
- Overhead for zone redundancy
- 63.6%
Step by step
- 1Values used
- 2Server count
- 3Servers to provision= 18
- 4Servers for capacity alone= 11
- 5Servers per availability zone= 6
- 6Overhead for zone redundancy= 63.6
Assumptions and limits
- The result is derived only from the 4 inputs listed above. Anything not among them is not part of the calculation.
- In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
- Figures are rounded for display to 1 decimal place, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.
How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.
Frequently Asked Questions
Why do three zones cost less overhead than two?
Losing one of three zones removes a third of capacity, so the survivors need 50 percent more each. Losing one of two removes half, so survivors need to double.
Should autoscaling replace this calculation?
Autoscaling handles growth but not instantaneous zone loss, since scaling takes minutes. The redundant capacity must already be running.
Related Calculators
You might also need
- Paging Load CalculatorAlso in DevOps, SRE & Performance
- Partition Count CalculatorAlso in DevOps, SRE & Performance
- IaC Resource Count CalculatorAlso in DevOps, SRE & Performance
- Infrastructure Drift Count CalculatorAlso in DevOps, SRE & Performance
- Load Test Concurrency CalculatorAlso in DevOps, SRE & Performance
- Peak Load Forecast CalculatorAlso in DevOps, SRE & Performance
