Peak Load Forecast Calculator

Work out peak load forecast instantly with clear inputs, formula shown and shareable results.

About Peak Load Forecast Calculator

Capacity must be sized for peak, not average, so the peak-to-average ratio is as important as the growth rate. Consumer services commonly run 3 to 5 times average at peak, and internal tools much higher because usage clusters in working hours. Compounding the average and then applying the ratio keeps the two factors separate and visible.

Formula

Peak forecast

future average = current average x (1 + annual growth)^(months/12); forecast peak = future average x peak-to-average ratio

Worked example

With the default values below, Peak Load Forecast Calculator returns Forecast peak load of 3,712. Change any input above and every figure here is recalculated in your browser.

Inputs used

Current average loadminimum 0.01 req/s: 800 req/s
Peak to average ratioaccepts 1–50 x: 3.2 x
Annual growthaccepts 0–1,000 %: 45 %
Forecast horizonaccepts 1–60 months: 12 months

Results

Forecast peak load: 3,712 req/s
Forecast average load: 1,160 req/s
Current peak load: 2,560 req/s
Growth factor over the horizon: 1.450 x

Step by step

1Values used
2Peak forecast
3Forecast peak load= 3,712 req/s
4Forecast average load= 1,160 req/s
5Current peak load= 2,560 req/s
6Growth factor over the horizon= 1.450 x

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 or tax into account.
Figures are rounded for display to 3 decimal places, 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

Does the peak-to-average ratio stay constant as traffic grows?

It usually falls slightly as the user base spreads across time zones, so assuming it is constant is mildly conservative — which is the safe direction.

Should marketing events be modelled separately?

Yes. A launch or campaign can produce a spike well beyond the organic ratio, and it should be planned as a discrete event with its own capacity reservation.