Instance Type Comparison Calculator

Work out instance type comparison instantly with clear inputs, formula shown and shareable results.

About Instance Type Comparison Calculator

Comparing instances on hourly rate alone is meaningless without performance. Cost per unit of performance — rate divided by a benchmark score representative of your workload — is the only fair comparison, and it frequently favours newer or ARM-based instances that cost more per hour but deliver disproportionately more throughput.

Formula

Price-performance

cost per performance unit = hourly rate / benchmark score; lower is better

Worked example

With the default values below, Instance Type Comparison Calculator returns Better price-performance of Instance B gives better price-performance. Change any input above and every figure here is recalculated in your browser.

Inputs used

Instance A hourly rateminimum 0: 0.192
Instance A benchmark scoreminimum 0.001: 100
Instance B hourly rateminimum 0: 0.154
Instance B benchmark scoreminimum 0.001: 115

Results

Better price-performance: Instance B gives better price-performance
Cost per performance unit, instance A: $0.001920
Cost per performance unit, instance B: $0.001339
Price-performance advantage: 30.3%

Step by step

1Values used
2Price-performance
3Better price-performance= Instance B gives better price-performance
4Cost per performance unit, instance A= 0.001920
5Cost per performance unit, instance B= 0.001339
6Price-performance advantage= 30.3

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 6 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

Which benchmark should be used?

Your own workload, measured. Synthetic scores mislead badly for memory-bound, I/O-bound or vectorised code, where instance families differ by far more than their nominal specifications.

Why do ARM instances often win?

One vCPU is a full physical core rather than a thread, and the price per vCPU is lower. Where the software stack supports ARM, price-performance gains of 20 to 40 percent are common.