High Knowledge · RZ-KPI-Lab
Data centre metrics.
Understand. Calculate. Put into context
Compare lifetime costs and calculate availability, rack density, PUE, water use and heat reuse with your own assumptions. The Data Centre KPI Lab makes calculation methods and measurement boundaries transparent, supporting planning, monitoring and discussions with your operations team.
DATA CENTRE METRICS · AVAILABILITY · POWER DENSITY
Calculate availability and rack power.
Use a few assumptions to calculate annual downtime, average rack density and the total power demand of your scenario. Results provide an initial plausibility check, not guaranteed project figures.
Rack density here is the average derived from IT power and rack count. Individual high-density racks can differ significantly. Connection capacity, redundancy and reserves must be assessed separately.
ENERGY · WATER · WASTE HEAT
PUE, WUE and ERF from annual figures.
Use figures for the same reporting period and consistent measurement boundaries. Water usage is shown in m³/MWh, numerically equivalent to litres per kWh. Heat recovery within the data centre boundary does not count as external reuse here.
A low PUE alone does not assess productive IT utilisation or the overall environmental impact. Water demand and heat actually reused provide additional context. These metrics do not establish certification or compliance with legal requirements.
Formulas, measurement boundaries and source
PUE = total energy ÷ IT energy. WUE = water inflow in m³ ÷ IT energy in MWh. ERF = externally reused energy ÷ total energy, also displayed as a percentage. Infrastructure energy = total energy − IT energy.
Die Grundformeln entsprechen Anhang IV der Commission Delegated Regulation (EU) 2024/1364. The recording and reporting required for your data centre depend, among other factors, on its size and the applicable scope. This calculator does not constitute a complete regulatory report.
Investment · operations · option comparison
What will your data centre cost over its lifetime?
Compare two options with the same IT requirements: investment, energy, ongoing operation and selected replacement investments. The calculation shows annual cash flows and present value over your chosen period.
All prefilled figures are illustrative assumptions, not market prices, benchmarks or quotations. Replace them with project-specific figures. The calculator does not estimate construction costs from floor area or availability class.
A positive cost advantage means B costs less in this model. A negative value means higher costs. Availability, risk, quality and expansion options are not monetised.
| Metric | Option A | Option B |
|---|
How do your assumptions affect costs?
The comparison shows how different electricity prices, discount rates or a higher PUE affect the cost advantage. Only one assumption changes at a time. These scenarios are not price forecasts.
| Changed assumption | Cost advantage of B |
|---|
View annual cash flows
| Year | Cash flow A (€) | Cash flow B (€) | Present value A (€) | Present value B (€) |
|---|
Method, scope and technical basis
Initial investment = (building + technical systems + design/other capital costs + IT) × (1 + contingency/100). Annual energy = average IT load × operating hours × PUE. Electricity cost in year t = annual energy × initial electricity price × (1 + annual price change/100)^(t−1). Other operating costs are escalated in the same way.
Present value = initial investment + sum of annual net cash flows/(1 + discount rate/100)^t. Recurring cash flows are assumed to occur at year end. Replacement is included once in the chosen year; residual value is deducted at the end. Both inputs are already nominal euro amounts for the relevant year and are not escalated again.
The model assumes constant average IT load and PUE; it does not model phased expansion. Financing, taxes, depreciation, outage costs, decommissioning and heat revenue are not modelled separately. Include additional costs in the appropriate input item. This is not a DIN 276 cost estimate or proof of regulatory or standards compliance.
Background on present-value analysis: NIST Handbook 135 (2025). US-Vorgabewerte werden nicht übernommen. Sämtliche Preise und Zinssätze bestimmen Sie selbst.
For a robust investment decision, we clarify load profiles, measurement boundaries and the actual scope in our feasibility study and develop the selected concept through Data centre design.
Modellstand: 30. September 2026. Die Berechnung erfolgt im Browser. Nur beim angeforderten CSV-Versand werden Kontaktdaten und Berechnungsannahmen an High Knowledge übermittelt. CSV-Dateien enthalten Ihre Projektdaten.
FROM MEASUREMENT TO IMPROVEMENT
Useful metrics need a sound measurement approach.
01 · Consistent boundaries
Document measurement points, energy sources and time periods. Meaningful comparisons require comparable data.
02 · Actual load profiles
Consider annual averages alongside peak loads and periods of partial load. Power supply and cooling must also cover operational extremes.
03 · Track the effects
Use monitoring to assess improvements. Temperature, IT utilisation and heat users affect the results.
Turn measurements into effective improvements.
High Knowledge brings data centre planning, energy and cooling design together with monitoring and operations. We help you define measurement boundaries, assess metrics and plan suitable improvements.
Start a project enquiry → · Operate and monitor your data centre · Data centre and edge AI capacity calculator
Ihre Eingaben bleiben für die Berechnung im Browser. Nur beim angeforderten CSV-Versand werden Ihre Kontaktdaten und Berechnungsannahmen an High Knowledge übermittelt. Verfügbarkeit wird mit einem 365-Tage-Jahr berechnet.