CUE in Data Center

Carbon Usage Effectiveness (CUE) describes how many CO₂ equivalents are caused by the overall operation of a data centre, taking into account the energy source and the emissions factor of the electricity mix, relative to the energy consumption of the IT infrastructure (servers, storage and network equipment).

CUE is typically expressed in kg CO₂e per kWh of IT energy.

Short definition: CUE quantifies the CO₂ emissions generated per unit of IT energy in a data centre.

How CUE Works

CUE brings together two dimensions that are often assessed separately: energy efficiency and the carbon intensity of the electricity supply.

While metrics such as PUE measure how much total energy is required, CUE shows how carbon-intensive that energy is — depending on the grid mix, cooling demand and operational control.

In practice, CUE depends on three key factors:

• Total energy consumption of the data centre
• The share of IT load within that energy consumption
• The emissions factor of the electricity used

Because cooling demand and grid carbon intensity change throughout the day and across seasons, CUE is a time-dependent metric, not a fixed value.

Key Metrics and Components

A meaningful CUE assessment requires a clear understanding of energy flows within the data centre:

• Total electrical energy consumption
• IT load delivering the actual compute output
• Carbon intensity of the electricity supply

More advanced approaches also consider:

• time-based emissions factors
• regional differences in electricity generation mixes
• contributions from on-site generation and renewable energy

Relevance for Energy Efficiency and Cooling

Cooling often accounts for a significant share of non-IT energy consumption in data centres and therefore has a direct impact on CUE.

Even with highly efficient IT equipment, oversized or poorly controlled cooling systems can significantly worsen the carbon footprint.

With targeted cooling strategies — such as shifting loads to periods with lower carbon intensity or applying adaptive setpoints — both energy use and emissions can be reduced. This makes CUE a practical indicator for linking operational control to decarbonisation goals.

Typical Challenges

A common weakness is the use of annual average emissions factors, which can hide daily or hourly fluctuations in grid carbon intensity and lead to misleading conclusions about real carbon performance.

Other typical pitfalls include:

• treating CUE as a static KPI
• optimising PUE without considering carbon impact
• insufficient transparency on cooling and facility energy consumption

Best Practices

CUE delivers the most value when used as an operational control metric — not only as a reporting figure. Analysing time-resolved energy and carbon data reveals when and why emissions are particularly high.

Recommended best practices include:

• separating cooling energy from IT energy
• combining CUE with PUE, WUE and other efficiency KPIs
• continuously optimising control strategies to reduce emissions without compromising availability or performance

Example: CUE in Data Centre Cooling

A data centre may operate with a relatively stable IT load while cooling demand fluctuates throughout the day. During periods of high outdoor temperatures and a carbon-intensive electricity supply, cooling energy consumption increases — and so does the CUE value.

By adjusting setpoints and implementing intelligent control logic, energy consumption can be reduced during these critical periods while still meeting the required cooling demand. The resulting improvement in CUE is achieved through operational optimisation rather than expensive infrastructure upgrades.

‍