What is Power Usage Effectiveness (PUE) & Water Usage Effectiveness (WUE)?
PUE and WUE indicate how much of your power and water budget is consumed by non-IT infrastructure — especially cooling.
Power Usage Effectiveness (PUE) is a metric used to measure the energy efficiency of a data center. It is calculated as the ratio of total facility energy consumption to the energy used by IT equipment.
A lower PUE indicates a more energy-efficient data center, with less energy wasted on cooling, lighting, and other overhead.
- PUE = Total Facility Energy ÷ IT Equipment Energy
- A PUE of 1.0 means all energy goes to IT equipment, the ideal scenario.
Water Usage Effectiveness (WUE) is a metric that measures how efficiently a data center uses water for cooling. It is calculated as the ratio of annual water usage to the energy consumed by IT equipment.
- WUE = Annual Water Usage ÷ IT Equipment Energy
- Annual Water Usage = all water consumed for cooling (evaporative cooling towers, humidification, etc.)
- IT Equipment Energy = energy consumed by servers, storage, and networking gear
- A lower WUE is better, it means the data center is using water efficiently.
Why is Optimising Data Center Cooling Essential for Improving PUE and WUE?
- Cooling Is the largest non-IT energy load in data centers.
- In tropical climates, high year-round temperatures increase the cooling required to keep IT servers within safe operating limits.
- Therefore, optimising data center cooling will have a direct impact on PUE and WUE.
- The increase in demand for cloud services, 5G expansion, and AI adoption lead to increase in server demand.
- This leads to higher heat generation, requiring additional power for cooling to prevent IT equipment from overheating.
- For safety reasons, large data centers usually have additional cooling capacity - 1.2 to 1.3 times of the IT load.
- This means that as IT load increases, insufficient cooling capacity or efficiency causes cooling energy to rise faster than IT energy, leading to a higher PUE (and perhaps WUE, depending on the data center’s cooling system design).
Issues Influencing Data Center Cooling Performance
& possible strategies to tackle them
With the rapid growth in data demand, data center operators are under increasing pressure to expand and optimise their cooling infrastructure. These challenges are multifaceted, spanning capacity scalability, energy efficiency, reliability, and sustainability, as outlined below.
In response, ebm‑papst provides targeted strategies to address these data center cooling challenges. Backed by decades of focused R&D, a global network of offices, and deep expertise in energy-efficient ventilation, we deliver solutions that optimise cooling performance, reduce energy consumption, and measurably improve PUE and WUE.
Space Constraints in High-Density Racks
Racks are becoming increasingly dense due to rising data demand, concentrating more heat in smaller spaces. This requires more powerful in-rack cooling to maintain stable temperatures, and without it, energy and water use rise, ultimately worsening both PUE and WUE.
This challenge can be addressed by...
Outdated Cooling Equipment
Legacy cooling systems are significantly less energy-efficient than modern solutions. They often operate at constant speeds, wasting energy during low-load conditions and delivering imprecise temperature and airflow control. This leads to over- or uneven cooling, forcing systems to run longer to maintain set conditions, ultimately increasing energy consumption, worsening PUE and WUE, and driving up operating costs.
This challenge can be addressed by...
Harmonics
Harmonics are an unavoidable by-product of non-linear loads commonly found in data center cooling systems. They introduce electrical losses and excess heat, forcing cooling equipment to consume more energy and water, ultimately worsening both PUE and WUE.
This challenge can be addressed by...
Complex, Interdependent Cooling Systems
Data center cooling systems have complex topologies. They are distributed across a large physical and control space, feature diverse interdependencies, and dynamically influence one another. These factors can create cascading effects, making it difficult to identify the efficient setpoints for the equipment involved. Consequently, energy and water efficiency, reflected in PUE and WUE, is negatively impacted.