What is Power Usage Effectiveness (PUE)?

Power Usage Effectiveness, or PUE, is a metric used to determine the energy efficiency of a data center. Introduced in 2007 and endorsed by The Green Grid, PUE has become a global standard for data center energy management. Essentially, PUE helps data centers optimize their energy consumption by comparing the total facility energy to the IT equipment energy usage.

How to Calculate PUE in Data Centers

To calculate PUE at your data center, you need to know:

1. The IT load: This includes the energy consumed by all IT equipment, typically measured at power distribution units (PDUs).
2. Total facility energy consumption: This encompasses all energy used at the facility, including cooling systems, lighting, and uninterruptible power supplies, usually measured from the utility metre.

The formula for PUE is: Total amount of Energy used / IT equipment energy usage = PUE data center ratio.

For instance, if a data center uses 50,000 kWh of total energy and 40,000 kWh of that energy is used by IT equipment, the PUE would be 1.25. A lower PUE indicates better power usage effectiveness, making it a critical metric for assessing how sustainable and efficient a data center is.

What is a Good PUE?

A good PUE is one that is close to 1.0. The closer the PUE value is to 1, the more efficient a data center is in utilizing its power solely for IT equipment, without wasting much on other infrastructures. According to the Uptime Institute, the average PUE was about 1.55 in 2022, showing a trend towards more efficient energy use in data centers over the years.

What is DCiE?

Data Center Infrastructure Efficiency (DCiE) provides a different perspective on a data center's energy usage compared to PUE. While PUE is expressed as a ratio, DCiE is presented as a percentage, which represents the IT equipment's energy consumption as a part of the total facility power usage.

Why is DCiE important?

Data centers use a lot of energy. This places a responsibility on data center providers to use less energy where possible. The widespread adoption of real-time business processes, machine learning, and high-speed connectivity means existing energy consumption could increase in the future.

So, with data centers accounting for 1 per cent of the world’s energy consumption, it’s critical to make improvements wherever possible. Both PUE and DCiE are vital tools in tackling unnecessary energy consumption.

How to calculate DCiE

To calculate DCiE, you reverse the formula used for PUE. Specifically, you divide the energy consumed by IT equipment by the total energy used by the facility (IT equipment energy / total facility usage = DCiE). For example, if a data center uses 50,000 kWh of total energy and 40,000 kWh of that is consumed by IT equipment, the DCiE would be calculated as 40,000kWh/50,000kWh x 100 = 80%). This percentage indicates the proportion of the total energy that is effectively used for IT operations, providing a clear metric to assess infrastructure efficiency.

Including DCiE alongside PUE provides a comprehensive view of a data center's energy efficiency, offering different perspectives on how effectively energy is being used relative to IT equipment. This dual approach allows data centers to monitor and improve their energy usage more effectively.

Why is PUE important?

Power Usage Effectiveness (PUE) is an important metric for several reasons, primarily because it measures the energy efficiency of a data center. By calculating PUE, data centers can assess how much of the energy they consume is used for computing functions versus how much is used for other infrastructure needs like cooling and lighting. This distinction is crucial as it helps identify areas where energy is not being used effectively, allowing them to make targeted improvements.

Reducing PUE not only helps in cutting down energy costs but also contributes to environmental sustainability by minimising unnecessary energy use and associated emissions. Furthermore, a lower PUE can enhance a data center's operational efficiency and reliability, providing better services to customers and gaining a competitive advantage in the market. In an industry where energy consumption is significant, having a low PUE is a marker of operational excellence and environmental responsibility.

How to use PUE

Data center management should regularly measure their PUE. Energy usage can vary with the time of day and season – taking regular measurements helps overcome these fluctuations. Over time, these results show how companies have improved and create future data center benchmarks.

According to the Uptime Institute, the average PUE value was 1.55 in 2022, meaning data centers used around 60 per cent of their energy consumption on IT equipment. This is a slight disimprovement on 2021’s score of 1.57 – but a significant gain from the 2007 average score of 2.5. By collecting PUE over time, data centers can prove their efforts to reducing PUE.

How to reduce PUE

Improving PUE involves several strategies:

• Cold Aisle Containment: Implementing cold aisle containment can significantly reduce the PUE by optimizing airflow and eliminating bypass air.
• Enhanced Cooling Technologies: Upgrading cooling systems to more efficient models can lower energy consumption, thereby improving the PUE.
• Regular Measurements: Continuous monitoring of PUE helps identify trends, make informed decisions, and benchmark against past performance.
  

Why Reducing PUE is Important

PUE and DCiE demonstrate how efficiently a data center uses energy. By understanding the amount of energy spent on different processes, companies can assess how to make improvements that save money, improve service, and reduce waste. Our colocation data centers are designed with efficiency in mind.

Conclusion

By understanding the amount of energy spent on different processes, companies can assess how to make improvements that save money, improve service, and reduce waste. Our colocation data centers are designed with sustainability in mind. Contact us today to find out more.