Subscribe to the Data Center Knowledge Newsletter
Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily.
What Is Data Center PUE? Defining Power Usage Effectiveness
PUE measures the energy efficiency of data centers. Here's why power usage effectiveness is important and how to reduce your data center's PUE.
Antonella Gina Fleitas
June 22, 2023
11 Min Read
Alamy
Table of Contents:
Data and energy are leading topics in today's conversations. With approximately 3.5 quintillion bytes (3.5 million terabytes) of data created daily and approximately 57% of companies changing their machinery and processes to more sustainable options, it's not hard to understand why these two words hold a valuable space in many current conversations across industries, niches, and even academic spaces. And their relevance isn't the only thing they have in common: There's a lot to understand about how managing the increasing amount of data that we create every day consumes energy and, consequently, requires its own sustainable approach.
PUE, which stands for power usage effectiveness, is a metric that helps understand how data centers are spending their energy and helps recognize any opportunity to improve this over time. In this article, we will explore the main concept behind PUE, its components, and what makes it so important in the data management space.
A Brief Introduction to PUE
PUE is a fundamental metric that measures the energy efficiency of dedicated data centers. The concept first appeared back in 2007 when The Green Grid published the book "PUE: A Comprehensive Examination of the Metric," in which it explained what power usage effectiveness is and how to calculate it.
Back then, web 2.0 was just starting and the data space took the first steps to becoming as complex as it is today. Gathering data was easier as time went by, but there was a new challenge: managing and storing all this information without damaging the world we live in due to the amount of energy required to do these tasks. This is where PUE came in.
Components of PUE
Power usage effectiveness is the ratio between the total energy amount a facility consumes and the energy specifically used by the IT equipment. However, to understand what that shows, we need to take a look at its components.
IT Equipment Power
This component of PUE focuses on the power consumed by the core IT equipment within the data center, including servers, switches, storage devices, and networking infrastructure. It encompasses the energy required for data processing, computation, and transmission.
Cooling Infrastructure Power
Data centers generate substantial heat due to the operational intensity of IT equipment. To maintain optimal temperatures and prevent equipment from overheating, cooling systems such as computer room air conditioning (CRAC) units, chillers, fans, and pumps are employed. The power consumed by these cooling mechanisms plays a crucial role in the overall PUE assessment.
Lighting and Miscellaneous Power
While individual lighting fixtures and miscellaneous electrical loads may seem insignificant, their cumulative energy consumption can impact the overall PUE significantly. This component encompasses the power used by lighting systems, security equipment, and other miscellaneous electrical devices present in the data center.
Uninterruptible Power Supply (UPS) Losses
UPS systems provide backup power during utility outages to ensure uninterrupted operations. However, the UPS units themselves introduce inefficiencies during power conversion and conditioning processes, resulting in power losses. These losses occur during charging, discharging, and maintaining batteries and are factored into the PUE calculation.
Power Distribution Losses
This last component of PUE refers to the power distribution infrastructure, including transformers, switchgear, power distribution units (PDUs), and cabling. Each of these components incurs electrical resistance and associated inefficiencies, leading to power losses during the transmission of electricity from the utility source to the IT equipment. These losses are taken into account when calculating PUE.
How Do You Calculate PUE?
The formula used to calculate power usage (PUE = Total Facility Energy / IT Equipment Energy) considers two factors: the total facility energy and the IT equipment energy.
Total facility power includes everything that guzzles power in the data center, like cooling systems, lights, and non-IT equipment. On the flip side, IT equipment power refers to the juice consumed by servers, storage devices, and networking gear. By crunching the numbers, we can gauge a data center's efficiency and pinpoint areas where we can cut down on energy waste. So, aiming for a lower PUE is the way to go to save energy and make data centers greener.
Benefits of Power Usage Effectiveness
As H. James Harrington wisely noted: "Measurement is the first step that leads to control and eventually to improvement. If you can't measure something, you can't understand it. If you can't understand it, you can't control it. If you can't control it, you can't improve it." Power usage effectiveness sets a strong foundation for improving the sustainability of data centers. But what are some of the main benefits data centers can gain when calculating and then improving PUE?
Cost savings: By adopting energy-efficient practices, companies can effectively reduce their energy consumption. This not only lowers their operational expenses but also translates into substantial cost savings over the long run. These savings can be allocated toward other critical areas of the business, fostering growth and innovation.
Environmental sustainability: Embracing efficiency measures in data centers goes hand in hand with minimizing the carbon footprint. A reduced PUE directly contributes to a more sustainable environment by curbing energy waste. This aligns with corporate sustainability initiatives and demonstrates a commitment to responsible business practices.
Enhanced reliability: Efficient data centers prioritize the optimization of cooling systems and power distribution. This meticulous approach ensures that critical infrastructure remains stable and functional, minimizing the risk of downtime. By maintaining high reliability, organizations can avoid disruptions and associated losses, safeguarding their operations and reputation.
Capacity optimization: An improved PUE not only reduces energy waste but also allows organizations to make the most of their existing infrastructure. By maximizing resource utilization, companies can delay the need for costly expansions or infrastructure upgrades. This flexibility in capacity management translates into significant cost efficiencies and better resource allocation.
Issues with Power Usage Effectiveness
When considering the PUE metric for assessing data center energy efficiency, it is important to be aware of potential challenges and drawbacks. While PUE offers valuable insights into energy utilization, it also comes with certain limitations that should be taken into account.
Focus: PUE measures only the ratio of total facility power to IT equipment power, excluding other energy-consuming components such as cooling and lighting systems. This narrow focus can lead to a misleading representation of energy efficiency, as it fails to capture the complete picture of energy consumption within the data center.
Inaccuracy: When calculating PUE, energy consumption values for various data center components are often estimated, which can lead to inaccuracies.
Complexity: Obtaining precise measurements requires robust monitoring systems and specialized tools, which can be costly to implement and maintain. Capturing accurate PUE values over time necessitates continuous monitoring and meticulous data collection, further adding to the complexity of the measurement process.
Cost: Even when sustainable solutions can be a smart investment in the long term, taking the first steps to implementing energy-efficient technologies requires a significant investment.
What Is the Ideal PUE Number?
The ideal PUE ratio is 1.0, as it signifies that every unit of power consumed is utilized solely by the IT equipment. In reality, however, most data centers fall within the range of 1.2 to 1.4 due to factors such as suboptimal equipment efficiency, inefficient cooling systems, and power losses in non-IT equipment.
While modern data centers leverage the latest technologies and best practices to achieve lower PUE values, facilities with older infrastructure might face some struggles in reaching this goal.
How to Reduce Your Data Center's PUE
Both middle-sized and large companies around the world are actively working to reduce the power usage effectiveness of their data centers, driven by a growing awareness of the significance of energy efficiency and sustainability in business operations. Microsoft, for example, "has become increasingly aware of its impact on the energy network as a large consumer of power," explained Senior Program Manager Dr. Christoph Mazur, and this is what led to a significant dedication by Microsoft to incorporating innovative cooling systems across their data centers.
Meta is another big name addressing the challenges of energy consumption, water usage, and greenhouse gas emissions. By adopting an innovative and proactive approach to data center design, Meta's recently completed data center buildings boast a PUE of 1.09.
5 Innovative Strategies for Reducing PUE
If your organization is looking to lower your PUE, here are five best practices:
Implement Efficient Cooling Techniques
One effective step is to utilize advanced cooling techniques such as hot and cold aisle containment, precision air conditioning, and airflow management. LinkedIn's data center in Hilltop, Oregon, is a great example of how to implement a waterside economizer system that reaches a low PUE (1.06). This advanced technology interacts with external air sensors, enabling the data center to leverage the naturally cool climate of the region for efficient and cost-effective cooling.
Optimize IT Equipment
Upgrading outdated and inefficient servers, storage devices, and network equipment with energy-efficient models is crucial. Techniques like server virtualization and consolidation can reduce the number of physical devices, resulting in significant energy savings. Microsoft, for example, saw remarkable energy efficiency improvements when it implemented server virtualization and increased server utilization.
Enhance Power Distribution and Energy Consumption
Optimizing power distribution infrastructure and implementing efficient energy management techniques, such as load balancing and virtualization, can enhance overall efficiency. These measures lead to higher energy efficiency, cost savings, and reduced environmental impact.
Strategically Adapting the Temperature
Some data centers are operating at temperatures below the recommended guidelines, but it is possible to gradually raise the temperature within a safe range as defined by industry standards, such as ASHRAE.
Operators can adjust the temperature settings of their cooling systems, such as CRAC or precision air conditioning (PAC) units, to increase the temperature setpoints slightly.
Locating Data Centers in the Arctic or Under the Sea
Locating data centers in the Arctic or underwater offers significant benefits for lowering power usage effectiveness. The Arctic's naturally cold climate provides a built-in advantage by enabling free cooling, which reduces the reliance on energy-intensive cooling systems. Following this same dynamic, underwater data centers take advantage of the surrounding water to efficiently dissipate heat, leading to improved energy efficiency and lower PUE. These innovative locations harness the power of nature, allowing for more sustainable and eco-friendly data center operations.
Learn More About PUE
Now you know some of the main aspects of power usage effectiveness and why it serves as a valuable benchmark for organizations that want to evaluate and refine their energy consumption practices. But there's a lot left to learn.
If you want to know more about PUE and how companies use this concept, check out the following links:
Things to Keep in Mind When Building a Data Center in 2023
Read more about:
Technical ExplainerAbout the Author
You May Also Like
.jpg?width=300&auto=webp&quality=80&disable=upscale)
