– Daniel Bodenski, Director of Strategic Solutions, Electronic Environments Corporation, says:
When analyzing the effectiveness of a data center, one of the most critical components to consider is Power Usage Efficiency (PUE), a metric used across the board by mission-critical facility implementation teams, data center owners and operators, and C-level executives to assess a data center’s current and potential energy efficiency. When used properly, this information can be leveraged to produce a stronger, more efficient mission-critical environment, gain a deeper understanding of competition, and open the door to exploring options for added efficiency improvements such as data center renovations, new builds, or migrations to the cloud or a third-party service provider. In order for a facility to both evolve and adapt to ever-changing industry climates and customer demands, a deep understanding of what lies under the hood is vital to its success, and data center PUE is no exception.
PUE is by definition the ratio of total energy consumption of a data center including all fuels, divided by the total energy consumption of IT equipment. It was first developed by the Green Grid Association in 2007 and quickly became the go-to metric adopted globally to definitively measure and track data center energy efficiency. Originally designed as an end-user tool to assist data center operators, today PUE has been implemented throughout all facets of a facility, providing many benefits to data center owners and operators for the duration of a data center’s lifecycle. Since PUE is such a versatile tool, understanding how to properly use it with respect to the different stages and segments of a facility’s lifecycle is an important step to reaping the highest capacity of its benefits.
Below, we will provide insight into how some of the key stakeholders in the data center’s lifecycle, including designers, operators and C-suite executives, view the importance of PUE and how each party uses PUE to satisfy their demands and the demands of their customers.
When designing a data center, those in charge of creating a plan that is elegantly simple yet highly effective are faced with a difficult task. The design features must drive energy efficiency and innovation, while upholding maximum uptime and simultaneously protecting against outage threats. Through the use of PUE metrics, a balanced approach can be developed with clearer insight into how a data center will eventually perform, making it easier to implement a fresh, energy efficient mechanical/electrical system.
Utilizing environmental factors as well as documented, low-risk strategies such as increasing the supply air temperature and/or chilled water temperatures are some of the ways that many designers effectively save overhead costs on energy every day. To produce optimal results, the design team should adhere to Green Grid’s PUE definition of components during design and analysis and properly define source energy to ensure their initial calculations will match ultimate operational results. With PUE in mind, designers have the right information and knowledge to expertly select the best options for reduction of overall mechanical system energy use, while reducing risks to uptime.
In today’s industry, a data center’s operations team is under immense pressure to reduce energy usage through solutions that not only yield the highest level of results, but fit within the framework of a live, operational facility. Though PUE is incredibly important to the operations team, their primary tasks are geared towards ensuring maximum availability for all critical infrastructure and management of real-time data center planning activities. As a key performance indicator metric for operations teams, PUE is used to report the overall effectiveness of the data center to management on a regular basis. By understanding what is currently happening in regards to energy usage, operators have the opportunity to determine new and effective ways of reducing power loss, while concurrently using PUE to justify the energy savings measures taken.
According to the Green Grid, there are three separate levels for measuring PUE, each providing a unique value proposition:
- Level 1, also referred to as “basic” measurement, allows for IT equipment energy measurement at Uninterruptible Power Supply (UPS) output on a weekly or monthly basis;
- Conversely, level 2, or intermediate measurement, allows for IT equipment energy to be measured at Power Distribution Unit (PDU) outputs;
- Finally, level 3 is regarded as the most accurate form of measurement, requiring a high level of technology coordination, data collection, and human interaction. For this level of accuracy, facilities may install PUE measuring devices such as a kWh meter with the help of an experienced firm that is fully capable of installing the equipment and working in a live data center environment.
C-level data center executives view PUE as a significant factor affecting and influencing the effectiveness of the facility’s overall IT strategy. For example, Total Cost of Ownership (TCO) is a key metric used by executives to evaluate strategies to operate within an enterprise data center, a third-party colocation facility, or cloud provider. While energy usage typically makes up more than half of total operational costs within a data center, C-level executives regard technology infrastructure and associated labor as the main contribution to TCO, and the energy efficiency portion (and associated PUE) at roughly 8 to 15 percent of the TCO. Additionally, within a typical data center setting, a CIO regards PUE as a main Key Performance Indicator (KPI) and requires regular monitoring of these levels for presentation to corporate clients or as a metric to potential third-party customers.
By understanding the different professional perspectives of PUE, you can see that this one, simple metric can be highly impactful in many facets of a mission-critical facility. Through combined efforts from key stakeholders in the data center’s lifecycle, understanding and using PUE can mean the difference between a facility that produces continuous long-term success and one that just falls short.