Industry Perspective is a regular Data Center Journal Q&A series that presents expert views on market trends, technologies and other issues relevant to data centers and IT.
This week, Industry Perspective asks Patrick Flynn about power usage effectiveness (PUE) in today's data center market. Patrick serves as Group Leader, Applied Intelligence & Sustainability at IO, originally joining IO as Lead Sustainability Strategist to identify, analyze and implement sustainability projects across the company's global footprint. Before joining IO, he received his MBA from the MIT Sloan School of Management and worked for Good Energies, a renewable-energy and energy-efficiency investor, where he primarily focused on venture investment opportunities and portfolio company support in the green buildings and energy-efficiency sector. Before his time with Good Energies, Patrick was an engineering consultant, working for Edwards and Zuck on high-rise building HVAC system design, and he also earned his Professional Engineering license and LEED (Leadership for Energy and Environmental Design) accreditation in the process. He graduated from Stanford University with a B.S. in mechanical engineering and a minor in economics.
Industry Perspective: Would you discuss the definition of power usage effectiveness (PUE) and its role in today’s market?
Patrick Flynn: PUE represents the ratio of total energy that comes into a data center to the portion that reaches and is used by information technology (IT) equipment. The energy that reaches the computing equipment is considered productive, while energy used for infrastructure (e.g., cooling, lighting, security and system inefficiency) is auxiliary and viewed as waste. This waste is one place to look for efficiency gains, but as we will discuss, it’s not always the best place to do so. Data centers strive for a PUE of 1.0, which represents a hypothetically perfect facility where energy is used exclusively to power IT, and there is no energy loss or overhead in the system. In today’s market, traditional PUE measurement still has a place in current customer assessments of data center efficiency and environmental sustainability, even though providers’ efficiency claims are infrequently validated and the measure itself does not account for the type of work being performed.
IP: Over the past few years, concerns have arisen regarding the limitations of PUE. What is the biggest myth about PUE, and how does dispelling that myth address some of these concerns?
PF: The biggest myth about PUE is that it measures efficient performance of a data center’s end purpose. Ultimately, the data center’s job is not to provide energy to IT equipment or infrastructure but to do useful and productive computing; that power is flowing smoothly to IT equipment does not mean that equipment is doing good work.
At IO, we recognize there is a performance gap between the way the industry measures PUE and the need for more-useful information to continue driving performance. Therefore, our job is two-fold: to validate PUE in actual deployments happening today and to evolve the calculation of the PUE metric so it becomes a more meaningful tool for business.
IP: What are the difficulties with the approach to PUE that many data center operators use today?
PF: PUE has industry-acknowledged shortcomings in terms of accuracy, and it rarely provides a reliable way to compare data centers. Without controlling for variables like location, size, design and data sets, PUE cannot tell you with accuracy which data center is performing better.
Over the past months, we have heard serious questions from our customers about the utility of today’s PUE. Typically, PUE is measured for entire facilities. A mixed-use building may house any number of functions, such as data centers, labs and offices. For these types of mixed-use environments, determining the power usage of just the data center environment is difficult, especially when some systems share power or cooling infrastructure.
PUE measurement is also a challenge in a colocation data center, which is a mixed-use facility in that it serves multiple customers. To Tenant A, the neighbors’ power, efficiency, equipment and overhead are unknown, and they do not contribute to the energy that reaches the IT gear of that tenant. For PUE to be useful to Tenant A, it must be specific to that tenant’s defined infrastructure and data systems, even if they are a small fraction of a larger shared data center infrastructure.
Further reducing the utility of today’s PUE is the fact that it is typically calculated retroactively, looking at energy consumption over a historical period and then calculating an implied average power usage over the period. Such an approach masks any volatility in PUE, and it fails to give operators timely feedback.
IP: How can these difficulties be addressed?
PF: IO recognizes the need to evolve to newer, enhanced measurement models for greater energy efficiency. Though a retroactive, building-averaged PUE may serve to confirm overall progress, it falls short of helping to pinpoint opportunities for improvement. Service providers must strive to give customers information that allows them to improve performance and make better business decisions. This is why IO is increasing the usefulness of energy-efficiency measurement through an evolved methodology called real-time PUE. Real-time PUE measures power efficiency instantaneously, and provides a level of granularity down to the individual server.
IP: How does real-time PUE benefit a specific organization?
PF: A specific organization can benefit from real-time PUE by using a data center (such as IO’s software-defined data center) that captures live data from across the infrastructure, enabling monitoring, measurement, benchmarking and continuous improvement.
Modular data center designs allow system administrators to pinpoint where power is being used and where improvements are possible. By taking the analytical lens down to the server level, we are one step closer to tying PUE to actual digital work.
We have had initial conversations with companies who are generating strong interest in implementing the real-time PUE methodology all the way into the application layer. When our industry achieves this level of measurement, we will have arrived at a truly comprehensive measure of efficiency, inclusive of the work output.
IP: How can you prove that real-time PUE is more effective and efficient than traditional PUE measurement?
PF: Arizona Public Service (APS) recently collaborated with IO on a comparative, independent third-party study that evaluated both construction-based data centers and modular data centers. IO specifically operates both traditional and modular, and this study showed how our modular products performed under almost identical conditions. APS specifically reviewed IO Phoenix, which is approximately 180,000 square feet and one of the largest Tier 3 design-certified data centers in the world, under technical scrutiny for 12 months and also monitored PUE for the calendar year 2012. Data was collected from both IO.Anywhere modules and the traditional build-out. As a result, APS found that IO’s manufactured, modular data center approach achieved 19 percent energy cost savings compared with the traditional construction-based environment, validating how a modular deployment can achieve even greater efficiencies.
IP: What happens to organizations if they do not convert to real-time PUE?
PF: The findings from our independent third-party survey with APS reveal that organizations that do not convert to real-time PUE will miss out on cost savings and efficiency gains, and they will have less information with which to assess true TCO.
IP: How do you see real-time PUE affecting data center efficiency in the future?
PF: At IO, we believe real-time PUE is the logical next step for the industry, especially as modular data center designs become the standard. Real-time PUE is just one critical step on the path to achieving breakthrough cost and efficiency gains. Although providing data center users with more-useful efficiency measurement enables better decision making, it is analytics work that yields recommendations to take action. Evolving our category’s thinking on PUE is an instrumental step in improving data center efficiencies and will serve CIOs, CFOs, facilities managers, end users and the planet.