Tracking Data Center Efficiency: What PUE Can Say and Where We Can Look To Better Understand Energy Performance  

Education Type: 
1.5 Hours
0.2 CEU
Sponsored by: 

DOE Federal Energy Management Program - FEMP

LBNL will present its latest snapshot of efficiency in data centers, based on nearly 60 assessments of energy use and efficiency opportunities from 2006 to 2020. From those assessments, LBNL reports the standard metric for efficiency in data centers, Power Usage Effectiveness or PUE. They examine the relationship of average PUE to time, climate and data center type. In general, PUE has remained flat or declined slightly (suggesting an improvement in efficiency). Where a decline is really seen when PUE is weighted by information technology (IT) capacity. That is, the larger the data center, the greater the likelihood that it is more efficient or that its efficiency has improved over time.

Owners and operators of smaller data centers tend to have less time and access to capital, so smaller data centers present a tougher target for saving energy. The data suggests that is still the case. LBNL will contextualize those findings on PUE with a discussion of its strengths and weaknesses as the key industry metric for efficiency. They will look at efforts to enrich the industry's understanding of its energy use and emissions. The discussion will include alternative energy metrics and touch on the new Open Data Initiative (ODI) called for in the Energy Act of 2020. The training will close with an invitation to industry actors to work with FEMP and LBNL on developing the primary use cases for ODI and identifying which variables and metrics best serve those use cases.


Jefferey Murrell, Program Manager for the Federal Energy-Intense Facilities (EIF), FEMP  

Jefferey Murrell is program manager for the Federal Energy-Intense Facilities (EIF) program in U.S. DOE's Federal Energy Management Program (FEMP) and is a licensed professional engineer. He also serves as a Program Manager for the Federal Metering and Federal Energy Management and Information System (EMIS) programs. He currently supports the Federal Healthy Building Toolkit (HBT) and Federal Energy Efficient Product Procurement (EEPP) programs. Jeff graduated from Vanderbilt University with a B.S. in General Engineering and graduated from Columbia Southern University with a M.B.A. in Public Administration. He is currently matriculating at California Southern University in the Doctorate in Business Administration (DBA) program, with a focus on energy resiliency and EMIS development/planning.

Ian Hoffman, Senior Scientific Engineering Associate, Lawrence Berkeley National Laboratory  

Ian Hoffman is a with the Lawrence Berkley National Laboratory Center of Expertise for Energy Efficiency in Data Centers and other teams in the Building and Industrial Applications department. His work focuses on efficiency and resilience in data centers, individual and societal behavior on energy and efficiency, and utility customer-funded energy efficiency programs.

Steve Greenberg, Senior Energy Management Engineer in the High Tech and Industrial Systems Group, Lawrence Berkeley National Laboratory  

Steve Greenberg has researched and applied energy-efficient building and industrial systems for a variety of clients on three continents over the past 27 years. He has been involved in design, design review, commissioning, and retrofit of commercial and industrial buildings and performed detailed energy assessments on multiple federal data centers. Steve holds a BS in mechanical engineering and a master's degree in energy and resources--both from the University of California at Berkeley. He is a registered mechanical engineer in California. He is also a LEED Accredited Professional by the U.S. Green Building Council and a Certified Energy Manager.

Alex Newkirk, Research Associate, Lawrence Berkeley National Laboratory  

Alex Newkirk is a research associate in the Building Technology and Urban Systems Division at Lawrence Berkeley National Laboratory. Alex researches the diffusion of innovation, procurement of emerging technologies, and organizational behavior. He received a B.S. in Physics from Carleton College, and tries to bring an interdisciplinary approach to his work. Prior to joining the lab, Alex was an energy innovation policy consultant for IHS Markit, providing international innovation case study comparisons and analyzing the role of incubators and accelerators in the domestic innovation ecosystem.

Learning Objectives

Upon completion of this training, attendees will be able to:

  • Demonstrate an understanding of Power Usage Effectiveness (PUE) for data centers and which factors can influence the energy efficiencies of data centers;
  • Identify the Open Data Initiative for data centers as an effort to define the key variables and metrics needed to quantify energy use and efficiency in data centers;
  • Recognize that the metrics for energy performance of data centers or data center systems are power density, server power usage efficiency, and IT utilization;
  • Identify how, according to an international standard, PUE should not be used to compare different data centers; and
  • Identify how, although the data center industry often uses PUE to compare data centers, the Energy Act 2020 is advocating a different energy efficiency standard.
Federal Agencies and Facility Criteria: 
Building Types: