Pittsburgh, PA
Data centers are one of the most energy-intensive building types, consuming 10 to 50 times the energy per floor space of a typical commercial office building. About 2% of U.S. electricity fuels data centers, and for many Federal agencies, the proportion is much higher. This session will cover multiple approaches and technologies for shovel ready ways to improve energy efficiency while achieving resilience in these significant energy users. The speakers will include conceptual design considerations, retrofit opportunities, and operational innovations. Case studies addressing water-cooled chip systems and modular high performance computing design and construction will be shared.
Instructors
Otto VanGeet, Principal Engineer, National Renewable Energy Laboratory Read Bio
Otto Van Geet is a Principal Engineer at NREL. He has been involved in the design, construction, and operation of energy-efficient research facilities such as laboratories and data centers, office and general use facilities, and low-energy-use campus and communities. Van Geet was one of the founding members of the Labs21 (Smart Labs) program and his experience also includes renewables screening and assessment, PV system design for on- and off-grid applications, energy audits, and minimizing energy use. Van Geet has authored many technical reports and conference papers and been recognized with many awards from professional associations, including the 2007 Presidential Award for Leadership in Federal Energy Management and the 2011 GreenGov Green Innovation Presidential Award for the NREL Research Support Facility data center.
Ian Hoffman, Technology Researcher, Lawrence Berkeley National Laboratory Read Bio
Ian Hoffman leads the data center technical assistance and energy assessments team with the joint DOE/Berkeley Lab Center of Expertise for Energy Efficiency in Data Centers, based at Lawrence Berkeley National Laboratory. His research includes the technical and behavioral dimensions to energy management and decarbonization, particularly in the IT sector. He has an MS in Energy and Resources from the University of California at Berkeley.
Bill Thigpen, High-End Computing Capability (HECC) Component Manager, NASA Read Bio
As manager of the High-End Computing Capability (HECC) Component, William (Bill) Thigpen leads key supercomputing environment development efforts-including operation of energy-efficient systems in the Modular Supercomputing Facility (MSF)-to significantly reduce the water and power required to cool today's supercomputers and provide more computing capability to NASA users. Thigpen oversees the deployment and testing of groundbreaking technologies such as machine learning, artificial intelligence, and cloud tools to further enhance NASA high-end computing users' experience. He is also evaluating the potential benefits of quantum computing on NASA's mission challenges. Thigpen's extensive industry career supporting government contracts includes software engineering and mission support management. He led technical teams that provided HEC capabilities to NASA, and developed a wide variety of software solutions including operating systems, network and messaging protocols, machine control software and air traffic control software. He holds a bachelor's degree in computer science from the University of Nebraska Omaha.
Learning Objectives
Upon completion of this course, attendees will be able to:
- Identify the basics of data center energy efficiency;
- Recognize the different types of data center cooling technologies and their pros/cons with regards to operations and energy efficiency;
- Identify options to retrofit or establish operational efficiencies to make an existing data center more energy efficient;
- Select strategies to work with IT managers to reduce unnecessary IT loads in a data center.
