•  
  •  
 

Abstract

Project Mentor(s): Susan Rivera, PhD

The US is confronting a growing energy demand crisis, with analysts projecting a 2.4% rise in data center energy consumption by 2030. Data centers are major energy consumers, with primary usage concentrated in IT equipment and extensive cooling systems. While Power Usage Effectiveness (PUE) remains the leading metric for assessing energy efficiency, it is limited to assessing total facility energy and IT energy use without accounting for environmental and social impacts. Supplementary metrics of data center energy consumption include Water Usage Effectiveness (WUE) and Carbon Usage Effectiveness (CUE), which measure water use and carbon emissions, respectively, per kilowatt-hour (kWh) of IT energy use. These metrics provide insight into the environmental cost of energy sources, however, are limited by variability based on geographic location, energy grid composition, and a lack of standardization. Emerging technologies like the COOLERCHIPS (Cooling Operations Optimized for Leaps in Energy, Reliability, and Carbon Hyperefficiency for Information Processing Systems) program aim to enhance the energy performance of data centers by improving microchip efficiency and optimizing hybrid cooling systems that use both air and liquid coolants. These innovations show promise in reducing energy metric scores, but a more integrated framework is necessary to capture the complexities of sustainable data center operation. This review explores the pressing issue of data center energy consumption in the U.S., critically examines current and emerging efficiency metrics, and emphasizes the need for more holistic, adaptable solutions. As digital infrastructure expands, prioritizing energy and environmental efficiency is vital to ensuring sustainable growth in large-scale industries.

SOURCE Form ID

127

Download

Share

COinS