Chanwoo Park, a University of Missouri researcher and professor, has devised a new cooling system to reduce temperatures and improve efficiency in data centers as the need for them continues to expand.
Data centers and their extensive racks of servers account for about 4% of all electricity use in the U.S. as of 2022, as Tech Xplore reported. It takes about 40% of that energy to run cooling systems, which are essential for operational efficiency.
These power-hungry facilities could eventually account for 9% of all electricity generation in the country by 2030, according to a study by the Electric Power Research Institute. Much of that is expected to be driven by an increase in AI applications and processing. Â
"Cooling and chip manufacturing go hand-in-hand," said Park, a professor of mechanical and aerospace engineering, in the report. "Without proper cooling, components overheat and fail. Energy-efficient data centers will be key to the future of AI computing."
Currently, data centers employ a variety of cooling methods, usually involving air-moving fans or liquid transfer of heat, something akin to how many heat pumps work.
Park's method is a two-phase cooling system that aims to "dissipate heat from server chips through phase change, such as boiling liquid into vapor in a thin, porous layer," the report noted, with Park adding more detail.
"The liquid goes in different directions and evaporates on a thin metal surface," Park said. "Using this boiling surface, we're able to achieve very efficient heat transfer with low thermal resistance."
This is important work and dovetails with a recent federal effort to find new "transformational, highly efficient, and reliable cooling technologies for data centers," led by the Advanced Research Projects Agency-Energy.
The International Energy Agency states that since 2010, the worldwide internet user head-count has more than doubled and internet traffic has grown to twenty times its previous size. Luckily, energy efficiency improvements and more sustainable sources, like clean solar and wind, have kept pollution down to only modest growth since that time.Â
Although it can sometimes get a bad rap, artificial intelligence has been leveraged to assist in many greener endeavors. It's helped to make wind turbines safer for wildlife, expedite solar panel applications for homeowners, improve high-speed rail, and optimize recycling efforts.
Park's plan to reduce energy consumption while improving data center cooling systems has a lot of potential.
"We're trying to get ahead of the curve and have something ready and available for the future of AI computing," Park shared in the report. "This is a futuristic cooling system."Â
His campus is also adding a new Center for Energy Innovation, where researchers can collaborate across disciplines to solve issues around energy concerns and AI.
"The center will allow us to explore additional ideas and innovations around energy-efficient processes," Park stated. "These are complex problems that require different areas of expertise. I look forward to future collaborations."
Join our free newsletter for weekly updates on the latest innovations improving our lives and shaping our future, and don't miss this cool list of easy ways to help yourself while helping the planet.
Cool Picks
