Yujui Lin
TomKat Graduate Fellow for Translational Research
Research Lab: Kenneth E. Goodson
Year Awarded: 2024
Yujui Lin is a PhD candidate at the Department of Mechanical Engineering at Stanford University, advised by Prof. Kenneth Goodson and Prof. Mehdi Asheghi.
He received his B.S. in Power Mechanical Engineering at National Tsing Hua University and his M.S. in Mechanical Engineering at Stanford University. He was a visiting student researcher at UC Berkeley during his undergrad and worked as a research assistant at National Taiwan University before joining Stanford.
His research focuses on heat transfer, electronics cooling, and Micro Electrical Mechanical Systems (MEMS). He is interested in the microfluidics heatsink embedded high-power-density devices such as power electronics or data center chips to advance the efficiency and performance of the cooling system.
Translational Research in Cooling Technology for the Sustainable and Energy Efficient Operation of AI Data Centers
A growing number of economic sectors, infrastructure, communication, and daily life activities rely on cloud services that require data centers to store and process information. However, the inefficient cooling system that does not match the surged chip performance in the data center impacts energy consumption significantly. The data centers consume 100 BWh/year, or 2% of US electricity production. Between 30-40% of the total power is for the cooling infrastructure. Moreover, the data centers consume 650 billion liters of water for electricity generation and refrigeration systems. Therefore, the sustainable operation of data centers requires innovative and transformative technologies in cooling systems.
The inefficiency of the cooling system comes from the high thermal resistance between the chip (~70-90°C) and the ambient, which forces the facility to supply coolant at lower temperatures (10-45°C) to keep the chips below the temperature limits, thus requiring bulky refrigeration systems that lead to considerable energy consumption. We propose to develop an advanced cooling and packaging solution, such as the force-convective and evaporative microchannel cooler embedded in the electronics, to reduce the thermal resistance by 10×. Therefore, coolants at higher temperatures could be used for the chip to eliminate the need for refrigeration infrastructure, resulting in energy and water savings and promoting efficient and sustainable data centers.