Google Presents Liquid Cooling Solutions for AI Datacenters at Hot Chips 2025
By
giuliomagnifico
An everything bagel for the brain. Substantive, layered, well-seasoned.
Summary
Google presented on liquid cooling technology for datacenters at Hot Chips 2025, highlighting how water's superior thermal conductivity (4000x better than air) makes it essential for cooling high-power AI and machine learning chips. The article discusses the growing adoption of liquid cooling in datacenters due to increasing power demands and heat output from modern processors.
Key quotes
· 3 pulledWater has a thermal conductivity about 4000 times that of air, making it an attractive solution to deal with the cooling demands associated with the current AI boom.
Liquid cooling has taken an increasing role in datacenters, amid increasing power draw and correspondingly high heat output from the latest chips.
Machine learning in particular has an insatiable appetite for power and cooling.
You might also wanna read

Microsoft Develops Microfluidic Cooling Technology for More Efficient Data Centers
Microsoft is developing a microfluidic cooling technology that uses liquid coolant flowing directly into silicon chips to improve heat dissi

Tech Companies Explore Space-Based Data Centers as AI Infrastructure Demands Grow
The article discusses how tech billionaires and major AI companies are exploring space-based data centers as a solution to Earth's limitatio
Scientists and engineers race to reduce AI's growing energy consumption
This article explores the massive and growing energy consumption of AI systems, particularly data centers powering large language models lik

Google Launches Private AI Compute Cloud Platform for Privacy-Focused AI Processing
Google is launching Private AI Compute, a cloud-based platform that enables advanced AI features on devices while maintaining privacy levels

Google's Project Suncatcher: Space-Based AI Data Centers Powered by Solar Satellites
Google has announced Project Suncatcher, a 'moonshot' research initiative to launch AI data centers into space using solar-powered satellite
General Compute Launches ASIC-Based Inference Cloud for Faster AI Agent Performance
General Compute is an inference cloud built on ASICs (purpose-built alternatives to Nvidia GPUs) designed specifically for AI inference, not
