Data Center Cooling with Kelley Mullick - #30

In this episode I talk with Dr. Kelley Mullick a data centerexpert and the CEO and co-Founder of Avayla Consulting Group. The focus of thisepisode is on the technology and methods used to cool data centers. We startthe episode discussing Kelley’s academic background as a chemist and breakinginto the data center field at Intel.Then we get into data center cooling. Kelley explains themetrics used to evaluate the thermal performance of data centers, the overallincrease in compute power usage in the AI age, common cooling methods such asair and liquid cooling, how data center operators determine the optimal coolingmix for their facility, the costs of retrofitting a data center with moreadvanced cooling technology and much more. I hope you enjoy! ARTICLES & PAPERS:Kelley’s PhD dissertation: https://rave.ohiolink.edu/etdc/view?acc_num=osu1324247960CPU Cooler: Liquid Cooling Vs. Air Cooling: https://www.intel.com/content/www/us/en/gaming/resources/cpu-cooler-liquid-cooling-vs-air-cooling.htmlAlibaba is developing its own immersion cooling tech: https://www.datacenterdynamics.com/en/news/alibaba-is-developing-its-own-immersion-cooling-tech/Hybrid Liquid Cooling Is the Only Way to Stop AI fromBoiling Over by Kelley Mullick: https://thesiliconreview.com/2025/11/hybrid-liquid-cooling-is-the-only-way-to-stop-ai-from-boiling-overCool It: Team Tackles the Thermal Challenge Data CentersFace: https://blogs.nvidia.com/blog/liquid-cooling-doe-challenge/Americans Oppose AI Data Centers in Their Area: https://news.gallup.com/poll/709772/americans-oppose-data-centers-area.aspxHow is Switch Building AI-Ready Sustainable Data Centres?: https://datacentremagazine.com/news/how-is-switch-building-ai-ready-sustainable-data-centres BOOKS:Nancy Drew SeriesMUSIC INTRO/OUTRO:massobeats - chamomile (https://youtu.be/YFuAgsA4bNI) CONNECT: LinkedIn:https://www.linkedin.com/in/adrian-dolinay-frm-96a289106/ GitHub: https://github.com/ad17171717 X:https://twitter.com/DolinayG Odysee: https://odysee.com/@adriandolinay:0 Medium: https://medium.com/@adriandolinay PODCAST: Apple Podcasts:https://podcasts.apple.com/us/podcast/the-aspiring-stem-geek/id1765996824 Audible: https://www.audible.com/podcast/The-Aspiring-STEM-Geek/B0DC73S9SNiHeart Radio: https://iheart.com/podcast/202676097/ Spotify: https://open.spotify.com/show/60dPNJbDPaPw7ru8g5btxV |-Video Chapters-|0:00 – Intro 1:32 – Kelley’s interest in STEM and the arts 2:23 – Studying chemistry in undergrad and grad school 6:17 – Defining what a data center is 7:13 – Kelley’s time at Intel 9:33 – Why compute needs to be cooled 10:33 – Metrics to measure the thermal performance of a data center 13:39 – Colocation operator vs hyperscalers 14:55 – AI and increased compute power usage 20:15 – Components outside of compute that contribute to a data center’sthermal load 20:48 – Data center cooling methods 26:16 – How data center operators determine the optimal cooling architecture 29:17 – Data centers outside of the United States 31:40 – The potential best path for data center cooling 34:28 – Feasibility of retrofitting data centers with more advanced cooling 37:54 – The public’s view on data centers 40:43 – Kelley’s ideal data center policies 44:25 – Things to do in Ohio 46:48 – Book recommendations 48:16 - Conclusion