Elon Musk’s Tesla Gigafactory in Texas is expanding to accommodate a cluster of AI supercomputers, and Supermicro’s CEO is a big fan of the cooling solution. Supermicro founder and CEO Charles Liang took to X (formerly Twitter) to celebrate Musk’s use of Supermicro’s liquid cooling technology for Tesla’s new cluster and the similar xAI supercomputer, which is also on the way.
Pictured together amid server racks, Liang and Musk are looking to “bring liquid cooling technology to large AI data centers.” Liang estimates that the impact of Musk’s adoption of liquid cooling technology in AI data centers “could lead to saving 20 billion trees for our planet,” evidently referring to the improvements that could be achieved if liquid cooling were adopted in every data center around the world.
AI data centers are well-known for their massive power consumption, and Supermicro is hoping to reduce that pressure by promoting liquid cooling. The company claims that direct liquid cooling can deliver up to 89% reduction in cooling infrastructure electricity costs compared to air cooling.
Thanks to @elonmusk for pioneering liquid cooling technology in large AI data centers! This could save 20 billion trees for our planet❤️ pic.twitter.com/oJ48Dw3YVFJuly 2, 2024
In a previous tweet, Liang clarified that Supermicro’s goal is to “boost DLC [direct liquid cooling] “Adoption of Supermicro cooling technology is growing from <1% to over 30% in one year.” Musk is deploying Supermicro cooling at scale for his Tesla Gigafactory supercomputer cluster. The new expansion of the existing Gigafactory will house 50,000 Nvidia GPUs and more Tesla AI hardware to train Tesla’s fully autonomous driving feature.
The expansion is turning heads thanks to the supermassive fans being built to cool the liquid cooling, something Musk also recently highlighted in an X-rated post of his own (expand the tweet below). Musk estimates that the Gigafactory supercomputer will consume 130 megawatts when deployed, with growth to 500 MW planned after Tesla’s proprietary AI hardware is installed. Musk says construction of the facility is nearly complete and it should be ready for deployment in the coming months.
Sizing for ~130MW of power and cooling this year, but ramping up to 500+MW in the next 18 months or so. Goal: ~50% Tesla AI hardware, 50% Nvidia/other. Play to win or don’t play at all.June 20, 2024
Tesla’s Gigafactory supercomputer cluster shouldn’t be confused with Elon Musk’s other supercomputer cluster, the multi-billion dollar X/xAI supercomputer, which is also under construction. In fact, Musk is building not one, but two of the world’s largest GPU-powered AI supercomputer clusters. The xAI supercomputer is a bit more well-known than Tesla’s, with Musk having already ordered 100,000 H100 GPUs from Nvidia. xAI will use its supercomputer to train GrokAI, X’s original AI chatbot alternative, available to X Premium subscribers.
The xAI supercomputer, which is expected to be ready “within months,” will also be liquid-cooled by Supermicro and is already planning an upgrade to 300,000 Nvidia B200 GPUs next summer. Recent reports suggest that getting the xAI cluster online is a slightly higher priority for Elon Musk than it is for Tesla, as Musk reportedly ordered Nvidia to ship thousands of GPUs originally ordered for Tesla to X in June. The move reportedly delayed the buildout of Tesla’s supercomputer cluster by months, but as with so much Elon Musk-centric news, it’s likely an exaggeration.
