- Nvidia’s Rubin generation AI servers run coolant that is often hotter than a hot tub, as it enters a closed loop at a temperature of up to 45°C and exits at around 55°C, with no performance penalty.
- The principle used here is that of a temperature gradient in which the server chips still operate at a higher temperature than the coolant, allowing heat transfer to occur smoothly.
- The approach allows Nvidia to conserve about 2.6 million gallons per megawatt per year to near zero and could save a 50 MW site more than $4 million per year in costs.
Nvidia’s Rubin generation is the first of its kind in multiple ways, but the one that arguably stands out is the fact that it is 100% liquid-cooled, implementing it as a core design feature at the platform level with every chip and networking component covered by a closed loop.
Nvidia’s reference design aims to enable AI factories that effectively consume no water by implementing a closed-loop circuit that does not take advantage of evaporative water cooling 99% of the time.
The approach goes a step further by potentially removing industrial chiller plants from the equation by using liquid cooling where the temperature is up to 45 degrees Celsius for the refrigerant entering the system and approximately 55 degrees Celsius when leaving the system.
A game of numbers for a data center that analyzes them at an industrial level
Nvidia’s approach is not the first of its kind in terms of design or thermal implementation; However, IBM beat it by 16 years, using 60-degree coolant to cool its supercomputer prototype at ETH Zurich.
The premise was the same: a much smaller carbon footprint, significantly less energy consumption and a circuit that is filled once and left alone. What Nvidia adds is scale: its Vera Rubin-based DSX design extends the approach to an entire facility rather than a single machine.
“The NVIDIA DSX reference design for AI factories has zero water consumption – we have eliminated massive amounts of power use and virtually all water use,” said Ali Heydari, director of data center infrastructure and cooling at Nvidia.
The coolant is a 3:1 mixture of water and propylene glycol and easily extracts heat from the chips which get much hotter than the coolant itself.
The savings aren’t just water-based: Industrial cooling plants reduce their energy costs by about 4% for every degree they increase their operating temperature, and Nvidia estimates that a 50-megawatt hyperscale facility could save more than $4 million a year in cooling-related energy and water costs if it adopted its design.
With cooling historically accounting for up to 40% of a data center’s total power bill, Nvidia’s approach is ambitious, but should pay for itself quickly: by eliminating server fans entirely and allowing coolers to run only when necessary, and at considerably less power than a circuit that demands colder coolant.
It should be noted that DSX is a reference design: a set of best practices for building an AI factory, not a census of what the industry has actually built, where many designs still consume large amounts of water, even as AI data centers continue to pop up in drought-stricken regions of the US.
It might seem like Nvidia is making the case for a “greener” AI data center for the future, at least in its own designs, but that’s only part of the story, and the part it leaves out is a little less flattering. Nvidia didn’t adopt hot water cooling because it’s stylish, green, or smart, although it’s all three. It adopted it because its accelerators became so energy dense that moving air through them was no longer physically viable, and once you go all-in on liquid cooling, heating it is simply the efficient way to do it.
The sustainability victory is real and is a consequence of a thermal problem that Nvidia created for itself by building the most energy-dense chips in the industry. Hot water is not the trick. It is the overdue invoice, paid in the most efficient currency available.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
