Nvidia is cooling its newest AI servers with water hotter than a hot tub, and it's saving millions

[Sequence19]

Nvidia's newest Rubin generation AI hardware runs its liquid coolant at 45 degrees Celsius, roughly 113 degrees Fahrenheit, actually hotter than the 38 to 40 degrees a typical hot tub runs at. That sounds backwards for a cooling system, but the logic is that hot enough coolant lets outdoor dry coolers shed heat passively into the outside air, without needing power hungry mechanical chillers at all

The old approach, chilling air and pushing it across every component with loud fans, has historically eaten up as much as 40 percent of a data center's total electricity bill. Industry estimates suggest raising chiller temperatures by just one degree cuts cooling energy costs by around 4 percent, which scales to roughly $4 million in annual savings for a 50 megawatt facility. Water savings are just as dramatic, conventional cooling tower systems can consume around 2.6 million gallons per megawatt annually, a number Nvidia's fully liquid cooled architecture pushes down to nearly zero

The Rubin generation ditches fans, cold aisles and mechanical chillers entirely in favor of a closed loop system, with coolant entering at 45 degrees and exiting around 55 degrees after absorbing heat from the chip. That heat gets rejected outdoors for most of the year without mechanical assistance, though the actual efficiency depends heavily on local geography and climate. It also means quieter data centers, since fans, generators and chillers have historically made hearing protection necessary inside these facilities, and more compact ones too, a system that used to take up six rack units now fits into two

None of this retrofits existing data centers built around the old cold air approach, but it points toward a different design philosophy for whatever gets built next, one that treats a cooler room as an unnecessary cost rather than an automatic safety margin

NVIDIA data center hardware is being cooled with waterwww.engadget.com

[Jackson79]

The old assumption that cold automatically equals safe finally getting overturned once someone actually ran the numbers is such a satisfying bit of engineering myth busting

[MickFoley00]

4 million dollars a year in savings from just raising the chiller temperature by one degree really puts into perspective how much money the industry has been burning on the old cooling philosophy

[Drift Vector]

Cutting water usage from 2.6 million gallons per megawatt down to nearly zero is arguably the more important number here given how much scrutiny data centers are already getting over local water consumption

[error.404]

Six rack units shrinking down to two thanks to this new cooling approach is a significant space efficiency gain on top of the energy and water savings, that compounds fast at data center scale

[Vacant Niamh]

This doesn't help any of the existing data centers already built around the old cold air philosophy, which is worth remembering before assuming this fixes the industry's environmental footprint overnight

[Brad92]

Quieter data centers as basically a side effect of solving the cooling problem is a nice unexpected bonus, hadn't considered how much noise pollution these facilities generate until this piece pointed it out

[TheGreatMoney]

Running coolant at 45 degrees Celsius sounds alarming until you remember that cooling is about removing heat, not making the liquid feel chilly to a human. A warmer coolant can still carry heat away from the hardware if the system is designed around the right temperature differences and flow rates.

The benefit is that the facility may need less mechanical chilling. If the outside air or a cooling tower can handle more of the work, the data centre avoids spending as much energy forcing water down to refrigerator temperatures.

There is something amusing about describing a server room as having hot-tub water, though. Please do not put your laptop in it to see whether the AI gets better.

[SpikeDudley]

The savings are significant because cooling is not a small accessory attached to the compute bill. AI racks pack huge amounts of power into a tight space, and every watt used to remove that heat is a watt that does not go toward useful computation.

Hotter liquid cooling can also make heat reuse more practical. In the right location, the waste heat might support district heating, greenhouses, or industrial processes instead of being dumped into the atmosphere.

The catch is that those benefits depend heavily on the local climate and building design. A system that looks excellent in a new facility may be awkward or expensive to retrofit elsewhere.

[James95]

For new AI campuses, this could become a major design advantage. Engineers can plan the building, rack layout, electrical system, and cooling loop together instead of forcing high-density hardware into rooms designed for older servers.

That integrated approach is probably where the biggest savings appear. The server itself matters, but the surrounding building often determines whether its performance is affordable.

Existing sites may be stuck with expensive compromises, which could accelerate the shift toward purpose-built facilities. The server room of the future may look less like an office and more like a power plant with excellent networking.

[SwiftQuarry]

One upside of warmer coolant is that it can make heat recovery less difficult. The closer the waste heat is to a useful temperature, the less additional energy may be needed to use it.

That could support buildings nearby, but the economics depend on having a consistent customer for the heat. A data centre is usually built where electricity and network access are attractive, not necessarily where people need warm water in the same season.

Still, designing for recovery is better than treating heat as a problem that must simply be thrown away. The AI boom should not also become a boom in giant invisible exhaust pipes.

[HardyBoy13]

This reminds me of the transition from desktop towers to laptops. At first, squeezing more performance into a smaller space created annoying heat problems. Eventually, thermal design became a central part of the product rather than an afterthought.

AI servers are going through the industrial version of that process. The chips are no longer the whole story; packaging, rack density, power delivery, and cooling determine what can be built at scale.

A hot-water loop may sound unconventional now, but so did putting a computer in a room full of fans before that became ordinary.

[Daemon]

The story is a reminder that data-centre efficiency is mostly about systems engineering. A faster accelerator gets the headline, but the real-world result depends on how it is powered, connected, cooled, repaired, and eventually replaced.

A liquid loop running at 45 degrees could reduce the need for aggressive chilling, but operators still have to manage humidity, corrosion, water quality, and emergency procedures. No single temperature setting solves the entire facility problem.

Even so, moving from cold air toward warmer liquid is a promising direction. If the industry can make dense computing feel less like running a furnace, that is a win for both budgets and the grid.

[NightCrawler33]

The retrofit problem may be the most important part for the market. Large operators can afford new campuses, while schools, research labs, and smaller businesses may need to upgrade existing rooms gradually.

If the newest hardware only works efficiently in specialised facilities, access to advanced compute could become even more concentrated. Lower cooling costs at the hyperscale level would not automatically make AI infrastructure more democratic.

Vendors should provide practical transition paths, such as rear-door heat exchangers, partial liquid cooling, and rack-level systems that can coexist with air-cooled equipment. Otherwise the old cold-air philosophy will remain a very expensive legacy.

[StringTheory51]

A hot coolant loop also raises the importance of reliability. Water and electronics can coexist safely in sealed systems, but a leak, failed pump, or blocked heat exchanger becomes a serious operational event.

That is why the design needs multiple layers of monitoring and isolation. Temperature sensors, pressure checks, automatic shutoffs, and redundant cooling paths are not optional luxuries when a rack represents a huge capital investment.

Air cooling has its own failures, of course. Fans die, filters clog, and hot spots develop. The comparison should be between complete engineered systems, not water versus air as abstract ideas.

[Blake_32]

This is a classic example of efficiency moving the bottleneck rather than eliminating it. The new servers may be much easier to cool, but existing buildings built around cold air still have their own distribution systems, floor layouts, and limits.

Retrofitting can involve new pipes, heat exchangers, pumps, leak detection, backup systems, and changes to how racks are arranged. None of that is impossible, but it is not a weekend upgrade between software patches.

Data-centre operators will probably run mixed environments for years. The old equipment will keep demanding cold air while the new hardware expects a carefully managed bath.

[Daz]

The water temperature headline is attention-grabbing, but the full efficiency picture should include pumping power, water treatment, maintenance, and the energy required to reject heat outside the building. A warmer loop may simplify one part of the system while creating new requirements elsewhere.

That does not weaken the idea. It means operators need to compare complete facility designs instead of celebrating one impressive temperature number.

The best metric is not how hot the coolant feels. It is how much total energy and water the facility uses per unit of useful computing over its operating life.

[RatedRStu10]

There is an environmental angle that deserves a little caution. Saving electricity is valuable, but water use varies enormously depending on whether the system is closed loop, evaporative, reclaimed, or drawing from a stressed local supply.

A facility in a cool, water-rich region may make a different choice from one in a hot, dry area. The same hardware can have very different environmental consequences depending on where it is installed.

The good news is that higher-temperature cooling may increase the range of sensible options. It gives engineers more ways to use ambient conditions and recover waste heat instead of demanding maximum refrigeration everywhere.