- Nvidia’s closed-loop liquid cooling system virtually eliminates water waste
- Direct on-chip cooling transfers heat more efficiently than air
- It enables higher performance per watt and higher rack densities
Data centers are not without criticism: power-intensive computing raises temperatures, and giant campuses consume considerable amounts of air and/or water to operate optimally.
Land scarcity and financial incentives also pushed new developments closer to high-risk areas, including drought-prone regions, which ultimately led to even higher cooling needs.
But Nvidia knows this, and knows that traditional air cooling has all but reached its limits as AI hardware becomes denser and denser.
Closed-loop cooling virtually eliminates water waste
With cooling now a critical part of AI infrastructure design, Nvidia’s latest liquid-cooled AI systems promise higher thresholds for reducing load, thereby reducing water and power consumption.
By running the coolant at higher temperatures – 45°C or 113°F, to be precise – this helped simplify cooling systems and reduce electricity consumption. Nvidia’s concept uses 75% water and 25% glycol as a coolant, noting that it can operate at around 5-7°C hotter than hot tubs.
Compared to traditional evaporative cooling towers, Nvidia’s latest proposal involves a closed-loop system in which coolant continuously circulates through servers to remove excess heat from the chips. The hot coolant then circulates through external dry coolers, leaving virtually no water evaporation.
The company boasted that cooling-related water consumption can be reduced by up to 100% in suitable climates subject to occasional extreme days, with the complete elimination of cooling towers.
“The Nvidia DSX reference design for AI factories has zero water consumption – we have eliminated massive amounts of power consumption and virtually all water consumption,” said Ali Heydari, director of data center cooling and infrastructure.
The system’s efficiency comes primarily from direct-on-chip liquid cooling, where liquid flows directly through cold plates attached to CPUs and GPUs. This captures and expels heat exactly where it is produced.
Not only is it more efficient than cooling entire rooms, but the liquid also promises to transfer heat thousands of times more efficiently than air.
Major improvements in water consumption, energy efficiency and energy use efficiency (PUE) all help improve sustainability, but the benefits are of a different order.
Nvidia claims it can increase rack density and performance
Nvidia has acknowledged that chip power consumption and rack density continue to increase. So, by implementing liquid-cooled data centers, companies can add more GPUs per rack, use higher rack power, and ultimately pack larger AI clusters into the same building footprint.
The company explained that its Rubin systems now fit into two racks, instead of six, representing a major space saving.
At the same time, air cooling became ineffective. “Once watts per chip rose above a certain level, liquid cooling became mandatory,” said Richard Whitmore, CEO of Motivair.
Independent testing entirely separate from Nvidia’s latest announcement shows that its H100 systems deliver approximately 17% better performance when water-cooled, compared to air-cooled. Under sustained AI workloads, GPU temperatures fluctuated between 41 and 50°C when water-cooled and between 54 and 72°C when air-cooled.
In addition to improving immediate and lasting performance, greater thermal efficiency could also increase longevity.
The new higher temperature closed-loop water cooling model is expected to be used in upcoming Rubin deployments this year.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds.
