NVIDIA A100 PCIe GPU Accelerator Gets Liquid-Cooled, Offers Same Performance While Consuming 30% Less Energy


NVIDIA has unveiled its brand new A100 PCIe GPU accelerator which now comes with a liquid-cooling option for higher energy efficiency.

NVIDIA Adds Liquid-Cooled A100 PCIe GPUs for Sustainable, Efficient Computing

Press Release: Born in the mainframe era, liquid cooling is maturing in the age of AI. It’s now widely used inside the world’s fastest supercomputers in a modern form called direct-chip cooling. Liquid cooling is the next step in accelerated computing for NVIDIA GPUs that already deliver up to 20x better energy efficiency on AI inference and high-performance computing jobs than CPUs.

NVIDIA Launches GeForce GTX 1630: An Entry-Level Graphics Card For The Sub-$150 US GPU Segment

Efficiency Through Acceleration 

If you switched all the CPU-only servers running AI and HPC worldwide to GPU-accelerated systems, you could save a whopping 11 trillion watt-hours of energy a year. That’s like saving the energy more than 1.5 million homes consume in a year.

Today, NVIDIA adds to its sustainability efforts with the release of our first data center PCIe GPU using direct-chip cooling. Equinix is qualifying the A100 80GB PCIe Liquid-Cooled GPU for use in its data centers as part of a comprehensive approach to sustainable cooling and heat capture. The GPUs are sampling now and will be generally available this summer.

Saving Water and Power

“This marks the first liquid-cooled GPU introduced to our lab, and that’s exciting for us because our customers are hungry for sustainable ways to harness AI,” said Smith.

Colorful’s GeForce GTX 1630 Graphics Card Leaks Out, Entry-Level Design With Dual-Fan Cooler

Datacenter operators aim to eliminate chillers that evaporate millions of gallons of water a year to cool the air inside data centers. Liquid cooling promises systems that recycle small amounts of fluids in closed systems focused on key hot spots.

Same Performance, Less Power

In separate tests, both Equinix and NVIDIA found a data center using liquid cooling could run the same workloads as an air-cooled facility while using about 30 percent less energy. NVIDIA estimates the liquid-cooled data center could hit 1.15 PUE, far below 1.6 for its air-cooled cousin.

Liquid-cooled data centers can pack twice as much computing into the same space, too. That’s because the A100 GPUs use just one PCIe slot; air-cooled A100 GPUs fill two

At least a dozen system makers plan to incorporate these GPUs into their offerings later this year. They include ASUS, ASRock Rack, Foxconn Industrial Internet, GIGABYTE, H3C, Inspur, Inventec, Nettrix, QCT, Supermicro, Wiwynn, and xFusion.

A Global Trend

Regulations setting energy-efficiency standards are pending in Asia, Europe, and the U.S. That’s motivating banks and other large data center operators to evaluate liquid cooling, too. And the technology isn’t limited to data centers. Cars and other systems need it to cool high-performance systems embedded inside confined spaces.

The Road to Sustainability

“This is the start of a journey,” said Smith of the debut of liquid-cooled mainstream accelerators.

Indeed, we plan to follow up the A100 PCIe card with a version next year using the H100 Tensor Core GPU based on the NVIDIA Hopper architecture. We plan to support liquid cooling in our high-performance data center GPUs and our NVIDIA HGX platforms for the foreseeable future.

For fast adoption, today’s liquid-cooled GPUs deliver the same performance for less energy. In the future, we expect these cards will provide an option of getting more performance for the same energy, something users say they want.