Skip to main content

Nvidia’s new Polaris supercomputer to tackle climate, cancer, and more

Nvidia pulled back the curtain on a new supercomputer that’s set to take on some of the world’s biggest problems. It’s called Polaris, and Nvidia says it will be at Argonne National Laboratory near Chicago to tackle issues like climate change and offer insight into fields like neuroscience in 2022.

Polaris can handle complex simulations, charting, and machine learning — and that’s only part of what it can do. Nvidia says the machine is capable of 1.4 exaflops of theoretical A.I. performance and over 44 petaflops of FP64 performance. Those numbers put it within the top 10 most powerful supercomputers in the world based on the most recent TOP500 list, which tracks supercomputers around the globe.

Nvidia A100 accelerator.
Image used with permission by copyright holder

For context, Nvidia’s recently announced DGX SuperPod is capable of around 100 petaflops of A.I. performance, and it’s designed for large companies handling complex A.I. problems. Polaris is part of the Argonne Leadership Computing Facility (ALCF), which provides computing resources to researchers. Simply put, a lot of people will be using Polaris, so it needs a lot of power.

Recommended Videos

“Harnessing the huge number of Nvidia A100 GPUs will have an immediate impact on our data-intensive and A.I. HPC workloads, allowing Polaris to tackle some of the world’s most complex scientific problems,” ALCF Director Michael E. Papka said.

At the heart of Polaris are 2,240 Nvidia A100 Tensor Core GPUs. These are the same cores that power the Grace supercomputer at Texas A&M University. The A100 core is powered by the same Ampere architecture in RTX 3000-series graphics cards, but tuned for scientific research, A.I., and data analytics.

Hewlett Packard (HP) is building the machine, and Nvidia says it will be available sometime in early 2022. It’s broken up into 560 nodes, each of which contain four A100 GPUs. In a press release, Nvidia said researchers will put these GPUs to use for things like exploring cancer treatments, finding sources of clean energy, and running complex physics simulations.

Polaris isn’t the first GPU-accelerated system Argonne has deployed. Last year, the research institute used a GPU-accelerated supercomputer to research COVID-19. In a press briefing, Nvidia said that Polaris is an extension of this type of machine, and that it helps bridge the gap to Argonne’s upcoming Aurora exascale computer.

The Aurora supercomputer.
Image used with permission by copyright holder

Exascale computing is an emerging concept, noting the number of floating point operations per second (FLOPS) that a supercomputer is capable of. For exascale, the computer needs at least 1 exaflop of power, or 1 quintillion FLOPS. Aurora was originally set to be the world’s first exascale supercomputer, but it has suffered numerous delays.

Exascale was first achieved by the Folding@Home project in March 2020. Instead of a single machine, Folding@Home took advantage of volunteer computing resources to reach the exascale mark. With Aurora’s delays, the first exascale supercomputer in the U.S. is set to be Frontier, which is scheduled to go online this year.

Jacob Roach
Former Digital Trends Contributor
Jacob Roach is the lead reporter for PC hardware at Digital Trends. In addition to covering the latest PC components, from…
Nvidia might break with tradition for the RTX 5060
Two RTX 4060 graphics cards stacked on top of each other.

Although Nvidia has already established the flagship RTX 5090 as one of the best graphics cards you can buy, most PC gamers are eagerly waiting for Team Green's more budget-conscious offerings. According to a new rumor, Nvidia's RTX 5060 and RTX 5060 Ti will ditch the 16-pin power connector that Nvidia has used for the past few generations in favor of a standard 8-pin power connector.

The rumor comes from Brother Pan Talks Computers, a Chinese media outlet that VideoCardz reports has ties to Zotac. Nvidia has yet to announce the RTX 5060 and RTX 5060 Ti, but they'll be some of the most important entries in Nvidia's latest RTX 50-series lineup. Traditionally, Nvidia's 60-class graphics cards are among the most popular GPUs on the market, and last-gen's RTX 4060 still tops the charts in the Steam hardware survey.

Read more
Nvidia’s new GPUs show up in prebuilts, but the RTX 5090 is missing
iBUYPOWER RTX for AI PCs side view of pre-built on sale hero

Nvidia's upcoming RTX 5080 and RTX 5070 Ti just appeared in several iBUYPOWER gaming PCs. This is the first U.S. retailer to list Nvidia's RTX 50-series in prebuilt systems. The listings are interesting, with performance figures that really don't add up. Still, the biggest question is: Where's the GPU that's bound to beat all the current best graphics cards? Yes, we're talking about RTX 5090.

The listings have already been taken down, but they were preserved by VideoCardz. A total of five systems were listed by iBUYPOWER, but they all contained the same two GPUs -- either the RTX 5080 or the RTX 5070 Ti. Both cards are said to come with 16GB of memory, and we expect them to be announced on January 6 during the CES 2025 keynote held by Nvidia's CEO, Jensen Huang.

Read more
Nvidia may have found a new way to bypass GPU export restrictions
The RTX 4090 graphics card sitting on a table with a dark green background.

Chinese gamers are expected to have limited access to Nvidia's best graphics card due to strict export restrictions. However, Nvidia may have found a way around it without cutting down its flagship GPU. A new leak suggests that the RTX 5090D will have the same hardware specifications as the worldwide version, and the solution lies in firmware adjustments.

As a quick refresher, the "D" in RTX 5090D stands for "Dragon." It marks GPUs made by Nvidia to bypass the export restrictions imposed by the U.S. on China, which limit the sale of high-performance graphics processors. We first saw it appear in the RTX 4090D as a remedy for the fact that the base RTX 4090 exceeds the performance thresholds set by these regulations.

Read more