In the realm of supercomputers, there are efforts to develop a high-performing machine that would be 30x more powerful than the current fastest supercomputer, Frontier.

As per reports, AMD, a California-based semiconductor company, has been approached by a client to create an AI training cluster of a 1.2 million GPUs.

Only a few thousand GPUs connected via high-speed interconnect across several local servers make current training clusters. Notably, Frontier has 37,888 Radeon GPUs.

Advanced AI training cluster

For the advanced AI training cluster, the client has approached AMD, which supplied only 2% of the data center GPU shipments in 2023. The market leader, Nvidia, supplied the other 98%.

Forrest Norrod, AMD’s GM of Datacenter Solutions, stated that very sober people are considering spending up to one hundred billion dollars on AI training clusters.

This shouldn’t come as a shock, as the past few years in the tech world have been defined by the explosion in AI advancements. It seems that companies are ready to invest significant sums in AI and machine learning to remain competitive, reported Tech Spot.

MI300 is the fastest ramping product of AMD

MI300 is claimed to be the fastest ramping product of AMD. According to the company, MI300 Series accelerators are uniquely well-suited to power even the most demanding AI and HPC workloads, offering exceptional compute performance, large memory density, high bandwidth memory, and support for specialized data formats.

Designed to deliver leadership performance for Generative AI workloads and HPC applications, AMD Instinct MI300 Series accelerators are built on AMD CDNA 3 architecture, which offers Matrix Core Technologies and support for a broad range of precision capabilities—from the highly efficient INT8 and FP8 (including sparsity support for AI), to the most demanding FP64 for HPC.

Forrest Norrod stated that AMD will keep making progress in its software. “We’re keen to keep making progress on the hardware. I feel really good about the hardware, I feel pretty good about the software roadmap as well – particularly because we have a number of very large customers that are helping us out.”

“And it’s clearly in their best interest to promote an alternative and to get differentiated product for themselves as well. So we’re going to try to harness the power of the open ecosystems as much as we possibly can, and grow it as fast as we can,” Norrod told The Next Platform.

AMD’s MI300X enhances Blender’s real-time viewport performance, making tasks such as sculpting, animation playback, and texture painting more fluid and responsive. And Nvidia’s H100 offers improvements in viewport interactivity, allowing for smoother manipulation of high-poly models and complex scenes.

While both GPUs are highly capable, the MI300X offers advantages in memory-intensive tasks like large scene rendering and simulations. In contrast, the H100 could excel in AI-enhanced workflows and ray-traced rendering performance, according to Cudo Compute.


The Blueprint Daily

Stay up-to-date on engineering, tech, space, and science news with The Blueprint.


Prabhat Ranjan Mishra Prabhat, an alumnus of the Indian Institute of Mass Communication, is a tech and defense journalist. While he enjoys writing on modern weapons and emerging tech, he has also reported on global politics and business. He has been previously associated with well-known media houses, including the International Business Times (Singapore Edition) and ANI.

Leave a Reply

Your email address will not be published. Required fields are marked *