Spiria logo.

Microsoft and Nvidia team up on supercomputing

November 16, 2022.

Nvidia H100 SXM GPU.

H100 SXM. © Nvidia.

Nvidia announced a partnership with Microsoft to build a supercomputer focused on artificial intelligence that will be among the most powerful in the world. This machine will use tens of thousands of high-end Nvidia GPUs to operate applications such as deep learning and large language models. It will use what is arguably the most powerful GPU in the world, the Hopper H100, launched by Nvidia in October, as well as A100 GPUs. Its architecture will be based on the Quantum-2 InfiniBand network platform, which transfers data at 400 gigabits per second between servers and links them in a powerful cluster. For its part, Microsoft will contribute its Azure cloud infrastructure and its ND- and NC-series virtual machines.

“AI technology advances as well as industry adoption are accelerating. The breakthrough of foundation models has triggered a tidal wave of research, fostered new startups and enabled new enterprise applications,” said Nvidia vice president of enterprise computing Manuvir Das. “Our collaboration with Microsoft will provide researchers and companies with state-of-the-art AI infrastructure and software to capitalize on the transformative power of AI.”

Ars Technica, Benj Edwards, “Nvidia and Microsoft team up to build massive AI cloud computer.”