NVIDIA Tensor Core GPUs Power HPE's Newest AI Supercomputer, Champollion
A brand new supercomputer born from a collaboration between Hewlett Packard Enterprise and NVIDIA, Champollion, will bring a powerful machine learning hub to French AI researchers.
Created on May 27|Last edited on May 27
Comment
Hewlett Packard Enterprise has unveiled a collaboration project with NVIDIA, the newest in AI supercomputers, Champollion. This supercomputer, named after Jean-François Champollion (known for the decipherment of Egyptian hieroglyphs), resides at the HPE Center of Excellence in France and will be available for use by researchers at several institutions across France as part of the French government-backed "AI for Humanity" initiative.
Powered by the latest NVIDIA technologies, Champollion will help lead the way for AI research based in France, and generally push the global AI community forwards.
What hardware does Champollion use?
Champollion sports 160 NVIDIA A100 Tensor Core GPUs linked together by NVIDIA Quantum InfiniBand, all housed within HPE Apollo 6500 Gen10 Plus Systems, allowing for up to 100 petaflops of performance across the full cluster.
The supercomputer is driven by HPE's machine-learning optimized software stack, HPE Machine Learning Development Environment, as well as being supported by NVIDIA's NGC Catalog of tools for accelerating the learning and inference processes.
Find out more
Add a comment
Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.