Nvidia on June 22 announced the Vera Rubin supercomputing platform aimed at high-intensity HPC and AI workloads such as climate modeling, computational fluid dynamics and energy exploration, according to 36Kr. The platform combines Rubin GPUs with Vera CPUs and uses an integrated design featuring NVLink, InfiniBand and liquid cooling.
A single system is rated to deliver more than 7 exaflops of AI compute and about 5 PF FP64 scientific computing performance, and supports up to 144 GPUs per rack. Dell, HPE and Supermicro are also set to launch high-density supercomputing systems based on the architecture, with availability expected to begin rolling out from Q4 2026.
Nvidia Launches Vera Rubin Supercomputing Platform for HPC-AI Workloads
2026-06-23 00:07:40
Disclaimer:
1. The information provided does not constitute investment advice. Investors should make independent decisions and bear all risks themselves.
2. The copyright of this content belongs to the original author. The views expressed herein are solely those of the author and do not represent the stance or position of this website.
