NVIDIA Unveils Vera Rubin Platform: Single-Rack Supercomputer to Deliver 7 Exaflops of AI Horsepower

BlockBeats News, June 22nd, NVIDIA announced the launch of the Vera Rubin Supercomputing Platform, targeting the scientific and high-performance computing (HPC) fields. The single-rack system can provide over 7 Exaflops of AI computing power and 5 Petaflops of FP64 double-precision performance, officially described as "achieving TOP500-level supercomputing capability with a single rack."

The platform integrates the Rubin GPU and Vera CPU architecture, combined with technologies such as NVLink-C2C, ConnectX-9 SuperNIC, and BlueField-4 DPU, supporting the unified computing needs of AI training, scientific simulations, and data-intensive research.

NVIDIA stated that Vera Rubin will focus on high-complexity scientific computing tasks such as climate modeling, computational fluid dynamics, quantum chemistry, and energy exploration, supporting an integrated workflow of "AI + Simulation + Data Analytics."

At the application level, institutions such as the Leibniz Supercomputing Center, the National Energy Research Scientific Computing Center (NERSC) of the U.S. Department of Energy, and Los Alamos National Laboratory have planned to adopt this platform to build next-generation supercomputing systems.

In addition, vendors such as Bull, Dell Technologies, HPE, and Supermicro will also introduce high-density liquid-cooled rack-based supercomputing solutions based on Vera Rubin. NVIDIA CEO Jensen Huang stated that the platform aims to integrate AI, simulation, and data processing into a "new scientific tool," accelerating scientific discovery and industrial innovation.