NVIDIA introduces Vera Rubin for scientific computing

Key Takeaways

• NVIDIA introduces the Vera Rubin platform for scientific computing, combining native double-precision performance, NVIDIA CUDA-X libraries, and AI platform software components
• The platform supports AI, simulation, and data-intensive research in a rack-scale system for scientific and industrial computing
• Vera Rubin provides over 7 exaflops of AI performance, 5 petaflops of native FP64 performance, and memory bandwidth across configurations with up to 144 GPUs
• The platform is designed for simulation, AI, and data analytics workloads, including climate modeling, computational fluid dynamics, and energy exploration
• Supercomputing centers and industrial users are planning systems based on Vera Rubin for AI and high-performance computing workloads

Introduction to NVIDIA Vera Rubin

NVIDIA has announced the Vera Rubin platform, a comprehensive solution for scientific computing that integrates native double-precision (FP64) performance, NVIDIA CUDA-X libraries, and the software components of the NVIDIA AI platform. This platform is designed to support AI, simulation, and data-intensive research in a rack-scale system for scientific and industrial computing.

Vera Rubin Architecture and Performance

The Vera Rubin platform combines NVIDIA GPUs and CPUs connected through NVIDIA NVLink-C2C, NVIDIA ConnectX-9 SuperNICs, and NVIDIA BlueField-4 DPUs in a direct liquid-cooled architecture. This platform provides native FP64 support for simulation workloads that require numerical precision, along with AI performance for surrogate models, scientific foundation models, and AI-assisted analysis. With over 7 exaflops of AI performance, 5 petaflops of native FP64 performance, and memory bandwidth across configurations with up to 144 GPUs, Vera Rubin is capable of handling demanding workloads such as climate modeling, computational fluid dynamics, and energy exploration.

Comparison of Vera Rubin with Other Supercomputing Systems

Supercomputing System Deployments

Supercomputing centers and industrial users are planning systems based on Vera Rubin for AI and high-performance computing workloads. For example, the Leibniz Supercomputing Centre (LRZ) is developing the Blue Lion system, which will use the NVIDIA Vera Rubin platform and second-generation exascale-class HPE Cray architecture. Scheduled to come online in 2027, the system is expected to provide about 30 times the computing power of LRZ's current system and support research in astrophysics, environmental science, and life sciences.

Bottom Line

The NVIDIA Vera Rubin platform is a powerful solution for scientific computing, providing native double-precision performance, AI performance, and memory bandwidth to support demanding workloads such as simulation, AI, and data analytics. With its ability to handle large models, increase simulation detail, and reduce processing time, Vera Rubin is poised to revolutionize the field of scientific computing and enable new breakthroughs in research and industry. As supercomputing centers and industrial users begin to deploy systems based on Vera Rubin, we can expect significant advancements in fields such as climate modeling, energy exploration, and life sciences.