NVIDIA has announced its Vera Rubin NVL4 system, designed for demanding HPC and AI workloads like climate modeling and fluid dynamics. This platform integrates Rubin GPUs with Vera CPUs, featuring NVLink, InfiniBand, and liquid cooling for a unified design. It offers over 7 exaflops of AI compute and approximately 5 PF of FP64 scientific computing power, with a high density of up to 144 GPUs per rack. Major manufacturers including Dell, HPE, and Supermicro will offer high-density supercomputing systems based on this architecture, with availability expected in Q4 2026. AI
IMPACT This system's high AI compute power and integrated design could accelerate advancements in complex fields like climate modeling and energy exploration.
RANK_REASON NVIDIA announced a new supercomputing platform with specific compute capabilities and a release date. [lever_c_demoted from frontier_release: ic=2 ai=1.0]
Read on Mastodon — fosstodon.org →
- Dell
- Hewlett Packard Enterprise
- InfiniBand
- NVIDIA
- NVLink
- Rubin GPU
- Supermicro
- Vera CPU
- Vera Rubin NVL4 System
- high-performance computing
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
