🚀 What drives 78% of the world’s TOP500 supercomputers? NVIDIA networking and GPUs are at the core of global innovation. ✅ Powering 388 systems on the new list ✅ Connecting 362 platforms with InfiniBand or Ethernet ✅ Accelerating 7 of the top 10 Green500 systems 👉 See how NVIDIA is leading the world in supercomputing performance: https://nvda.ws/4nUKxiB #SC25
NVIDIA Networking
Computer Networking Products
Santa Clara, California 45,410 followers
Accelerated Networks for Modern Workloads
About us
Data center networking is essential in confronting the challenges of modern-day computing. NVIDIA networking delivers accelerated software-defined networking, storage, security, and management services that are turning the data center into one compute unit.
- Website
-
https://www.nvidia.com/en-us/networking/
External link for NVIDIA Networking
- Industry
- Computer Networking Products
- Company size
- 10,001+ employees
- Headquarters
- Santa Clara, California
Updates
-
NVIDIA Networking is delivering 2,200 exaflops of AI performance and accelerating discovery at every scale for Argonne National Laboratory. Read more ⬇️ #SC25
ICYMI: ⚡️NVIDIA, Oracle, and U.S. Department of Energy (DOE) are powering breakthroughs with Solstice and Equinox at Argonne National Laboratory. With 110,000+ #NVIDIABlackwell GPUs, 2,200 exaflops of #AI performance, and NVIDIA networking technologies, we're bringing open science breakthroughs for AI, security, and energy. A new era for scientific discovery begins. Learn more ➡️ https://nvda.ws/4oNilQ8 #SC25
-
-
NVIDIA Quantum-X800 InfiniBand powers Japan’s next AI and quantum supercomputers — fueling research from life sciences to climate innovation. Read the News 👇
At #SC25, NVIDIA and RIKEN announced Japan’s next AI and #quantum supercomputers, powered by NVIDIA GB200 systems and Quantum-X800 InfiniBand networking. Together, RIKEN's new systems will feature over 2,000 #NVIDIABlackwell GPUs, supporting Japan's sovereign AI strategy and expanding their infrastructure to advance research in life sciences, climate and weather forecasting, quantum computing, manufacturing, and more. 🔗 Read the announcement now: https://nvda.ws/4oNgNFP
-
-
🧠 What does the next chapter of AI supercomputing look like? That's what everyone at #SC25 is asking and why NVIDIA BlueField-4 is the talk of the town. NVIDIA BlueField-4 DPUs—powered by a 64-core NVIDIA Grace CPU and NVIDIA ConnectX-9 SuperNIC—free CPUs/GPUs for compute, unlocking zero-trust security and real-time, multi-tenant protection. ✔️ Accelerates networking, storage, security ✔️ Drives intelligent, scalable AI factories Check out what leaders at SC25 are building with BlueField-4 and ConnectX-9: https://nvda.ws/4461X4O
-
-
🌐 “Networking is the fundamental future of all AI.” – John Furrier, Executive Analyst & Co-founder, SiliconANGLE & theCUBE In this episode, Gilad Shainer, SVP of Networking at NVIDIA, joins John Furrier to discuss how AI factories are transforming into data centers of the future. The conversation dives into: ✅ The crucial role of networking, compute, and storage integration in scaling AI workloads ✅ How co-packaged optics and energy-efficient design are redefining performance and reliability ✅ The evolution of AI infrastructure from data centers to the edge 🎥 Watch the full conversation: https://nvda.ws/3JSC69M
-
-
🔥 Reminder to register for the #SC25 Fireside Chat on Monday Hear from NVIDIA's Ian Buck, along with global #supercomputing leaders at Argonne, TACC, RIKEN, & Jülich as they explore breakthroughs in scientific computing. Register now 👇 https://nvda.ws/47GpfAA
-
-
Microsoft is expanding its Fairwater AI Superfactory network from Wisconsin to Atlanta, connecting datacenters built on NVIDIA Spectrum-X Ethernet with GB200 and GB300 NVL72 systems, driving a new level of scale and performance for AI training. Read the full news 👇
Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more: https://lnkd.in/e82Dm2jQ
-
🚀 What does it take to scale from a single GPU to 100,000—and eventually millions? Michael Kagan, CTO of NVIDIA and co-founder of Mellanox, shares how the $7B Mellanox acquisition reshaped NVIDIA from a chipmaker into the architect of modern AI infrastructure. 🔍 He breaks down how networking drives system performance at scale and why the next leap, toward million-GPU systems, requires rethinking architecture from the ground up. He also shares his view of AI as humanity’s “spaceship of the mind,” enabling discoveries we haven’t yet imagined. 🎥 Watch the full interview with Michael Kagan: https://nvda.ws/3JXlfm9
Nvidia CTO Michael Kagan: Scaling Beyond Moore's Law to Million-GPU Clusters
https://www.youtube.com/
-
NVIDIA Networking reposted this
In the latest #MLPerf Training v5.1 round, NVIDIA swept all seven benchmarks. The NVIDIA Blackwell Ultra GPU made its MLPerf Training debut, delivering large leaps for LLM pretraining and LLM fine-tuning performance, supercharged by the world’s first FP4 training submissions with NVFP4. Learn more: https://nvda.ws/4oxGv0T
-
NVIDIA Networking reposted this
📢 Join NVIDIA's Ian Buck along with global supercomputing leaders at Argonne National Laboratory, Texas Advanced Computing Center (TACC), RIKEN, & Jülich Supercomputing Centre (JSC) as they explore breakthroughs in scientific computing. Gain insights on how innovations in AI, #HPC, quantum computing, and next-generation infrastructure are redefining what’s possible at the intersection of supercomputing and science. 🔗 Register now for the #SC25 Fireside Chat: https://nvda.ws/4hWqSNO
-