NVIDIA’s Ethernet Networking Platform for AI Available Soon
End-to-end platform features latest NVIDIA Spectrum-X networking for customers to transform business wiith AI.
November 21, 2023
NVIDIA announces that Dell Technologies, Hewlett Packard Enterprise (HPE) and Lenovo will be the first to integrate NVIDIA Spectrum-X Ethernet networking technologies for artificial intelligence into their server lineups.
Purpose-built for generative AI, Spectrum-X offers enterprises a new class of Ethernet networking for higher networking performance for AI communication, NVIDIA reports.
The new systems from these three system makers bring together Spectrum-X with NVIDIA Tensor Core GPUs, NVIDIA AI Enterprise software and NVIDIA AI Workbench software.
“Generative AI and accelerated computing are driving a generational transition as enterprises upgrade their data centers to serve these workloads,” says Jensen Huang, founder and CEO of NVIDIA. “Accelerated networking is the catalyst for a new wave of systems from NVIDIA’s leading server manufacturer partners to speed the shift to the era of generative AI.”
“Accelerated computing and networking are key to building systems to meet the demands of large language models and generative AI applications,” says Michael Dell, chairman and CEO of Dell Technologies. “Through our collaboration, Dell Technologies and NVIDIA are providing customers with the infrastructure and software needed to quickly and securely extract intelligence from their data.”
“Generative AI will undoubtedly drive innovation across multiple industries,” says Antonio Neri, president and CEO of HPE. “These powerful new applications will require a fundamentally different architecture to support a variety of dynamic workloads.”
“Generative AI can power unprecedented transformation but places unprecedented demands on enterprise infrastructure,” says Yuanqing Yang, chairman and CEO of Lenovo. “Working closely with NVIDIA, Lenovo is building efficient, accelerated systems with the networking, computing and software needed to power modern AI applications.”
Networking Purpose-Built to Accelerate AI
Spectrum-X combines the performance of the Spectrum-4 Ethernet switch; the NVIDIA BlueField-3 SuperNIC, a class of network accelerators for supercharging hyperscale AI workloads; as well as acceleration software. Spectrum-X complements BlueField-3 DPUs, an infrastructure computing platform.
Spectrum-4 is a 51Tb/sec Ethernet switch for AI, providing data throughput at scale and under load. Its intelligent, fine-tuned routing technology enables maximum use of network infrastructure at all times.
BlueField-3 SuperNICs are designed for network-intensive, parallel computing, offering up to 400Gb/s RDMA over Converged Ethernet (RoCE) network connectivity between GPU servers. They also enable secure, multi-tenant data center environment. BlueField-3 SuperNICs are made for enterprise-class servers.
NVIDIA AI Enterprise provides frameworks, pretrained models and development tools for secure, stable and supported production AI. NVIDIA AI Workbench allows developers to quickly create, test and customize pretrained generative AI models on a PC or workstation — then scale them to virtually any data center or cloud.
NVIDIA Israel-1 Supercomputer
Spectrum-X also enables the NVIDIA Israel-1 supercomputer, a reference architecture for next-generation AI systems. Israel-1 is a collaboration with Dell Technologies, using Dell PowerEdge XE9680 servers powered by the NVIDIA HGX H100 eight-GPU platform and BlueField-3 DPUs and SuperNICs with Spectrum-4 switches.
New systems from Dell, HPE and Lenovo featuring the complete NVIDIA AI stack are expected in the first quarter of next year.
Sources: Press materials received from the company and additional information gleaned from the company’s website.
More Dell Coverage
More Hewlett Packard Enterprise Coverage
More Lenovo Coverage
More NVIDIA Coverage
About the AuthorDE Editors
DE’s editors contribute news and new product announcements to Digital Engineering.
Press releases may be sent to them via DE-Editors@digitaleng.news.