Constructing and deploying AI purposes at scale requires a brand new class of computing infrastructure — one that may deal with the large quantities of knowledge, compute energy and networking bandwidth wanted by generative AI fashions.
To raised guarantee these fashions carry out optimally and effectively, NVIDIA is teaming with Cisco to allow enterprise generative AI infrastructure.
Cisco’s new Nexus HyperFabric AI cluster resolution, developed in collaboration with NVIDIA, offers a path for enterprises to operationalize generative AI. Cisco HyperFabric is an enterprise-ready, end-to-end infrastructure resolution to scale generative AI workloads. It combines NVIDIA accelerated computing and AI software program with Cisco AI-native networking and a sturdy VAST Knowledge Platform.
“Enterprise purposes are remodeling into generative AI purposes, considerably growing information processing necessities and total infrastructure complexity,” mentioned Kevin Wollenweber, senior vice chairman and normal supervisor of knowledge heart and supplier connectivity at Cisco. “Collectively, Cisco and NVIDIA are advancing HyperFabric to advance generative AI for the world’s enterprises to allow them to use their information and area experience to rework productiveness and perception.”
Powering an Enterprise-Prepared AI Cluster Resolution
Foundational to the answer are NVIDIA Tensor Core GPUs, which offer the accelerated computing wanted to course of large datasets. The answer makes use of NVIDIA AI Enterprise, a cloud-native software program platform that acts because the working system for enterprise AI. NVIDIA AI Enterprise streamlines the event and deployment of production-grade AI copilots and different generative AI purposes, making certain optimized efficiency, safety and software programming interface stability.
Included with NVIDIA AI Enterprise, NVIDIA NIM inference microservices speed up the deployment of basis fashions whereas making certain information safety. NIM microservices are designed to bridge the hole between complicated AI improvement and enterprise operational wants. As organizations throughout varied industries embark on their AI journeys, the mix of NVIDIA NIM and the Cisco Nexus HyperFabric AI cluster helps all the course of, from ideation to improvement and deployment of production-scale AI purposes.
The Cisco Nexus HyperFabric AI cluster resolution integrates NVIDIA Tensor Core GPUs and NVIDIA BlueField-3 SuperNICs and DPUs to boost system efficiency and safety. The SuperNICs provide superior community capabilities, making certain seamless, high-speed connectivity throughout the infrastructure. BlueField-3 DPUs offload, speed up and isolate the infrastructure companies, making a extra environment friendly AI resolution.
BlueField-3 DPUs may run safety companies just like the Cisco Hypershield resolution. It permits an AI-native, hyperdistributed safety structure, the place safety shifts nearer to the workloads needing safety. Cisco Hypershield is one other notable space of collaboration between the businesses, specializing in creating AI-powered safety options.
Be a part of NVIDIA at Cisco Stay
Study extra about how Cisco and NVIDIA energy generative AI at Cisco Stay — operating by way of June 6 in Las Vegas — the place the businesses will showcase NVIDIA AI applied sciences on the Cisco AI Hub and share finest practices for enterprises to get began with AI.
Attend these periods to find methods to speed up generative AI with NVIDIA, Cisco and different ecosystem companions:
- Keynote Deep Dive: “Harness a Daring New Period: Remodel Knowledge Middle and Service Supplier Connectivity” with NVIDIA’s Kevin Deierling and Cisco’s Jonathan Davidson, Kevin Wollenweber, Jeremy Foster and Invoice Gartner — Wednesday, June 5, from 1-2 p.m. PT
- AI Hub Theater Presentation: “Speed up, Deploy Generative AI Wherever With NVIDIA Inference Microservices” with Marty Jain, vice chairman of gross sales and enterprise improvement at NVIDIA — Tuesday, June 4, from 2:15-2:45 p.m. PT
- WWT AI Hub Sales space: Thought management interview with NVIDIA’s Jain and WWT Vice President of Cloud, Infrastructure and AI Options Neil Anderson — Wednesday, June 5, from 10-11 a.m. PT
- NetApp Theater: “Accelerating Gen AI With NVIDIA Inference Microservices on FlexPod” with Sicong Ji, strategic platforms and options lead at NVIDIA — Wednesday, June 5, from 1:30-1:40 p.m. PT
- Pure Storage Theater: “Accelerating Gen AI With NVIDIA Inference Microservices on FlashStack” with Joslyn Shakur, gross sales alliance supervisor at NVIDIA — Wednesday, June 5, from 2-2:10 p.m. PT
Join generative AI information to remain updated on the most recent breakthroughs, developments and applied sciences.
