Cisco Nexus Delivers New AI Innovations with NVIDIA

[ad_1] AI is reworking each {industry}—redefining how we innovate, compete, and create worth. This shift calls for a brand new class of knowledge middle infrastructure: one which’s high-performance, safe, and constructed to adapt to always altering workloads. The community is on the coronary heart of this transformation—it should evolve to match AI’s tempo, enabling organizations to unlock new efficiencies, insights, and alternatives.

The truth of AI networks As organizations scale their AI environments, they’re discovering that conventional networks merely weren’t constructed for the job. Managing east-west GPU-to-GPU visitors for AI coaching—whereas additionally supporting frontend, storage, and administration visitors—requires a basically new strategy.

Clients are asking for a unified working mannequin that spans on-premises and cloud, automation that works throughout open and built-in methods, and platforms tuned from silicon to software program for AI-scale efficiency. The frequent Nexus working mannequin—mixed with Nexus Dashboard and Nexus Hyperfabric—simplifies operations throughout numerous environments, giving community operators a single, constant technique to handle, automate, and acquire visibility throughout all their networks.

What’s wanted is a community designed to scale massively, adapt seamlessly to combined workloads, and guarantee constant efficiency—precisely what our new architectures ship.

<img loading="lazy" decoding="async" class="lazy lazy-hidden wp-image-480575 size-medium_large" data-lazy-type="image" src="https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1-768x379.png" alt="Diagram illustrating N9100 infrastructure with its features, foundational elements, such as software, optics, and a model. Benefits highlighted for GPU solution." width="768" height="379" srcset="" sizes="auto, (max-width: 768px) 100vw, 768px"/><img loading="lazy" decoding="async" class="wp-image-480575 size-medium_large" src="https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1-768x379.png" alt="Diagram illustrating N9100 infrastructure with its features, foundational elements, such as software, optics, and a model. Benefits highlighted for GPU solution." width="768" height="379" srcset="https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1-300x148.png 300w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1-1024x505.png 1024w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1-768x379.png 768w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1-1536x757.png 1536w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-1-1.png 2000w" sizes="auto, (max-width: 768px) 100vw, 768px"/> Determine 1: New calls for on AI community materials Constructing future-ready AI community infrastructure requires greater than efficiency—it requires a strong working mannequin. I’m thrilled to announce the enlargement of our Cisco Nexus portfolio and our partnership with NVIDIA.

Introducing Cisco N9100 Collection Switches Constructing on its February 2025 announcement on deepening its NVIDIA partnership to speed up AI adoption, Cisco is increasing its Nexus knowledge middle networking portfolio with the launch of the Cisco N9100 Collection Switches.

The Cisco N9100 Collection augments the broader Cisco Nexus portfolio, which incorporates switches powered by Cisco Silicon One and Cloud Scale silicon, by incorporating NVIDIA Spectrum-X Ethernet change silicon—all unified below the Nexus working mannequin for seamless integration with Cisco Nexus Dashboard.

With the N9100 Collection, clients additionally get the selection of working methods—Cisco’s confirmed NX-OS for enterprise-grade reliability and operational consistency, or SONiC for organizations that choose open networking requirements with automation and operations with Nexus Dashboard.

<img loading="lazy" decoding="async" class="wp-image-480576 size-medium_large" src="https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-2-1-768x360.png" alt="Image of the N9100 series switch" width="768" height="360" srcset="https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-2-1-300x141.png 300w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-2-1-1024x480.png 1024w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-2-1-768x360.png 768w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-2-1-1536x720.png 1536w, https://blogs.cisco.com/gcs/ciscoblogs/1/2025/10/AI-Scale-Data-Centers_blog_figure-2-1.png 2000w" sizes="auto, (max-width: 768px) 100vw, 768px"/> Determine 2: Introducing Cisco N9100 Collection Switches Reference architectures for ease of deployment To help the evolving wants of AI infrastructure, Cisco combines high-performance {hardware} and unified operations with confirmed reference architectures. This complete strategy ensures that clients can scale, adapt, and function with confidence.

Cisco Cloud Reference Structure with Silicon One

Our Cisco Silicon One platform stays the cornerstone for patrons requiring most architectural flexibility. With industry-leading scale and programmability, Silicon One delivers unmatched efficiency throughout numerous workloads. Cisco Cloud Reference Structure (CRA) with Nexus Dashboard and Nexus Hyperfabric AI showcases scalable networking for large-scale clusters. Nexus Hyperfabric AI, now typically out there, simplifies and accelerates AI journey for patrons with a simple begin operational expertise for deploying and managing full-stack AI clusters. It leverages Cisco Silicon One’s programmability to allow NVIDIA Spectrum-X Ethernet expertise on Cisco switches that adhere to NCP design tenants for neocloud and sovereign cloud clients.

“We’re proud to accomplice with Cisco to carry next-generation networking improvements into our AI-native framework,” states Tarun Raisoni, CEO and Co-Founding father of Gruve. “Gruve helps enterprises construct the infrastructure and intelligence layer that powers enterprise AI adoption. Cisco’s 800G Silicon One and NVIDIA Spectrum-X silicon-based Nexus switches set a brand new normal for efficiency, scalability, and programmability, enabling us and the ecosystem to scale the muse of contemporary AI infrastructure. Collectively, we’re constructing the clever, environment friendly, and future-ready connectivity that defines the period of AI.”

NCP reference structure compliance with N9100

Cisco will supply an NVIDIA Cloud Associate (NCP) compliant reference structure with the brand new Cisco N9100 Collection Switches for neocloud and sovereign cloud clients. These switches mix the strengths of NVIDIA AI expertise with Cisco enterprise-grade reliability and streamlined administration. Orderable earlier than the tip of 2025, it's typically out there early in 2026.

“BlueSky Compute is happy to be one of many first neoclouds to deploy Cisco’s N9100 Collection Switches for the scale-out cloth in our AI coaching clusters—powering our imaginative and prescient to show AI into ROI for the world’s enterprises by constructing bigger, extra interconnected B300 clusters, quicker. Cisco’s NCP compliant reference structure, based mostly on N9100 Collection Switches whereas sustaining the operational simplicity with Nexus Dashboard, is a game-changer.”—Ian Hartley, CEO, BlueSky Compute

Cisco and NVIDIA: Powering the subsequent wave of AI Cisco and NVIDIA’s partnership continues to speed up as we work collectively to empower AI at scale—securely and at ranges that may be noticed, managed, and trusted. The developments introduced at present with the Cisco N9100 Collection Switches which are NCP RA compliant and the Cisco CRA with Nexus Dashboard and Nexus Hyperfabric AI are a testomony to Cisco’s relentless pursuit of innovation to drive AI adoption throughout enterprises, neoclouds, and sovereign clouds. Cisco Safe AI Manufacturing unit with NVIDIA encompasses these improvements to ship complete framework the place compute, storage, safety, and observability work in live performance.

By constructing on a versatile, scalable, and safe community basis, you unlock new alternatives for AI innovation—enabling your group to appreciate its full potential of observable and future-ready AI success.