Sign In 
Last Updated on November 28, 2025 by SmartNet
The AI infrastructure landscape transformed dramatically in 2025 as Cisco and NVIDIA forged an expanded partnership combining hardware, software, and professional certification programs. This collaboration addresses the critical skills gap threatening to slow enterprise AI adoption while delivering integrated infrastructure solutions that dramatically simplify AI deployment.
Between March and November 2025, the partnership produced validated reference architectures, new networking hardware, and comprehensive certification pathways that are reshaping how enterprises build and manage AI infrastructure. This comprehensive analysis explores these developments and their implications for AI infrastructure professionals.
Watch: Cisco and NVIDIA Unveil Secure AI Factory Architecture
See the complete Cisco Secure AI Factory with NVIDIA demonstration, including the new N9100 switch, Nexus Hyperfabric AI management, and integrated security features. This video from Cisco’s GTC 2025 presentation explains how enterprises can deploy GPU clusters using validated reference architectures. Watch the announcement:
The Strategic Partnership: Addressing Enterprise AI Infrastructure Challenges
The expanded Cisco-NVIDIA partnership announced in February 2025 and unveiled at GTC in March directly responds to enterprise struggles deploying production AI systems.
Organizations face mounting pressure to implement AI capabilities but encounter significant barriers. AI infrastructure complexity overwhelms traditional IT teams lacking specialized GPU computing knowledge. Skills shortages prevent companies from building and maintaining AI systems effectively. Security concerns around AI data and model protection require new architectural approaches. Integration challenges emerge when connecting AI systems with existing datacenter infrastructure.
Chuck Robbins, Cisco’s Chair and CEO, framed the partnership’s significance: “AI can unlock groundbreaking opportunities for the enterprise. To achieve this, the integration of networking and security is essential. Cisco and NVIDIA’s trusted, innovative solutions empower our customers to harness AI’s full potential simply and securely.”
Jensen Huang, NVIDIA’s founder and CEO, emphasized the security imperative: “AI factories are transforming every industry, and security must be built into every layer to protect data, applications and infrastructure. Together, NVIDIA and Cisco are creating the blueprint for secure AI—giving enterprises the foundation they need to confidently scale AI while safeguarding their most valuable assets.”
The partnership combines Cisco’s networking and security expertise with NVIDIA’s GPU computing and AI software leadership, creating end-to-end solutions that address the full stack of AI infrastructure requirements.
Cisco Secure AI Factory with NVIDIA: Integrated Architecture
The centerpiece of the collaboration is the Cisco Secure AI Factory with NVIDIA, unveiled at GTC in March 2025 and continuously enhanced through the year.
This architecture integrates multiple technology layers into cohesive solutions. Compute platforms feature Cisco UCS AI servers based on NVIDIA HGX and MGX for accelerated computing, providing the GPU resources powering model training and inference. Networking infrastructure includes Cisco Nexus Hyperfabric AI and Nexus networking solutions powered by Silicon One, combined with NVIDIA Spectrum-X Ethernet networking for high-bandwidth, low-latency GPU cluster communication.
Storage integration brings high-performance storage from certified partners Pure Storage, Hitachi Vantara, NetApp, and VAST Data, ensuring data pipelines can feed GPUs at the speeds required for efficient training. Security and observability embed throughout the stack with solutions like Cisco AI Defense, Hybrid Mesh Firewall, and integration with NVIDIA NeMo Guardrails.
The architecture offers flexible deployment models. The Nexus Hyperfabric AI option provides cloud-managed infrastructure with Cisco 6000 Series data center switches managed by the Nexus Hyperfabric AI controller. Alternative on-premises configurations use Cisco Nexus 9000 Series Switches with traditional datacenter management approaches. Both options leverage validated reference architectures that enterprises can deploy directly or adapt to their specific requirements.
The reference architecture approach dramatically reduces deployment complexity. Cisco and NVIDIA build and test complete configurations internally before releasing them to customers. Organizations can replicate these validated designs one-to-one, bringing infrastructure online in hours rather than weeks. Alternatively, they can extract elements of the reference architecture and integrate them with existing infrastructure components.
Packaged offerings based on the Secure AI Factory architecture became available before the end of 2025, with many individual components like Cisco 9000 and 6000 switches shipping throughout the year.
Revolutionary Hardware: Cisco N9100 Series Switch
October 2025 brought a hardware milestone with the Cisco N9100 Series switch launch, the first NVIDIA partner-developed datacenter switch based on NVIDIA Spectrum-X Ethernet switch silicon.
The N9100 delivers impressive specifications addressing AI networking requirements. With 51.2 Tbps of bandwidth, it handles the massive data transfers between GPU servers during distributed training. The switch combines NVIDIA Spectrum-X Ethernet silicon with Cisco’s management capabilities, offering customers flexibility in operating system choice between Cisco NX-OS and open-source SONiC.
This switch enables Cisco to offer NVIDIA Cloud Partner-compliant reference architectures, critical for neocloud and sovereign cloud deployments. The NCP certification ensures the infrastructure meets NVIDIA’s stringent requirements for multi-tenant AI clusters with thousands of GPUs.
The N9100 integrates NVIDIA BlueField data processing units and ConnectX SuperNICs, handling security and data transfer at high speeds. Everything operates through unified management via Cisco’s Nexus Dashboard, providing IT teams with consistent oversight across their AI infrastructure.
Early customer adoption validates the approach. Blue Sky Compute became one of the first customers deploying the N9100-based reference architecture, demonstrating real-world viability.
Nexus Hyperfabric AI: Simplified Cluster Management
Cisco’s Nexus Hyperfabric AI, becoming orderable in November 2025, represents a paradigm shift in AI infrastructure management.
Traditional GPU cluster deployment requires weeks of specialized engineering, manually configuring switches, servers, GPUs, and storage systems. Hyperfabric AI automates this entire process through cloud-managed infrastructure that provisions clusters as complete units.
The platform provides turnkey capabilities comparable to managed cloud services but for on-premises and colocation deployments. Install it in a colocation facility, and you have a managed AI cloud cluster at your disposal. For enterprises, this means less custom engineering and faster time to production.
Key capabilities include auto-configured network fabrics that provision switches, servers, and GPUs seamlessly. Unified dashboards provide monitoring and scaling across compute, networking, and storage. Deployment times compress from weeks to hours. The system scales efficiently as organizations expand their AI capabilities.
Customer testimonials highlight the impact. Dan Mons, CTO at Sharon AI, noted: “Sharon AI is excited to test Cisco Nexus Hyperfabric AI because it offers a turnkey, cloud-managed, full-stack AI infrastructure solution that simplifies deployment and accelerates AI innovation, perfectly aligning with our mission to power sovereign AI solutions and AI factories.”
Rajeev Khanolkar, Chief Strategy Officer at Gruve Inc, added: “As a Cisco partner, Gruve views Cisco Nexus HyperFabric AI as a major step forward in simplifying AI infrastructure—integrating compute nodes, fabric interconnects, and VAST storage into a cohesive, scalable framework. Our partnership with Cisco enables us to accelerate customer deployments, enhance throughput, and ensure secure, high-performance AI operations.”
Addressing the AI Skills Gap: New Certification Programs
Parallel to hardware innovations, Cisco launched comprehensive certification programs addressing the critical shortage of AI infrastructure professionals.
The AI Technical Practitioner certification, announced in November 2025 with full availability by mid-December, targets network engineers, system administrators, solution architects, and IT professionals who need to understand how AI impacts enterprise infrastructure. The certification provides foundational knowledge and practical experience with AI technologies and workflows, focusing on technical integration of AI across enterprise environments.
Training covers infrastructure design for AI workloads, security considerations specific to AI systems, data management for AI applications, and AI operations and monitoring. The program rolls out in phases, with the Generative AI Essentials module available immediately through Cisco U. Additional modules addressing AI security and compliance, model customization, and AI agents and orchestration follow throughout late 2025 and early 2026.
The AI Infrastructure Specialist Certification sits within the CCNP Data Center track, targeting engineers, architects, operations teams, and service providers running AI workloads reliably at scale. This professional-level certification validates comprehensive knowledge in designing, implementing, operating, and troubleshooting AI solutions across Cisco infrastructure.
The certification exam first became available in February 2026 at Cisco Live Amsterdam, with Cisco recommending candidates complete the AI Solutions on Cisco Infrastructure Essentials Learning Path beforehand. This certification specifically addresses deploying, migrating, operating, monitoring, and troubleshooting AI-based workloads on Cisco infrastructure.
The AI Business Practitioner badge complements technical certifications, providing business-focused AI education. Available from November 2025, it helps non-technical professionals understand AI’s business implications and work effectively with technical teams.
These certifications establish industry benchmarks for validating AI infrastructure expertise. As organizations struggle to find qualified AI infrastructure professionals, Cisco’s certification programs provide clear pathways for existing IT professionals to transition into AI-focused roles.
Security-First Architecture: Cisco AI Defense Integration
Security concerns frequently delay enterprise AI deployments. The Cisco-NVIDIA partnership addresses this through integrated security capabilities embedded throughout the infrastructure stack.
Cisco AI Defense, launched earlier in 2025, discovers shadow AI usage by scanning cloud environments for LLM API calls across AWS, Azure, and GCP. By October 2025, it integrated with NVIDIA NeMo Guardrails, delivering robust cybersecurity for AI applications. The integration enables security and AI teams to protect AI models and applications while limiting sensitive data exposure outside organizational boundaries.
AI Defense now supports on-premises data-plane deployment, crucial for enterprises requiring complete control over sensitive AI workloads. This enables organizations to protect AI models and applications within their own datacenters, addressing data sovereignty and compliance requirements.
The architecture also incorporates Cisco Hypershield running on embedded data processing units in switches, providing scalable firewall services directly within the switching fabric. This transforms datacenter security from perimeter-based approaches to AI-native, hardware-accelerated, distributed security fabric.
Splunk Observability Cloud and Enterprise Security provide comprehensive monitoring and security analytics. Teams gain real-time insights into AI infrastructure health, performance metrics across the AI application stack, security event detection and response, and cost tracking for expensive GPU resources.
Market Impact and Customer Adoption
The Cisco-NVIDIA partnership’s market impact became evident through multiple indicators throughout 2025.
Cisco reported over $2 billion in AI-related orders during fiscal year 2025, underscoring strong market demand. Analysts anticipate a multi-year growth phase for Cisco driven by enterprises renewing and upgrading networks specifically for AI workloads.
Customer adoption spans multiple segments. ClusterPower, a European cloud services provider, committed to the partnership to drive datacenter operations with AI/ML solutions foundational for client infrastructure. Blue Sky Compute deployed early N9100-based reference architectures. Sharon AI and Gruve Inc actively test and deploy Hyperfabric AI solutions.
The partnership attracted ecosystem participation beyond Cisco and NVIDIA. Storage vendors Pure Storage, Hitachi Vantara, NetApp, and VAST Data provide certified storage solutions. Red Hat contributes containerization and automation offerings optimizing AI and containerized workloads. This ecosystem approach ensures customers access complete, validated solutions rather than cobbling together components from multiple vendors.
Industry analysts emphasize the partnership’s strategic significance. The shift to “AgenticOps” positions networking providers like Cisco at the forefront as critical infrastructure becomes a key AI driver. The consensus view holds that AI-ready networks have moved from theoretical to present reality, with Cisco leading this transformation.
Future Directions: 6G and AI-Native Infrastructure
The Cisco-NVIDIA collaboration extends beyond immediate datacenter needs into next-generation connectivity.
October 2025 brought unveiling of the industry’s first AI-native wireless stack for 6G, developed by Cisco, NVIDIA, and additional telecom partners. This initiative integrates sensing and communication technologies into mobile infrastructure, paving the way for AI capabilities at the network edge.
The 6G work reflects a broader vision of truly AI-native infrastructure that is self-optimizing and deeply integrated with AI capabilities. Rather than retrofitting AI onto existing systems, future infrastructure will embed AI throughout, enabling autonomous operations, predictive maintenance, and intelligent resource allocation.
Cisco envisions infrastructure that automatically optimizes itself based on workload patterns, predicts and prevents failures before they impact operations, and intelligently allocates resources across heterogeneous compute resources. This AI-native approach represents the logical evolution from current AI-ready infrastructure.
What This Means for AI Infrastructure Professionals
The Cisco-NVIDIA partnership developments create both opportunities and imperatives for infrastructure professionals.
Demand for AI infrastructure expertise continues exceeding supply. Organizations need professionals who understand GPU computing, high-performance networking, and ML operations. The certification programs provide clear pathways for acquiring validated skills.
Existing network engineers and datacenter professionals possess foundational knowledge transferable to AI infrastructure. Understanding Ethernet networking translates to configuring Spectrum-X fabrics. Experience with datacenter operations applies to GPU cluster management. The leap to AI infrastructure is substantial but achievable for motivated professionals.
The validated reference architectures lower barriers to AI infrastructure deployment. Organizations no longer need to assemble specialized expertise to design every aspect of their AI infrastructure. Following Cisco-NVIDIA reference architectures provides proven blueprints, allowing infrastructure teams to deploy confidently while learning AI-specific technologies.
Security and operations skills become increasingly critical. As AI moves from experimentation to production, organizations need professionals who can secure AI systems, monitor performance, troubleshoot issues, and optimize resource utilization. These operational capabilities often prove more valuable than cutting-edge technical knowledge.
Continuous learning remains essential in this rapidly evolving field. Hardware, software, and best practices advance constantly. Professionals committed to ongoing education through certifications, hands-on experimentation, and community participation will find themselves well-positioned for long-term success.
Preparing for AI Infrastructure Careers
Infrastructure professionals can take concrete steps to position themselves for AI infrastructure opportunities.
Pursue relevant certifications strategically. Cisco’s AI Technical Practitioner certification provides accessible entry for networking professionals. Cloud platform AI certifications from AWS, Azure, or Google Cloud demonstrate platform-specific expertise. NVIDIA’s AI Infrastructure and Operations certifications validate GPU computing knowledge.
Gain hands-on experience with AI workloads. Many organizations deploy AI systems without specialized infrastructure support. Volunteering to support these initiatives provides valuable exposure. Cloud providers offer free credits and trials for experimenting with GPU instances and AI services.
Build foundational knowledge systematically. Understanding Kubernetes becomes essential for modern AI deployments. Familiarity with containerization and orchestration enables supporting MLOps platforms. Python scripting ability facilitates automation and tooling development.
Engage with the AI infrastructure community. Online forums, local meetups, and virtual conferences provide learning opportunities and professional connections. Following thought leaders and practitioners on social media surfaces emerging practices and technologies.
For comprehensive training covering AI infrastructure fundamentals, GPU computing, MLOps platforms, and deployment patterns, SmartNet Academy’s AI Infrastructure and Operations Training provides practical education designed for infrastructure professionals entering AI roles. The program addresses the full technology stack that working engineers encounter in production AI environments.
Complete the AI Infrastructure and Operations Training course at SmartNet Academy to master the technologies that power AI systems. Receive a Certificate of Completion to validate your expertise in AI infrastructure and set yourself apart in the growing job market.
The AI Infrastructure Transformation Accelerates
The Cisco-NVIDIA partnership represents a watershed moment in enterprise AI infrastructure. By combining validated hardware architectures, integrated software platforms, and comprehensive training programs, the collaboration addresses the primary barriers that have slowed AI adoption.
The technical innovations—N9100 switches, Hyperfabric AI management, Secure AI Factory architecture—provide enterprises with proven pathways to deploy GPU clusters at scale. The reference architecture approach dramatically reduces deployment complexity and risk.
The certification programs create structured pathways for infrastructure professionals to acquire AI-specific skills. As organizations struggle to find qualified AI infrastructure talent, these programs will quickly become industry benchmarks for validating expertise.
For infrastructure professionals, these developments signal both opportunity and urgency. The AI infrastructure skills gap creates exceptional career opportunities for those who systematically build relevant expertise. The availability of certifications, training programs, and validated architectures lowers barriers to entry while providing clear progression paths.
The partnership’s market momentum—evidenced by Cisco’s $2 billion-plus AI-related orders and broad ecosystem participation—confirms enterprise commitment to AI infrastructure investment. Organizations are moving from AI experimentation to production deployment, creating sustained demand for infrastructure expertise.
The future of enterprise IT infrastructure is AI-native. Infrastructure professionals who embrace this transformation, acquire relevant skills, and gain hands-on experience will find themselves at the forefront of technology’s most significant shift. The tools for succeeding in this transition—certifications, training programs, and proven architectures—are now available. The opportunity lies in taking action.