Enterprises can buy data center infrastructure. What's hard is combining compute, data, and networking infrastructure into something that's actually validated to run in production, at the scale the deployment calls for: a single enterprise environment, a large-scale AI factory, or a sovereign AI initiative that can provide its own data residency and governance requirements.
That's the gap NetApp and Supermicro are closing together, built on NVIDIA Reference Architectures for AI Factories. NVIDIA provides full-stack guidance for production-ready AI infrastructure, with design principles spanning GPU-accelerated compute, high-performance networking, AI software, and efficient data pipelines for consistent behavior across AI training and inference. Combining Supermicro's AI-optimized compute and rack-scale systems with the NetApp Platform, the two companies help customers accelerate deployment, improve GPU utilization, strengthen data governance, and reduce operational complexity.
This collaboration goes beyond pairing compute with high-performance storage. It extends Supermicro's AI Factory architecture, itself built on NVIDIA Reference Architectures, with NetApp Platform's unified data management, cyber resilience, hybrid-cloud leadership, and disaggregated scale, layered onto Supermicro's rack-scale compute, networking, power, and liquid-cooling expertise, validated end to end rather than assembled layer by layer.
For our customers, the distinction matters. Because the compute and data infrastructure are tested and validated as a unified system, teams get an architecture that deploys quickly, evolves across generations of compute, and delivers a consistent data experience, from a single enterprise environment to the world's largest AI factories. That consistency is what shortens design and integration cycles, reduces deployment risk, and moves AI environments into production faster than assembling and validating each layer independently.
The first concrete expression of this collaboration is the NetApp Verified Architecture for Supermicro AI Factory. The design combines NetApp-verified data infrastructure with Supermicro AI Factory configurations based on NVIDIA Reference Architectures. The applicable Supermicro compute and NVIDIA Spectrum-X Ethernet networking configurations are NVIDIA Design Review Board–endorsed designs for Infrastructure Configuration and Spectrum-X across NVIDIA RTX PRO 6000 Blackwell Server Edition, HGX B200, and HGX B300 AI factory designs, giving customers prescriptive guidance for integrating compute, networking, and enterprise data infrastructure.
NetApp's contribution is extending the reference architecture with the data-management depth enterprises actually need in production: governance, cyber resilience, and hybrid-cloud consistency, layered on top of the compute foundation Supermicro provides.
Every AI deployment context has different constraints, but the same underlying requirements: GPUs that stay utilized instead of waiting on data, storage and data management that scale without becoming a bottleneck, and infrastructure that can be audited and governed with confidence. The NetApp-Supermicro collaboration is built to answer those requirements across the full range of where AI actually gets deployed today:
For technical teams evaluating an AI infrastructure build today, the NetApp Verified Architecture for Supermicro AI Factory is the place to start: the design guide lays out the validated compute, storage, and networking configuration in full. For the broader view of how this collaboration fits into Supermicro's AI Factory portfolio, see the AI Factory page below.