OpenNebula Systems today announced a collaboration with NVIDIA to validate OpenNebula as an orchestration platform integrated with NVIDIA Spectrum-X Ethernet Networking. This validation enables enterprises and service providers to deploy AI-ready, multi-tenant cloud infrastructures that meet the scale, performance, and operational requirements of modern AI workloads. Support and validation for NVIDIA Spectrum-X Ethernet, as well as GB200- and GB300-based systems, are provided in OpenNebula 7.2.

By combining OpenNebula’s open cloud and edge orchestration capabilities with NVIDIA Spectrum-X Ethernet, organizations can build and operate high-performance AI infrastructures with consistent multi-tenancy, advanced scheduling, and integrated visibility, while maintaining flexibility and avoiding vendor lock-in.

Why This Matters for AI Infrastructure

As AI environments scale, infrastructure complexity grows rapidly. Organizations need to support multiple teams and workloads on shared accelerated platforms—without compromising performance, security, or governance.

By integrating natively with NVIDIA Spectrum-X Ethernet, OpenNebula enables:

• Predictable performance for AI training and inference workloads
• Secure multi-tenancy across compute, GPU, and network layers
• Simplified operations through automated provisioning and lifecycle management
• Lower deployment risk, with validated architectures ready for production.

This combination makes it easier for organizations to move from pilot projects to large-scale AI production, while maintaining control, operational consistency, and strong governance.

Hands-On Validation with NVIDIA Air

To support evaluation and large-scale testing, the OpenNebula control plane is fully operational on NVIDIA Air. Customers and partners can explore OpenNebula hands-on, validate AI Factory architectures, and simulate deployments before production. This reduces time to value and accelerates decision-making for complex AI infrastructure projects.

Built for AI Factories Across HPC, Neocloud, and Telco Environments

This validation is a significant step toward secure, high-performance AI-ready cloud infrastructure, particularly for enterprises and service providers in Europe. OpenNebula’s multi-tenant architecture and strong isolation capabilities allow multiple teams or customers to share AI platforms efficiently and safely—ideal for AI Factories, AI Gigafactories, and other large-scale accelerated deployments.

Meet Us at MWC, EuroHPC and GTC

We’d love to show you how OpenNebula and NVIDIA Spectrum-X Ethernet can transform your AI infrastructure. Join us for demos and follow-ups at:

• Mobile World Congress (MWC) – Discuss AI Factories, multi-tenant cloud infrastructures, and sovereign cloud opportunities for telecom operators
• EuroHPC Summit – Discuss AI Factories, HPC infrastructures, and sovereign cloud platforms for research and industry
• NVIDIA GTC San Jose – Explore OpenNebula’s integration with NVIDIA Spectrum-X Ethernet, GB200-based systems, and NVIDIA Air.

Reach out to schedule a meeting and discuss in person the deployment options tailored to your environment.

Read the full press release here | Learn more on the dedicated landing page