OpenNebula Systems has been validated by NVIDIA to deliver integrated, AI-ready cloud infrastructure built on NVIDIA Spectrum-X Ethernet.
OpenNebula provides a cloud management and virtualisation platform designed to support secure, multi-tenant AI environments, including AI Factories. The platform offers isolation, governance controls, and lifecycle management for accelerated infrastructure.
Spectrum-X Ethernet is designed for AI-scale networking, addressing performance constraints that can arise in traditional data centre networks. Large-scale AI training and multi-tenant inference workloads are sensitive to latency, congestion, and jitter, requiring optimised networking performance.
Through native integration with Spectrum-X Ethernet, OpenNebula delivers a software-defined cloud platform intended for AI workloads. The platform supports multi-tenancy across compute, GPU, and network layers on a shared Spectrum-X Ethernet fabric, with direct GPU and SuperNIC passthrough to support predictable performance and workload isolation.
The integration is automated, with OpenNebula orchestrating tenant provisioning, network configuration, and device attachment across the Spectrum-X environment to reduce operational complexity.
The OpenNebula control plane is also operational on NVIDIA Air, providing an environment for testing, integration, and validation. This enables customers to evaluate the integration with Spectrum-X Ethernet using simulation tools, automation workflows, and configuration rendering, including for AI Factory and AI Gigafactory proof-of-concept deployments.
The validation supports the delivery of AI-ready cloud infrastructure for enterprise and service provider environments, including large-scale AI Factory and AI Gigafactory deployments in Europe. It also enables segmented, multi-tenant operation on shared AI infrastructure platforms.
