OpenNebula
OpenNebula is a hybrid cloud platform enabling enterprise-grade AI workloads with multi-tenancy, NVIDIA integrations, LLM support, and 5G edge AI processing.
OpenNebula is a cloud management and orchestration platform designed with advanced features and integrations to build and operate enterprise-grade hybrid clouds that support multi-tenant AI Factories. It enables the on-demand, as-a-service execution of Large Language Models (LLMs) for training, and inference—meeting the demands of modern AI workloads. Models can be deployed within virtual machines or on virtualized Kubernetes instances, providing the flexibility required for a wide range of deployment scenarios.
OpenNebula incorporates NVIDIA best practices for deploying high-performance and cost-effective AI inference and training workloads, including the following features:
- Robust multi-tenancy
- Tight integration with hardware accelerators: DPUs, GPUs, CPUs, and InfiniBand
- Support for confidential computing
- High-performance Kubernetes instances
- Seamless integration with LLM libraries such as Hugging Face and vLLM
- Native integration with NVIDIA software like NVIDIA Dynamo
- Cloud interface to monetize AI workloads
- Built-in 5G support for ultra-low latency edge AI processing
- Hybrid cloud computing combining on-premises and cloud resources
Certification Overview
Solution Version: OpenNebula
Category: AI/ML
Certification Detail: 08-APR-2025 PowerEdge XE8640 with NVIDIA H100 GPU
Additional Information
- Company webpage: https://opennebula.io/
- Product description: https://opennebula.io/enterprise-ai/
Resources
- Learn more about Dell Technologies Open Telecom Ecosystem Lab
- Contact a Dell Technologies Expert
- View more resources
- Join the conversation with Dell @Dell
