At HPE Discover in Las Vegas, HPE announced new solutions to supercharge the creation, adoption and management of AI factories across the AI lifecycle and for every type of organisation.
HPE is expanding its Nvidia AI Computing by HPE portfolio of AI factory solutions with Nvidia Blackwell GPUs, including new composable solutions optimised for service providers, model builders and sovereign entities, as well as the next generation of HPE Private Cloud AI, the turnkey AI factory for enterprises.
“Generative, agentic and physical AI have the potential to transform global productivity and create lasting societal change, but AI is only as good as the infrastructure and data behind it. Organisations need the data, intelligence and vision to capture the AI opportunity and this makes getting the right IT foundation essential,” says Antonio Neri, president and CEO at HPE.
“HPE and Nvidia are delivering the most comprehensive approach, joining industry-leading AI infrastructure and services to enable organisations to realise their ambitions and deliver sustainable business value.”
Jensen Huang, founder and CEO of Nvidia, adds: “We are entering a new industrial era — one defined by the ability to generate intelligence at scale. Together, HPE and Nvidia are delivering full-stack AI factory infrastructure to drive this transformation, empowering enterprises to harness their data and accelerate innovation with unprecedented speed and precision.”
The signature offering in the Nvidia AI Computing by HPE portfolio, Private Cloud AI is a fully integrated turnkey AI factory offering with Nvidia accelerated computing, networking and software. Private Cloud AI will provide:
- Support for Nvidia Blackwell accelerated computing with HPE ProLiant Compute Gen12 servers. These servers rank #1 in over 23 AI tests[1] and provide secure enclaves to prevent tampering, as well as post-quantum cryptography and trusted supply chain capabilities at rack and server level.
- Investment protection and seamless scalability from one GPU generation to the next, including Nvidia H200 NVL and now with Nvidia RTX PRO 6000 Server Edition GPUs, to support a wide range of enterprise AI workloads including agentic and physical AI use cases. A new federated architecture unifies resource pooling to make new GPUs and resources available to all AI workloads.
- Air-gapped management for organisations with strict data privacy requirements.
- Multi-tenancy that enables enterprises to collaborate and partition resources across teams.
- The latest Nvidia AI Blueprints including the Nvidia AI-Q Blueprint for AI agent creation and workflows.
- A new “try and buy” program that lets customers test Private Cloud AI across Equinix’s global footprint of high-performance data centers before making a purchase.
New AI factory solutions combine the latest AI innovations and expertise from HPE, including five decades of industry-leading liquid cooling and HPE Morpheus Enterprise Software for a unified control plane. Unlike most competing solutions, these validated, end-to-end composable solutions are modular, full tech stacks integrated ahead of time for the customer, speeding time to value.
- Using AI to Observe AI: HPE OpsRamp Software provides full-stack observability across Private Cloud AI and new AI factory solutions. HPE OpsRamp is also now a validated observability solution for the Nvidia Enterprise AI Factory.
- New AI factory at scale is designed for organisations operating at large scale such as service providers and model builders. HPE offers HPE ProLiant Compute XD, Nvidia AI Enterprise software and blueprints, air- and liquid-cooling technologies and a full range of advisory and professional services.
- New AI factory for sovereigns is for nations, governments and public sector organisations and offers additional specialised capabilities such as air-gapped management and solutions and services that enable data, technological and operational sovereignty.
By leveraging the Nvidia Enterprise AI Factory validated design, AI factory solutions from HPE can be deployed using the latest Nvidia accelerated computing, Nvidia Spectrum-X Ethernet networking, Nvidia BlueField-3 DPUs, and Nvidia AI Enterprise software to ensure high performance, robust security, efficient storage acceleration, and scalable infrastructure for the next era of AI.
New HPE Compute XD690 supports eight Nvidia Blackwell Ultra GPUs and reflects HPE’s continued commitment to be a time-to-market leader with Nvidia’s latest innovations in accelerated computing for AI. HPE Performance Cluster Manager delivers fully integrated systems management and advanced infrastructure monitoring and alerting across large, complex AI environments that scale thousands of nodes.
To ensure AI factories and applications are continuously fueled with AI-ready unstructured data, HPE Alletra Storage MP X10000 will support Model Context Protocol (MCP) servers. Integrating MCP with the X10000’s built-in data intelligence accelerates data pipelines and enables AI factories, applications, and agents to process and act on intelligent unstructured data. Additionally, the X10000 supports the Nvidia AI Data Platform reference design and offers an SDK to streamline unstructured data pipelines for ingestion, inferencing, training and continuous learning
Through new Unleash AI ecosystem partners, the latest Nvidia AI Blueprints and internal professional services, HPE delivers more than 75 AI use cases to customers. The Unleash AI partner ecosystem now features software solutions for agentic AI, sovereign AI, smart cities, industrial and manufacturing applications, data governance and privacy, responsible AI, video analytics, responsible AI, security and cybersecurity.
[1] “HPE sets new AI inference world records: Continued excellence in performance,” April 02, 2025.