HPE launches Nvidia-powered autonomous infrastructure platform and self-driving networks for enterprise AI operations.
At HPE Discover Las Vegas 2026, Hewlett Packard Enterprise unveiled comprehensive innovations spanning agentic artificial intelligence, autonomous networking, hybrid cloud management, and sovereign AI infrastructure, positioning its hybrid cloud ecosystem as a foundation for enterprise AI deployment, next-generation networking, and large-scale data center modernization.
HPE significantly expanded HPE GreenLake Intelligence with an agentic AI framework designed to simplify hybrid cloud and AI operations. The platform introduces the HPE OpsRamp Operations Copilot, delivering advanced observability for large language models (LLMs), real-time infrastructure insights, and comprehensive monitoring of hybrid cloud consumption and spending. HPE also strengthened its virtualization strategy through HPE Morpheus Software, positioning it as an alternative to legacy virtualization platforms, with new capabilities including the HPE Morpheus Orchestration Copilot and HPE Morpheus Central for centralized management of heterogeneous IT environments. On the security front, HPE expanded its private cloud portfolio with the HPE Private Cloud PC3000 and PC7000 systems, supporting air-gapped deployments for sovereign environments and progressing toward Department of Defense Impact Level 4 (IL4) certification readiness.
In a significant infrastructure partnership, Vultr selected HPE and NVIDIA to power its next-generation AI-focused data centers, centered on the NVIDIA GB300 NVL72 by HPE platform—a liquid-cooled architecture engineered for large-scale AI training, inference, and private cloud workloads. The infrastructure leverages NVIDIA Spectrum-X networking, incorporating both 400GbE and 800GbE connectivity, with advanced switches, optical transceivers, and specialized SuperNICs designed to optimize rack-scale GPU communication while minimizing data bottlenecks.
HPE expanded its autonomous networking vision through the HPE Aruba Networking portfolio, aiming to reduce network downtime caused by human error by extending self-driving capabilities across edge, campus, data center, and AI factory environments. The HPE Aruba Networking Central platform serves as a unified management layer coordinating security policies, device configurations, observability, and operational controls, while integrated AI Network Agents continuously monitor network performance, diagnose issues, and automatically resolve anomalies. HPE also expanded its NVIDIA AI Computing by HPE portfolio to help enterprises move agentic AI into production, with HPE Private Cloud AI now including pre-configured Agentic AI workloads providing enterprise search capabilities and multi-agent customer service automation built on NVIDIA AI Enterprise and NVIDIA NIM microservices. Two new hardware platforms support these deployments: the liquid-cooled HPE ProLiant Compute XD685, optimized for multi-node AI clusters, and the HPE ProLiant Compute XD670, designed for dense single-node deployments. HPE also announced validated architectures for fully air-gapped sovereign AI clouds, enabling governments and highly regulated industries to run advanced AI workloads completely isolated from the public internet.
Demonstrating real-world enterprise adoption, Siemens Energy has selected HPE Private Cloud AI as the foundation for its global AI modernization initiative. The energy company is deploying the platform to accelerate engineering simulations, product development, workflow automation, and grid optimization activities as global electricity demand continues to increase, utilizing HPE Private Cloud AI and HPE Ezmeral Data Fabric to create a unified data architecture connecting engineering specifications, operational datasets, and enterprise knowledge sources to private AI models.