HPE announced an expansion of the NVIDIA AI Computing by HPE portfolio, introducing new solutions for secure and scalable AI factories, new AI data centre interconnect to optimise AI workload performance across clusters operating over long distances or within multiple clouds, and the first AI factory lab in the European Union (EU) for customers worldwide to test and validate their sovereign AI factories.
“HPE and NVIDIA continue to provide the foundation for secure AI factories at any scale, with new innovations that deliver a greater range of performance for more diverse workloads than ever before,” said Antonio Neri, president and CEO of HPE. “Together, HPE and NVIDIA are showcasing our unique strengths to deliver true full-stack AI infrastructures that provide enterprises with a greater range of performance for more diverse workloads.”
“Every nation and enterprise needs to own the production of its intelligence,” said Jensen Huang, founder and CEO of NVIDIA. “We’re transforming the data centre into an AI factory — a manufacturing plant for the new industrial revolution — and by deploying the full stack of NVIDIA accelerated computing and Spectrum-X Ethernet networking with HPE, we’re creating the template for sovereign AI. The new AI Factory Lab provides a foundry where customers can turn data into value, securely and at scale.”
HPE addresses customers’ needs for more control and autonomy over their AI infrastructure and data
HPE and NVIDIA are launching a new AI Factory Lab in Grenoble, France where customers can test and refine workloads on a sovereign, air-cooled AI factory environment. The lab is equipped with the latest NVIDIA AI Enterprise government-ready software, HPE servers, HPE Juniper Networking PTX and MX Series routers, NVIDIA accelerated computing, NVIDIA Spectrum-X Ethernet networking, and HPE Alletra storage. This new lab environment will also allow customers to validate performance on infrastructure located and running in the EU, designed to support scaled AI for the region, helping address data sovereignty and regulatory compliance needs of global enterprises operating within the EU. HPE is also partnering with Carbon3.ai to launch a Private AI Lab in London designed to accelerate UK enterprise AI adoption and built on HPE Private Cloud AI, NVIDIA AI Enterprise software suite and powered by NVIDIA AI infrastructure.
Expanded HPE Private Cloud AI capabilities to address data and operational sovereignty
As European markets are increasingly prioritising data and operational sovereignty to more easily adopt AI with a secure, private infrastructure, HPE Private Cloud AI has introduced new configurations, use cases, and capabilities, including:
- NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs on HPE Private Cloud AI, along with NVIDIA Hopper, give customers a choice of GPUs and provide greater flexibility across multiple workloads
- The integration of STIG-hardened and FIPS-enabled NVIDIA AI Enterprise in air-gapped environments offers an added layer of security, enabling a solution capable of satisfying numerous compliance standards
- HPE Private Cloud now supports GPU fractionalisation to optimise utilisation and lower costs, using NVIDIA Multi-instance GPU (MIG) technology
- New Datacenter Ops Agents provided by World Wide Technology (WWT), NVIDIA, and HPE simplify AI data centre management and strengthen HPE’s unified operations across agentic AI and hybrid cloud environments
HPE’s Sovereign AI factory solutions are now available with new system designs that incorporate country-specific compliance requirements, making it easier for organisations to comply with local regulations. HPE sets a new standard for security by offering adaptable solutions that maintain and strengthen security while meeting diverse country regulations. The new engineering-validated reference architectures also include robust software security measures and systems designed to support compliance audits. Additionally, HPE’s cybersecurity practice offers comprehensive advisory and professional services from workshops to management, and aligns with regulated industry needs, including solutions with NVIDIA technologies for sovereign compliance.
HPE and NVIDIA accelerate AI factory performance for data and data centres
HPE’s AI factory networking solutions include the NVIDIA Spectrum-X Ethernet networking platform and NVIDIA BlueField-3 data processing units (DPUs), providing customers with accelerated networking across various production scenarios – data centre to data centre, data centre to cloud. HPE solutions for AI factories have also been extended to include HPE Juniper Networking. This extension leverages the HPE Juniper Networks MX and PTX high-speed routing platforms to enable high-scale, secure and low latency connections from users, devices, and agents to AI factories and connections between clusters deployed across longer distances or across multiple clouds.
As AI matures, businesses need smarter, more integrated and secure ways to deploy AI, refining how they store, manage, and unlock value from distributed data. To meet this demand, HPE today announced the HPE Alletra Storage MP X10000 Data Intelligence Nodes, a next-generation architecture that transforms the X10000 into an active data layer that delivers inline intelligence by bringing NVIDIA accelerated computing to where information lives and enriching data in real time for AI pipelines. This builds on HPE Data Fabric capabilities that unify access to data across edge, core, and cloud.
By using the NVIDIA AI Data Platform reference design and running NVIDIA AI Enterprise software directly within the data path, the X10000 operates as a dynamic engine that analyses data as it is ingested and automatically infers patterns to feed AI factories. This powerful solution processes, classifies, and optimises data inline with RDMA for S3-compatible storage support and NVIDIA-accelerated performance.
Additional new innovations include the NVIDIA GB200 NVL4 by HPE, which delivers a compact, power-efficient alternative to larger AI and HPC platforms by enabling enterprises to quickly and securely deploy high-performance AI inference for large language models (LLMs) and other generative AI applications. The NVIDIA GB200 NVL4 by HPE combines two NVIDIA Grace CPUs with four NVIDIA Blackwell GPUs, providing exceptional density of up to 136 GPUs per rack and offering a cost-effective option for organisations needing high performance computing without massive power demands.
CrowdStrike, NVIDIA and HPE collaborate for unified AI security
HPE has selected CrowdStrike as an industry-leading security platform for HPE Private Cloud AI customers, unifying endpoint, identity, cloud, and data protection across hybrid and multicloud environments. The collaboration expands on Unleash AI partner CrowdStrike and HPE’s work to secure end-to-end AI innovation – including large language models (LLMs) accelerated by NVIDIA – and builds on CrowdStrike’s partnership with NVIDIA to deliver always-on AI agents for cybersecurity, advancing enterprises’ ability to safely operationalise AI at scale.
HPE collaborates with Fortanix to deliver NVIDIA Confidential Computing
HPE is working with Fortanix to leverage NVIDIA Confidential Computing and Fortanix Armet AI, a turnkey platform for running secure and sovereign agentic AI in AI factories and highly regulated environments. Technology from Fortanix, a new Unleash AI partner, can be deployed on HPE Private Cloud AI and HPE ProLiant Compute DL380a Gen12 servers equipped with NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs to help enterprises in EMEA and around the world build and run secure, scalable AI workloads on-premises, in the cloud, or in AI factories.