At the 2025 Open Compute Project Global Summit in San Jose, Hewlett Packard Enterprise announced updates to its server management platform, hardware design, and support for new AI networking technologies. The announcements include stronger encryption tools, modular server components, and expanded compatibility with NVIDIA’s latest infrastructure offerings.
One of the significant updates involves HPE’s Integrated Lights Out platform, iLO 7, which now includes new features designed to resist potential threats from quantum computing. Initially launched in February, iLO 7 uses a dedicated security processor and a secure storage area, known as a secure enclave, to protect HPE ProLiant Gen12 servers from unauthorized access.
The latest update supports longer encryption keys, up to 4096-bit RSA, meeting federal standards for post-quantum security. The platform has also been updated to function as a removable module, making it easier to replace hardware components without losing system settings. This change also helps verify that parts are original HPE components, thereby addressing risks associated with counterfeit hardware in the supply chain.
By 2026, HPE plans to introduce support for the Key Management Interoperability Protocol (KMIP) across both iLO 6 and iLO 7, allowing better integration with local and remote key management systems.
The company also highlighted its Gen12 ProLiant Compute server line, which follows modular hardware standards established by the Open Compute Project. These designs allow businesses to adjust their server configurations based on workload requirements. That flexibility can reduce costs and improve efficiency.
The product line includes several general-purpose servers (DL320, DL325, DL340, and DL345), as well as specialized models such as the DL380a for AI acceleration and the DL580 for large-scale applications.
HPE also announced that its ProLiant and Cray servers received 14 top rankings in the MLPerf Inference v5.1 benchmark tests. These results reflect improvements in how the systems handle machine learning workloads.
The company is expanding its support for NVIDIA technologies, including systems built for the new NVIDIA Kyber architecture and the upcoming Spectrum-XGS Ethernet platform. These technologies aim to improve data transfer speeds and reduce delays in AI data center operations. HPE said these updates will be available as soon as the technologies are released.
These announcements come during Cybersecurity Awareness Month, which has drawn more attention to how businesses are securing their IT infrastructure. HPE’s focus this year has been on making its systems more secure, more modular, and ready for new demands in AI and computing.
The company is demonstrating all of these servers and AI tools at booth A42 during the summit.
RAVEL and Strata Expanse have named RT-One as the first commercial customer of the AMPHIX AI Infrastructure Platform, a deployment…
At FMS 2026, Silicon Motion introduced the MonTitan SSD Reference Design Kit (RDK), a platform built around the company’s next-generation…
In late July, Liqid unveiled a scale-up AI platform built around AMD Instinct MI350P PCIe GPUs. The UltraStack 30 pairs…
DapuStor used FMS 2026 to show what it calls an industry-first 512TB SSD: a new top capacity for the R6060…
Volta has emerged from stealth with a vertically integrated AI infrastructure platform designed to combine capital formation, powered land, data…
IBM has announced a multi-year, $240 million agreement with Together AI to deploy a dedicated NVIDIA HGX B300-based AI inference…