HPE Launches AMD-Powered Solutions

Hewlett Packard Enterprise announces the HPE ProLiant Compute XD685 for complex artificial intelligence (AI) model training tasks, powered by 5th Gen AMD EPYC processors and AMD Instinct MI325X accelerators. The new HPE system is optimized to deploy high-performing, secure, energy-efficient AI clusters for use in large language model training, natural language processing and multi-modal training.

“Training large language models, and doing so efficiently, requires strong-scaling, massive parallel computing capabilities, and unique services that only HPE’s high performance computing solutions deliver,” says Trish Damkroger, senior vice president and general manager, HPC & AI Infrastructure Solutions at HPE. “As a leader in delivering the world’s most powerful and energy-efficient systems, together with AMD, we are extending this innovation to the new HPE ProLiant Compute XD685 to support a growing market of AI model builders and accelerate scientific and engineering breakthroughs across industries.”

“Through the powerful combination of the latest AMD EPYC processors and Instinct accelerators, HPE ProLiant Compute XD685 customers can accelerate time to market and experience true leadership in AI performance and efficiency,” says Forrest Norrod, executive vice president and general manager, Data Center Solutions Group, AMD. “Our collaboration with HPE continues to deliver flexible, highly performant solutions that help maximize AI efforts for increased demand in large language model training and advances competitive industry innovation.”

The HPE ProLiant Compute XD685 delivers speed to solution for large AI model training and tuning projects using a new HPE-designed modular chassis. The modular, compact 5U chassis shortens time to market for solutions and provides the flexibility to accommodate a wide range of GPUs, CPUs, components, software and cooling methods.

Powered by AMD CDNA 3 architecture, AMD Instinct MI325X accelerators offer AI  efficiency for training and inference tasks. The MI325X accelerators feature HBM3E memory capacity with 6 TB/s of memory bandwidth.

The direct liquid cooling option helps organizations meet power requirements, advance sustainability goals and lower operational costs.

HPE Services offers a full range of customizable and flexible services to install and deploy large AI clusters anywhere in the world with stability. Expert services build, integrate, validate, fully test and customize solutions in the factory, facilitating quicker onsite deployment.

Protect against growing security threats: Securely manage HPE ProLiant Compute XD685 systems with HPE iLO, delivering seamless management and built-in security that starts in the supply chain and is rooted in the silicon.

The HPE-designed 5U server chassis uses a compact eight-nodes-per-rack arrangement to maximize rack density for eight-way GPU systems. The HPE ProLiant Compute XD685 supports eight AMD Instinct MI325X or AMD Instinct MI300X accelerators, two of the latest AMD EPYC 9005 Series processors and air cooling or direct liquid cooling.

HPE Performance Cluster Manager is a fully integrated system management software that gets complex systems running quickly with automated setup from bare metal and keeps clusters healthy and operational with detailed telemetry and GPU stress tests.

Availability

The HPE ProLiant Compute XD685 is available to order through HPE and will be generally available in the first quarter of 2025.

Sources: Press materials received from the company and additional information gleaned from the company’s website.