HPE Launches New Purpose-built Solutions – Powered by AMD – to Accelerate Training for Large, Complex AI Models

New HPE ProLiant Compute XD685 supports eight AMD Instinct™ MI325X accelerators and two AMD EPYC^TM CPUs to deliver optimum performance and flexibility to efficiently build and train large language models 

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

The race is on to unlock the promise of AI and its potential to dramatically advance outcomes in workforce productivity, healthcare, climate sciences and much more. To capture this potential, AI service providers, governments and large model builders require flexible, high-performance solutions that can be brought to market quickly. 

“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,” said 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,” said 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.”

Speed to solution and flexibility increase competitive edge for AI

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. Additionally, the HPE ProLiant Compute XD685 brings together HPE’s decades of direct liquid cooling expertise, along with security innovation with HPE iLO, creating a high-performance solution designed to be sustainable and secure.

• Run the largest AI models with fewer GPUs: Powered by AMD CDNA™ 3 architecture, AMD Instinct™ MI325X, accelerators offer leading AI performance and efficiency for training and inference tasks. The MI325X accelerators feature leadership HBM3E memory capacity with 6 TB/s of memory bandwidth optimizes performance and lowers TCO. 
• Improve efficiency with optional direct liquid cooling: The direct liquid cooling option helps organizations meet escalating power requirements, advance sustainability goals and lower operational costs—all by leveraging HPE’s decades of expertise and market leadership in deploying the world’s largest liquid-cooled IT environments.
• Accelerate deployment and operations: HPE Services offers the full range of customizable and flexible services to install and deploy large AI clusters anywhere in the world with stability and operational excellence. 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, industry-leading security innovation that starts in the supply chain and is rooted in the silicon. 
• Optimize rack density: The HPE-designed 5U server chassis uses a compact 8-nodes-per-rack arrangement to maximize rack density for 8-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. 
• Simplify manageability of large AI clusters: HPE Performance Cluster Manager is a fully integrated system management software that gets complex systems up-and-running quickly with automated setup from bare metal and keeps clusters healthy and operational with detailed telemetry, GPU stress tests, and more.

Availability

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

Sign up for the free insideAI News newsletter.

Join us on Twitter: https://twitter.com/InsideBigData1

Join us on LinkedIn: https://www.linkedin.com/company/insideainews/

Join us on Facebook: https://www.facebook.com/insideAINEWSNOW