In this article
Today, at AMD Advancing AI 2026 in San Francisco, HPE showcased its latest engineering innovation in supercomputing with the HPE Cray Supercomputing GX5000 featuring 6th Gen AMD EPYC processors.
The GX5000 builds on the success of previous HPE supercomputing architecture that broke the exascale speed barrier, and dominates the world’s top 10 list of fastest supercomputers, to support the most complex scientific and engineering problems involving AI and quantum.
The GX5000 delivers the industry’s highest density performance with 40% more AMD EPYC 9996 cores per rack than the market’s next densest offering.1 The GX5000 also maximizes capacity for growing HPC and AI workloads and improves operational efficiency by reducing carbon footprint with 25% less floor space compared to the previous generation2 and an increase of warm facility water by 25%, with up to 40C heating capacity, through next-generation direct-liquid cooling capabilities from HPE.
HPE Cray Supercomputing GX5000: built for a new era of AI-driven scienceHPE Cray Supercomputing GX5000 will power next-generation exascale and leadership-class supercomputers that HPE builds as fully integrated systems spanning state-of-the-art CPUs, accelerators, storage, networking, software and 100% direct-liquid cooling.
The GX5000 was designed to support converged HPC workloads, like modeling and simulation, with AI, to accelerate scientific discovery across areas such as drug discovery, climate forecasting and materials science.
GX5000 systems also address growing requirements for secure multi-tenant capabilities to manage user control while enabling access to bolster collaboration.
World-leading supercomputing and research centers are adopting GX5000 with 6th Gen AMD EPYC processors to make breakthroughs in AI and quantum-driven applications scientific discoveries.
Early adopters of HPE Cray Supercomputing GX5000 take on new missions in AI and quantumWorld-leading supercomputing and research centers are adopting GX5000 with 6th Gen AMD EPYC processors to make breakthroughs in AI and quantum-driven applications scientific discoveries.
Early adopters include Oak Ridge National Laboratory (ORNL) and the High-Performance Computing Center of the University of Stuttgart (HLRS).
Oak Ridge National Laboratory expands exascale capabilities for open science with “Discovery”After ushering in the world’s first exascale supercomputer, Frontier, Oak Ridge National Laboratory (ORNL), HPE, and AMD are partnering again to deliver an even more powerful exascale machine with Discovery, an HPE Cray Supercomputing GX5000 system with new AMD EPYC processors.
Discovery will support a mix of converged simulation and AI workloads, AI model development and training, and serve as a testbed for ORNL’s ongoing efforts in quantum computing.
ORNL, HPE and AMD are also building a dedicated AI system, called “Lux,” using next generation HPE systems featuring 6th Gen AMD EPYC processors. Lux will accelerate AI development, as part of the U.S. Genesis Mission, to strengthen national security and competitiveness.
HLRS boosts Germany’s research and manufacturing efforts with next-generation supercomputer “Herder”HPE and HLRS, one of Germany’s leading research centers, are continuing a years-long partnership of designing and delivering cutting-edge supercomputers with HLRS’ upcoming flagship system, “Herder.”
Herder, a GX5000 system with new AMD EPYC CPUs, is expected to be 7X more powerful than HLRS’ current supercomputer, to bolster German and European research and advance industrial innovation.
The system will support a growing demand in computational science for converged simulation and AI capabilities, and advance AI model training.
HPE Cray Supercomputing GX5000 supports a range of mixed-precision and double-precision computing workloadsThe HPE Cray Supercomputing GX5000, featuring 6th Gen AMD EPYC processors, is offered as two blade options to support demands in traditional supercomputing workloads, like modeling and simulation, and AI capabilities like model development, training, and tuning.
HPE and AMD collaborate on AMD Helios™ AI rack scale system, featuring industry-first scale-up Ethernet networkingHPE is collaborating with AMD to deliver the AMD Helios AI rack scale system, which represents a new generation of AI infrastructure designed for the era of trillion-parameter models and large-scale AI environments.
The solution delivers 260 TB/s of scale-up bandwidth and 2.9 exaflops of FP4 performance for trillion-parameter AI training and high-volume inferencing. The purpose-built HPE Juniper Networking hardware and software is based on the open Ultra Accelerator Link over Ethernet (UALoE) standard, the first in the industry to supercharge AI training and inference performance using standards-based Ethernet.
By combining HPE's leadership in high-performance computing, networking, advanced direct liquid cooled technologies, system engineering, and AI deployment, with AMD state-of-the-art AI computing platform , HPE and AMD will help transform Helios from an architecture into a production-ready platform that delivers performance at scale, reliability, and faster time to insight.
To learn more about HPE Cray Supercomputing solutions and organizations adopting them to power breakthrough science, please visit: https://www.hpe.com/us/en/cray-exascale-supercomputing.html
HPE ProLiant – next breakthrough for enterprise computeHPE is proud to collaborate with AMD to deliver a portfolio of HPE ProLiant servers featuring the latest AMD EPYC™ 9006 Series Server CPUs, optimized for the most demanding enterprise and AI workloads. Our collaboration combines AMD processor innovation with the trusted and differentiated HPE ProLiant experience customers expect – not just for performance, but for industry-leading security and AI automation.
HPE ProLiant delivers security designed from the silicon-up, intelligent lifecycle management, AI-driven opportunities across a hybrid environment, energy-conscious infrastructure, and future ready I/O. These capabilities are designed to help customers simplify increasingly complex environments while preparing for the next generation of enterprise AI.
To learn more about HPE Compute solutions, please visit https://www.hpe.com/us/en/compute/hpe-proliant-compute.html
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。