The El Capitan supercomputer at Lawrence Livermore National Laboratory (LLNL), powered by AMD Instinct MI300A APUs and manufactured by Hewlett Packard Enterprise (HPE), has a High Performance Linpack (HPL) score of 1,742 exaflops, according to the latest Top500 list It is now the world’s fastest supercomputer. The El Capitan and Frontier system at Oak Ridge National Laboratory ranked 18th and 22nd, respectively, on the Green500 list, showcasing the impressive capabilities of AMD EPYC processors and AMD Instinct GPUs to deliver leading performance and energy efficiency for HPC workloads.
AMD Powers HPC and AI
AMD compute engines continue to power the most important supercomputers for businesses and national laboratories around the world and deliver outstanding performance in technical computing.
The latest AMD EPYC 9005 Series processors are the best server CPUs for enterprise, AI, and cloud, delivering up to 37 percent better next-gen IPC performance for HPC and AI workloads. These processors also deliver up to 3.9x faster time to insight compared to competitors for science and HPC applications that solve the world’s toughest problems.
AMD Instinct accelerators deliver leading performance for data centers of all sizes, from AI solutions to Exascale-class supercomputers. AMD Instinct MI300X and MI325X accelerators deliver leading AI performance and memory capabilities, while the AMD Instinct MI300A APU brings together CPU and GPU cores and stacked memory in a single package, enabling new levels of efficiency and performance for HPC and AI workloads.
Additionally, AMD EPYC Processors and AMD Instinct accelerators are being used to power many new supercomputer and AI projects and deployments, including:
- Italian energy company Eni recently announced its HPC 6 supercomputer powered by AMD EPYC CPUs and AMD Instinct GPUs. HPC 6 is one of the world’s most powerful supercomputers dedicated to industrial applications and is currently the fifth fastest supercomputer in the world.
- Paderborn University is in the process of receiving and installing a new supercomputer powered by the latest 5th Generation AMD EPYC CPUs.
- Sigma2 AS will replace two of the three national supercomputers in Norway with an HPE Cray Supercomputing EX system powered by 5th Gen AMD EPYC CPUs. When fully installed, this supercomputer is expected to be the fastest system in Norway.
- IBM and AMD announced a collaboration to distribute AMD Instinct MI300X accelerators as a service on the IBM Cloud. Expected to be available in the first half of 2025, this offering aims to improve performance and power efficiency for Gen AI models such as high-performance computing applications for enterprise customers. The collaboration will also enable support for AMD Instinct MI300X accelerators on IBM’s watsonx AI and data platform and support for Red Hat ® Enterprise Linux ® AI inference.
- AMD Instinct MI300A APUs will also power a next-generation supercomputer system for Japan’s National Institutes of Quantum Science and Technology (QST). Built by NEC Corporation, the system will use 280 AMD Instinct MI300A APUs to drive artificial intelligence and scientific research for the National Institutes of Quantum Science and Technology and the National Institute of Fusion Science.
“We Lead the Exascale Era”
As the only company running multiple exascale supercomputers, AMD continues to lead in performance and energy efficiency in HPC deployments worldwide.
The world’s most powerful supercomputer and the first exascale-class machine for the National Nuclear Security Administration (NNSA), El Capitan stands as the premier computing resource for NNSA Tri-Labs—LLNL, Los Alamos, and Sandia National Laboratories. It will be used to advance scientific discovery and national security, and will provide the massive computing power needed to ensure the safety, security and reliability of the nation’s nuclear deterrent without testing. This state-of-the-art system marks a giant leap forward in HPC, enabling unprecedented modeling and simulation capabilities required for NNSA’s Stockpile Management Program certifying aging nuclear stockpiles and other critical nuclear security missions such as nonproliferation and counterterrorism.
LLNL and other NNSA Tri-Labs are also using El Capitan and its companion system, Tuolumne, to drive AI and machine learning-powered data analysis, furthering LLNL’s AI-focused goals of building scientific models that are fast, accurate, and able to quantify the uncertainty in their predictions. It carries. El Capitan will apply AI to high energy density problems such as inertial confinement fusion research, while Tuolumne will be used for unclassified open science applications such as climate modelling, biosecurity/drug discovery and earthquake modelling.
Beyond El Capitan, AMD and HPE are also powering Frontier, the first exascale supercomputer. Located at Oak Ridge National Laboratory and powered by AMD EPYC CPUs and AMD Instinct GPUs, Frontier is the world’s second fastest computer with 1.35 exaflops of performance. Frontier continues to enable researchers to address complex scientific problems, from climate modeling to biomedical research to training large language models, further proving its significant contribution to advancing scientific discovery and supporting breakthroughs in artificial intelligence.
These systems provide tremendous computational power that contributes significantly to a wide range of research, including materials science, climate modeling and AI model development. By empowering researchers in diverse fields and supporting the development of AI models, El Capitan and Frontier are shaping the future of science and technology and providing solutions to address critical challenges around the world and demonstrate AMD’s commitment to providing the high-performance computing resources necessary to support the next generation of scientific discovery and innovation. emphasizes.
To learn more about AMD’s HPC solutions and meet with AMD experts, visit AMD booth #2731 at Supercomputing 2024.
Source: www.technopat.net
