SANTA CLARA, Calif., Nov. 14, 2023 (GLOBE NEWSWIRE) -- Today, AMD (NASDAQ: AMD) showcased its continued leadership in high performance computing (HPC) at Supercomputing 23. AMD EPYC™ processors and AMD Instinct™ accelerators continue to be the solutions of choice for many of the most innovative, energy efficient and fastest supercomputers in the world. AMD now powers 140 supercomputers on the latest Top500 list representing a 39% year-over-year increase. Additionally, AMD powers eight of the top ten most energy efficient supercomputers in the world based on the latest Green500 list.
“AMD technology continues to be at the forefront of solving the world’s most important challenges. Our partnerships with national labs and research facilities around the world accelerate scientific discoveries, while driving new levels of understanding for critical items like health, energy, physics and more,” said Forrest Norrod, executive vice president and general manager, Data Center Solutions Group, AMD. “Working together with the industry, we are helping to establish and drive new paradigms of performance and energy efficiency, while empowering scientists and researchers of the world to advance the groundbreaking work that seeks to answer humanity’s toughest questions.”
The Exascale Era is AMD
The Frontier supercomputer at Oak Ridge National Laboratory, powered by AMD EPYC processors and Instinct accelerators, remains the fastest computer in the world - for the fourth list in a row. Providing an impressive blend of the best performance and leadership energy efficiency, Frontier continues to drive impactful science through its first full year of user operations. The latest scientific projects on Frontier include new research on power grid optimization, new designs on airplane engines for better efficiency and capabilities as well as two out of six finalists for the Gordon Bell award.
The second exascale class supercomputer to be powered by AMD, El Capitan, has started its installation process at Lawrence Livermore National Laboratory. The system will be powered by the forthcoming AMD Instinct MI300A APU, which is the first data center APU to combine CPU and GPU cores and high-bandwidth memory, all in a single package. This innovative design is anticipated to provide dramatic increases in both energy efficiency and performance. When it comes online, El Capitan is expected to exceed two exaflops of double precision performance, becoming the second AMD powered supercomputer to surpass the exaflop barrier. You can see more about El Capitan at this video from Lawrence Livermore National Laboratory.
“We are making excellent progress on the installation of El Capitan, which will clearly be one of the world's best computing systems,” said Bronis R. de Supinski, the Chief Technology Officer for Livermore Computing at Lawrence Livermore National Laboratory. “This enormous undertaking of multiple partners is creating a system that will enable our scientists to achieve results which they previously could only dream about. El Capitan will offer incredible programmability, performance and energy efficiency through its use of the AMD Instinct MI300A APUs virtually eliminating the challenges of repeatedly moving data between CPUs and GPUs, thus helping us achieve our mission.”
Enabling More Energy Efficient Computing
In HPC, energy efficiency is a top priority for the industry to meet sustainability commitments and deliver performance to enable the next generation of supercomputers. AMD is setting a new pace of innovation to accelerate energy efficiency through its 30x25 goal, which aims to deliver a 30x energy efficiency improvement in processors and accelerators for AI-training and HPC by 2025i, from a 2020 baseline, through a holistic approach to chip design and system-level improvementsii.
Showcasing the impact of the energy efficiency capabilities of AMD EPYC processors and AMD Instinct accelerators, AMD powers eight of the top 10 systems on the latest Green500 list, including the two most powerful supercomputers on the top ten; Frontier and the Adastra supercomputer.
Convergence of AI and HPC
AMD is also providing the hardware and software portfolio needed to meet the rapidly growing demand for applications of AI within the HPC industry.
AMD Instinct accelerators and AMD EPYC processors are currently powering the top two systems, Frontier and LUMI, in the latest HPL-MxP mixed-precision benchmark, which highlights the convergence of HPC and AI workloads. Overall, AMD-based systems were the clear winners at the top of the list – with Frontier achieving an HPL-MxP score of 9.95 exaflops and the LUMI system reaching 2.35 exaflops – highlighting continued growth and capabilities of AMD EPYC CPUs and AMD Instinct accelerators in AI workloads.
In software, AMD continues to push an open software ecosystem with the AI and HPC communities, working together to support new applications, frameworks, languages and more through open software. The AMD ROCm™ open software platform continues to advance and be utilized across multiple large scale HPC and AI systems. The ROCm software stack also has “day zero” support for PyTorch 2.0 providing developers with an extensive array of AI models powered by PyTorch that are compatible and ready to use “out of the box” on AMD accelerators.
AMD Instinct MI300A Solution Partners
HPE recently announced the HPE Cray Supercomputing EX255a accelerator blade, featuring the AMD Instinct MI300A APU. Ideal for AI and HPC workloads, the MI300A APU will be the world’s first APU accelerator for AI and HPC. See the AMD-powered HPE Cray Supercomputing EX255a at HPE’s booth #913.
Additionally, GENCI, the French HPC and AI agency, will engage its first extension of the Adastra supercomputer based on AMD Instinct MI300A accelerators. This future AMD-based partition will offer French researchers support in the convergence of HPC and AI applications.
Finally, Eviden has recently developed an AMD Instinct MI300A-powered blade for the BullSequana XH3000 full DLC SuperComputer line and will be delivering the first AMD Instinct MI300A-based SuperComputer in H1-24 at Max Planck Data Facility (MPCDF) in Germany. Visit the Eviden Booth #655 showcasing the BullSequana XH3000 and AMD Technologies.
Visit the AMD booth #1117 at Supercomputing 2023 to learn more about AMD solutions for HPC and speak with AMD experts.
Supporting Resources
- Learn more about AMD EPYC Processors
- Learn more about AMD Instinct Accelerators
- Follow AMD on Twitter
- Connect with AMD on LinkedIn
About AMD
For more than 50 years AMD has driven innovation in high-performance computing, graphics and visualization technologies. Billions of people, leading Fortune 500 businesses and cutting-edge scientific research institutions around the world rely on AMD technology daily to improve how they live, work and play. AMD employees are focused on building leadership high-performance and adaptive products that push the boundaries of what is possible. For more information about how AMD is enabling today and inspiring tomorrow, visit the AMD (NASDAQ: AMD) website, blog, LinkedIn and Twitter pages.
AMD, the AMD Arrow logo, EPYC, AMD Instinct, ROCm and combinations thereof are trademarks of Advanced Micro Devices, Inc. Other names are for informational purposes only and may be trademarks of their respective owners.
i 2 Includes AMD high-performance CPU and GPU accelerators used for AI training and high-performance computing in a 4-Accelerator, CPU-hosted configuration. Goal calculations are based on performance scores as measured by standard performance metrics (HPC: Linpack DGEMM kernel FLOPS with 4k matrix size. AI training: lower precision training-focused floating-point math GEMM kernels such as FP16 or BF16 FLOPS operating on 4k matrices) divided by the rated power consumption of a representative accelerated compute node, including the CPU host + memory and 4 GPU accelerators.
ii EPYC-030a: Calculation includes 1) base case kWhr use projections in 2025 conducted with Koomey Analytics based on available research and data that includes segment specific projected 2025 deployment volumes and data center power utilization effectiveness (PUE) including GPU HPC and machine learning (ML) installations, and 2) AMD CPU and GPU node power consumptions incorporating segment-specific utilization (active vs. idle) percentages and multiplied by PUE to determine actual total energy use for calculation of the performance per Watt. 13.5x is calculated using the following formula: (base case HPC node kWhr use projection in 2025 * AMD 2023 perf/Watt improvement using DGEMM and TEC +Base case ML node kWhr use projection in 2025 *AMD 2023 perf/Watt improvement using ML math and TEC) /(2020 perf/Watt * Base case projected kWhr usage in 2025). For more information: www.amd.com/en/corporate-responsibility/data-center-sustainability.
