SANTA CLARA, Calif., – AMD (NASDAQ: AMD) showcased its ongoing high performance computing (HPC) capabilities at ISC High Performance 2024. For the third year in a row, the Frontier supercomputer housed at Oak Ridge National Lab – powered by AMD EPYC CPUs and AMD Instinct GPUs – remains the fastest supercomputer in the world with a High-Performance Linpack (HPL) score of 1.2 exaflops, based on the latest Top500 list.
The Top500 list also includes three new systems from Lawrence Livermore National Laboratories (LLNL), the El Capitan Early Delivery System (EDS), RZAdams, and Tuolumne, coming in at number 46, 47 and 48 respectively on the Top500 list with a single cabinet early submission score of 19.65 petaflops. These three systems are the first supercomputers on the Top500 list powered by the AMD Instinct MI300A APU, launched in December 2023.
AMD now powers 156 supercomputers on the latest Top500 list, a 29 percent increase from the previous year, and 157 systems on the Green500 list of the most efficient supercomputers in the world.
“The AMD Instinct MI300A APUs are setting the pace for innovation, delivering leadership performance and efficiency for critical workloads sitting at the convergence of HPC and AI,” said Brad McCredie, corporate vice president and general manager, Data Center and Accelerated Processing, AMD. “We are thrilled with the progress shown by LLNL today with their early submissions using Instinct MI300A. We look forward to our continued engagement with the teams at HPE and LLNL for further optimizations to drive increased performance as we help bring up El Capitan.”
Joining Frontier on the Top500 list, El Capitan EDS, RZAdams, and Tuolumne demonstrate that the El Capitan system is on track to exceed two exaflops of double precision performance, becoming the second supercomputer to surpass the exaflop barrier powered by AMD.
“RZAdams is a critical addition to LLNL’s existing and highly capable unclassified computing ecosystem,” said Bronis R. de Supinski, chief technology officer for Livermore Computing, LLNL. “Given the boost in performance we are already seeing from the AMD Instinct MI300A APUs, we expect this system to be a strategically significant resource and to move us forward across a wide range of scientific areas. RZAdams is supporting essential opportunities for code porting, optimization and software development for some of the applications that will eventually run on El Capitan. It is also allowing our scientists to push the boundaries of discovery with greater speed and efficiency than previous unclassified systems.”
New AMD Alveo V80 Accelerator for HPC and Compute-Intensive Workloads: To deliver optimized hardware-accelerated systems tailored for a wide spectrum of compute-intensive workloads, AMD launched the Alveo V80 Compute Accelerator Card. This new hardware adaptable accelerator is purpose-built to overcome performance bottlenecks for compute-intensive workloads with large data sets common to HPC, data analytics and network security applications.
Powered by the AMD Versal HBM Series adaptive SoC, the Alveo V80 accelerator card offers up to a 2X increase in memory bandwidth and logic density, and up to 4X the network bandwidth compared to the previous generationi It delivers massive hardware parallelism and 820GB/s memory bandwidth while hardware flexibility allows for real-time acceleration of diverse and custom workloads and data types for applications such as genomic sequencing, molecular dynamics, sensor processing, blockchain, and next-generation firewall and network security. Supermicro is integrating the Alveo V80 accelerator with AMD EPYC CPU-powered Supermicro servers.
“Supermicro’s strategic and long-standing collaboration with AMD enables us to address customer requirements for grappling with compute-intensive and memory-bound workloads at scale with superior time-to-delivery,” said Michael McNerney, SVP, Marketing and Network Security, Supermicro. “By leveraging the AMD Alveo V80 accelerator alongside our A+ server family, including the AS-4125GS-TNRT, customers can immediately take advantage of this compact 4U server solution for deployments where compute density and memory bandwidth are critical.”
Throughout the year, AMD EPYC processors and AMD Instinct accelerators have been leveraged to power many new supercomputers, including:
- Italian energy company Eni announced their new HPC6 supercomputer, powered by AMD EPYC CPUs and AMD Instinct GPUs. Once completed, HPC6 will be one of the world’s most powerful supercomputers dedicated to industrial applications.
- Lenovo and the University of Paderborn recently announced plans to build a new supercomputer powered by AMD EPYC CPUs. The installation is expected to start
by the second half of 2024.
- The High-Performance Computing Center of the University of Stuttgart (HLRS) has announced plans to build The Hunter system, which will be powered by the AMD Instinct MI300A accelerators.
- AGH University of Krakow unveiled a new supercomputer, powered by 4th Gen AMD EPYC CPUs, called Helios, which will be Poland’s fastest system to-date.
- AMD recently released ROCm 6.1 which consists of new features and optimizations to improve the stability and performance of applications powered by the latest AMD Instinct MI300 Series GPUs.