The Frontier supercomputer at Oak Ridge National Lab remains the world’s fastest supercomputer for the third consecutive year. Powered by AMD EPYC CPUs and AMD Instinct GPUs, the supercomputer achieved a staggering High-Performance Linpack (HPL) score of 1.2 exaflops, as confirmed by the latest Top500 list.
Three new systems from Lawrence Livermore National Laboratories (LLNL) — El Capitan Early Delivery System, RZAdams, and Tuolumne — secured spots at numbers 46, 47 and 48, respectively. Their single cabinet early submission scores reached an impressive 19.65 petaflops. They are the first to feature the AMD Instinct MI300A APU, which was launched in December 2023.
AMD now powers 156 supercomputers on the latest Top500 list, marking a 29 percent increase from the previous year. Additionally, it has 157 systems on the Green500 list, which ranks the most efficient supercomputers worldwide.
“The AMD Instinct MI300A APUs are setting the pace for innovation, delivering leadership performance and efficiency for critical workloads at the convergence of HPC and AI,” said Brad McCredie, Corporate Vice President and General Manager of Data Centre and Accelerated Processing at AMD.
El Capitan is on track to surpass the two exaflops mark for double precision performance. If successful, it will become the second supercomputer to break the exaflop barrier.
“RZAdams is a critical addition to LLNL’s existing and highly capable unclassified computing ecosystem. 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,” said Bronis R de Supinski, Chief Technology Officer of Livermore Computing, LLNL.
“RZAdams is supporting essential opportunities for code porting, optimisation 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,” he added.
Alveo V80 to revolutionise HPC
AMD has unveiled the Alveo V80 Compute Accelerator Card. Designed to tackle compute-intensive workloads with large data sets, Alveo V80 leverages the AMD Versal HBM Series adaptive SoC. With up to 2X increased memory bandwidth, 4X network bandwidth, and real-time acceleration capabilities, this hardware-adaptable accelerator promises to revolutionise HPC.
Supermicro will integrate the Alveo V80 accelerator with AMD EPYC CPU-powered servers, further enhancing the ecosystem of cutting-edge solutions.
“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,” said Michael McNerney, SVP of Marketing and Network Security at Supermicro.
Entelechy Asia 