Biomedical Research
Biological research teams are investigating next-gen sequencing and molecular modeling to design new pharmaceuticals and medical therapies.
Addressing the World’s Toughest Challenges
Supercomputers and high performance computing (HPC) systems are simulating the complex interactions behind weather, genomics, physics, and beyond. Exascale systems can execute a quintillion operations per second, enabling unprecedented time-to-insight. Explore how AMD solutions are powering these leading-edge systems and propelling innovation like never before.
Applications
Powering Top Supercomputers
AMD powers some of the world’s fastest and greenest supercomputers.
The El Capitan and Frontier exascale systems from the Department of Energy are leading us into a breakthrough era of efficiency in high-performance computing, artificial intelligence and a more sustainable, energy-efficient future.
El Capitan
AMD collaborated with the U.S. Department of Energy, Lawrence Livermore National Laboratory, and HPE to design El Capitan, which is currently the most powerful supercomputer in the world.1
El Capitan is built on AMD Instinct™ MI300A accelerated processing units (APUs), which offer exceptional compute density and efficiency, high-bandwidth memory, and advanced I/O sub systems. It is designed to excel at the most challenging HPC workloads and support AI and ML-based data analysis, enabling large models to run accurately.
Frontier
AMD collaborated with the U.S. Department of Energy, Oak Ridge National Laboratory, and HPE Cray to design the Frontier supercomputer. Featuring optimized 3rd Gen AMD EPYC™ processors and AMD Instinct MI250X accelerators, Frontier was the first supercomputer to officially cross the Exascale barrier with 1.1 exaflops of performance.
Frontier pushes the boundaries of scientific discovery by dramatically enhancing AI, analytics, and simulation performance—helping scientists pack in more calculations, identify new patterns in data, and get to insights faster.
LUMI
EuroHPC’s LUMI supercomputer is a powerful force for climate research, cancer research, and beyond. LUMI has granted researchers the unique ability to replicate Earth’s ecosystem down to the square kilometer, providing us with the knowledge to adapt to and potentially alter the course of climate change.
LUMI’s AI capabilities enable a neural network program that can detect cancer early and quickly simulate drug efficacy. This enables pathologists to diagnose cancer growth and simulate bespoke patient reactions to various treatments, so patients get the best personalized care as fast as possible.
Portfolio
Discover the AMD processors and accelerators that are supercharging scientific HPC workloads, as well as the software platforms that can be leveraged to optimize application performance.
AMD Instinct Accelerators
AMD Instinct accelerators are designed to supercharge HPC and AI, delivering leadership performance to handle the most demanding HPC applications and Exascale systems. Based on AMD CDNA™ architecture, these accelerators provide flexible, high-performance compute engines, high bandwidth memory, and scalable fabric and communications technologies.
AMD ROCm™ Open Software
AMD ROCm™ is open software for HPC application development targeting select AMD GPUs. It incorporates optimized drivers, compilers, libraries, support for OpenMP®, and the tools needed for large HPC systems—including the ability to code down to the metal, if desired.
AMD EPYC Processors
AMD EPYC processors are built to handle large scientific and engineering datasets, ideal for compute-intensive modelling and advanced analysis techniques. Certain workloads can also benefit from AMD EPYC processors with AMD 3D V-Cache™ technology. This technology can help accelerate simulations, boost engineering productivity, and speed product development.
AMD Zen Software Studio
Developers can optimize AMD EPYC server processor performance for HPC applications using the libraries and tools within AMD Zen Software Studio. The studio includes Optimizing C/C++ and Fortran Compilers (AOCC), uProf, Optimizing CPU Libraries (AOCL), ZenDNN, and Spack support.
AMD Alveo™ V80 Adaptive Accelerators
AMD Alveo™ adaptive accelerators combine hardware flexibility with low-latency acceleration for compute-intensive applications. The AMD Alveo V80 integrates FPGA fabric for dense compute, HBM2e for memory-bound algorithms, and 800Gb/s of network bandwidth for massive scale out over Ethernet.
AMD Vivado™ Design Suite
The AMD Vivado Design Suite™ provides a familiar design flow for FPGA developers for full customization via hardware programming languages or via IP building blocks available in the Vivado IP catalog. Developers can easily get started with a pre-validated base design that maps directly to Alveo hardware.
Find Solutions
Find a partner offering AMD hardware-based HPC systems.
Advancing AI
AMD is helping make the benefits of AI pervasive by delivering a broad solution portfolio, co-innovating across the open ecosystem, and driving compelling end-user experiences.
Stay Connected
Sign up for the newsletter to receive the latest data center updates or get in touch with an AMD Sales representative.
Footnotes
- 1. Source: Top500 supercomputer list November 2024
Footnotes
- 1. Source: Top500 supercomputer list November 2024