AMD announced multiple wins in the category of high-performance computing for AMD EPYC processors, including being the 7th fastest supercomputer in the world. Not only that, 4 of the 50 highest-performance systems on the biannual TOP500 list are powered by AMD. There’s no stopping AMD EPYC processors in the field of advanced science and health research as it continues to grow with new installations at Indiana University, Purdue University and CERN, as well as high-performance computing cloud instances from Amazon Web Services, Google and Oracle Cloud.
“The leading HPC institutions are increasingly leveraging the power of 2nd Gen AMD EPYC processors to enable cutting-edge research that addresses the world’s greatest challenges,” said Forrest Norrod, Senior Vice President and General Manager, Data Center and Embedded Systems Group, AMD. “Our AMD EPYC CPUs, Radeon Instinct accelerators and open software programming environment are helping to advance the industry towards exascale-class computing, and we are proud to strengthen the global HPC ecosystem through our support of the top supercomputing clusters and cloud computing environments.”
AMD EPYC processors are known for their improved performance, scalability, efficiency and low cost of ownership. It’s also known to power the upcoming world’s fastest exascale computers named Frontier and El Capitan.
To give recognition, here are 4 of the AMD EPYC-powered supercomputers that are now among the 50 highest-performance systems in the world, a total of 10 spots in the TOP500 list for AMD EPYC.
- Selene (#7) – An AMD EPYC 7742-based system in a DGX A100 SuperPOD platform from NVIDIA
- Belenos (#30) – One of the two BullSequana XH2000 supercomputers at Météo-France, the French national meteorological service, powered by 2nd Gen AMD EPYC processors
- Joliot-Curie (#34) – Moved up the list based on a new submission for the BullSequana XH2000 system using 2nd Gen AMD EPYC™ processors at GENCI, the French national high-performance computing organization
- Mahti (#48) – A 2nd Gen AMD EPYC-powered BullSequana XH2000 supercomputer at Finland’s Center for Science Information Technology
“Atos is proud to provide to its customers with cutting edge technology, integrating 2nd Gen AMD EPYC processors as soon as released, and demonstrating increased performance on HPC applications in production environments,” said Agnès Boudot, group senior vice president, Head of HPC and Quantum at Atos.
AMD-powered supercomputing systems are driving the research of the future. In a recent announcement, 2 universities are using research supercomputing systems powered by AMD EPYC processors in Dell EMC PowerEdge servers.
Indiana University will deploy Jetstream 2, an 8-petaflop distributed cloud computing system powered by upcoming 3rd Gen AMD EPYC processors. This will be used in a variety of fields such as AI research, social sciences and very timely, COVID-19 research. AMD EPYC is already being used in the campus with the Big Red 200.
“Jetstream 2 bundles computation, software and access to storage for individuals and teams of researchers across an array of areas of research,” said David Hancock, Director in Research Technologies, affiliated with the Pervasive Technology Institute at Indiana University. “With the next generation AMD EPYC processor, Jetstream 2 will provide 8 petaflops of cloud computing power, giving more access to high-end technologies to enable deep learning and artificial intelligence techniques.”
Purdue University will deploy Anvil, an AMD EPYC-powered supercomputer, which will provide advanced computing capabilities to support a wide range of computational and data-intensive research. AMD will also power Purdue University’s latest community cluster “Bell”, which is sceduled for deployment this fall.
In addition, CERN, the largest particle physics laboratory in the world, recently opted to use 2nd Gen AMD EPYC processors in Gigabyte servers to process the massive amounts of data from their latest Large Hadron Collider experiment to rapidly detect beauty quarks. In a case study, it detailed how combining the increased bandwidth of PCIe 4.0, DDR4 memory speed and the 64-core AMD EPYC 7742 processor allows researchers to collect the raw data streams generated by 40 terabytes of collision data occuring every second in the LHC.
As the high performance computing industry evolve to support new workload demands, especially at this time where working from home seems to be the new normal, cloud providers continue to adopt 2nd Gen AMD EPYC processors to provide unbridled performance and flexible solutions. With AMD’s recent wins that include Amazon Web Services, Google Cloud and Oracle Cloud, AMD is helping industry leaders push the boundaries in the new era of HPC and cloud computing.
Microsoft Azure continues its partnership with AMD with their recently announced HBv2-Series VMs for high performance computing workloads. The 2nd Gen AMD EPYC processors provide Azure customers with impressive core scaling, massive memory bandwidth and access to technologies like PCIe 4.0, a first in x86 server processors, to provide the best high performance computing experiences in the industry. Microsoft Azure aims to process real-world HPC workloads like CFD, explicit finite element analysis, seismic processing, reservoir modeling, rendering and weather simulation with AMD.
Community support continues to grow for AMD Radeon Open eCosystem (ROCm™), AMD’s open source foundation for heterogenous compute. Major development milestones in the latest update include:
- The HIP-Clang compiler is now up-streamed and reviewed by the LLVM™ community, providing a better open source experience for the developer,
- A new rocprofiler start/stop API to enable/disable GPU kernel HSA dispatch callbacks, increasing developer productivity and shortening the profiling run times,
- AMD RCCL compatibility with NVIDIA Communications Collective Library (NCCL) v2.6.4m,
- MIOpen provides an optional pre-compiled kernel package to reduce startup latency,
- A new CPU Affinity API is introduced for aiding applications to select the appropriate memory node for a given accelerator (GPU) and a given CPU,
- The new Radeon Performance Primitives library is a comprehensive high-performance computer vision library for AMD (CPU and GPU) with the HIP and OpenCL backend.
Join AMD CTO and Executive Vice President, Mark Papermaster, for a webinar on July 15th to discuss the full range of AMD solutions and upcoming innovations in HPC. Click the link for the time most convenient for you to register: 9 AM EDT, 12 PM EDT or 9 PM EDT.