In this video from SC18, Derek Bouius from AMD describes how the company’s new EPYC processors and Radeon GPUs can speed HPC and Ai applications.
It’s been a fantastic year in the supercomputing space as we further expanded the ecosystem for AMD EPYC processors while securing multiple wins that leverage the benefits AMD EPYC processors have on HPC workloads,’ said Mark Papermaster, senior vice president and chief technology officer, AMD. ‘As the HPC industry approaches exascale systems, we’re at the beginning of a new era of heterogeneous compute that requires a combination of CPU, GPU and software that only AMD can deliver. We’re excited to have fantastic customers leading the charge with our Radeon Instinct accelerators, AMD EPYC processors and the ROCm open software platform.
LLNL Selects EPYC Processors and Radeon Instinct Power a New System
Called Corona, the newest high-performance computing (HPC) system for Lawrence Livermore National Laboratory’s High Performance Computing Innovation Center will use both AMD EPYC processors and AMD Radeon Instinct compute GPUs.
With 170 nodes incorporating more than 300 AMD EPYC 7401 processors and 300 AMD Radeon Instinct MI25 GPUs, Corona is a 383 teraFLOPS cluster that will be used for machine learning and data analysis techniques to solve challenging problems in HPC and big data. It will be delivered in late November 2018 and is expected to be available for limited use by December 2018.
EPYC Processor Product Extension
Expanding on the growing adoption of AMD EPYC processors, AMD is announcing the new high-frequency AMD EPYC 7371 processor. Made for workloads that benefit from higher frequency, like electronic design automation, high-frequency trading and HPC, the AMD EPYC 7371 provides 16 cores and 32 threads at a 3.1 GHz base frequency, with a 3.6GHz all core boost frequency and a 3.8GHz max boost frequency for eight cores. It will be available for partners and customers in Q1 2019.
EPYC Processors – The Building Blocks of Supercomputing
AMD EPYC processors are available from a growing ecosystem of more than 50 OEMs, ODMs, and system integrators. Whether customers are targeting machine learning, computational fluid dynamics, simulation and crash analysis in aviation and automotive manufacturing, oil exploration or more, AMD EPYC processors support the memory bandwidth, core density and PCIe lane expandability needed for HPC workloads.
New customer wins include:
- The US Department of Energy’s NERSC– using a Cray Shasta system powered by future AMD EPYC processors
- Cray and HAAS F1 Racing– using an AMD EPYC processor powered Cray CS500 cluster for CFD simulations
- The High-Performance Computing Center of the University of Stuttgart (HLRS) – using the next generation AMD EPYC processor, codenamed Rome, to power what is anticipated to be the largest supercomputer in Europe designed to address the specific needs of applications in industrial usage.
AMD is also powering innovative cloud delivery models with Microsoft Azure, launching a preview of its new HB instance for HPC this week. Finally, the University of Notre Dame Center for Research Computing, Oregon State University and the National Institute for Nuclear Physics in Italy continue to benefit from the value of their AMD EPYC-based systems.
Accelerating Deep Learning, HPC and Cloud Computing
At AMD’s recent Next Horizon event, the company demonstrated how the AMD EPYC ‘Rome’ processor coming in 2019 and the new AMD Radeon Instinct MI60 and MI50 accelerators, the world’s first 7nm datacenter GPUs, are designed to deliver the compute performance required for next-generation deep learning, HPC, cloud computing and rendering applications.
AMD also highlighted the new version of the AMD ROCm open software platform for accelerated computing. Designed for scale, ROCm 2.0 allows customers to deploy high-performance, energy-efficient heterogeneous computing systems in an open environment.
