HP has announced that it will be delivering the world's fastest exascale supercomputer to the US Department of Energy (Doe) Homeland Security (NDSS).
A new supercomputer that can reach a record speed of two exaflops has been named El Capitan at DOW Livermore National Laboratory (possess). The supercomputer is expected to be delivered in early 2023 and will be managed and organized by Possess, Sandia National Laboratory and Los Alamos National Laboratory for use.
El Capitan will enable advanced modeling and simulation in order to maintain the reliability and security of the US nuclear arsenal. HPE is an optimization of supercomputers to power complex and time-consuming exploratory 3D simulations for NSA missions that today's supercomputers cannot successfully manage. El Capitan will allow researchers to explore new applications through new, resource-intensive workloads including modeling, modeling, analysis, and artificial intelligence to support future NUNS missions.
- Cray's new supercomputer will protect us from nuclear stores
- Nvidia and ARM teams on supercomputers
- AMD and Cray are home to the world's fastest supercomputer
Senior Vice President and General Manager of High Performance Computing and Critical Solutions (MSC) at HPE, Peter Ungaro explains in press release how El Capitan will be used to unlock solutions to the most difficult questions by saying:
“As an industry, and as a nation, we have achieved a major milestone in the development of computing technology. The WPO is honored to support the US Department of Energy and Livermore National Laboratory in a critical strategic mission to advance the United States' security and defense posture. The computing power and capabilities of this system represent a new era of innovation that will unlock solutions to society's most complex problems and answer questions we never thought were possible."
Powered by AMD
HPE has chosen AMD to help power the DOW El Capitan supercomputer and the firm will use its experience from the high performance computing industry on the new system.
El Capitan will use the next generation of AMD EPYC processors, codenamed "Genoa", which feature the "Zen 4" Core processor. These processors will support next generation memory and I/O subsystems for AI and complex computing.
In terms of graphics performance, the next generation of Radeon graphics cards instinct GPUs based on a new compute-optimized architecture for workloads including high-performance computing and artificial intelligence will be used in the supercomputer. These chips will use next generations of high bandwidth memory and have been designed for optimal deep learning performance.
All components from AMD in El Capitan will be connected through the 3rd generation Infinity AMD architecture, which will provide high bandwidth, low latency communication between four Radeon Instinct graphics cards and an AMD EPYC processor included in each node of the supercomputer.
Senior Vice President and General Manager of AMD's Data Center and Embedded Solutions Business Group, Forrest Norrod VPO thanked for choosing its components to power El Capitan, saying:
“We are excited to continue our longstanding journey with VPO in developing innovative technologies for a variety of solutions, and now for the world's most powerful, exascale-class supercomputers. We look forward to continuing our collaboration with customers to bring together the next generation of AMD EPYC processors and Radeon Instinct GPUs with the HPE Cray system to power complex, resource-intensive, and high-performance AI computing workloads on El Capitan that today's systems cannot manage.” .
- We've also highlighted the best laptops for programming