HPE continues its industry leadership building the fastest and most energy-efficient supercomputing systems in the world, enabling research institutions and large enterprise companies to handle larger workloads, accelerating discovery and innovation giving HPE the distinction of building the only three exascale systems in the world.
El Capitan, built for the National Nuclear Security Administration and Lawrence Livermore National Laboratory in conjunction with AMD, has reached 1.742 exaflops on the November 2024 edition of the TOP500, making it the world’s most powerful supercomputer and one of the top 20 most energy-efficient systems.
El Capitan joins the ranks of the world’s three verified exascale systems along with two other HPE-built supercomputers, Frontier at Oak Ridge National Laboratory, ranking No 2 and Aurora at Argonne National Laboratory ranking No 3.
“We are proud to have delivered not only the top 3 Supercomputers in the world, but also 3 of the top 4 supercomputers in Saudi Arabia. El Capitan showcases how monumental performance capabilities can accelerate AI-driven scientific discovery and make incredible breakthroughs. I expect to see further investments in supercomputing in the Middle East region,” said Mohammad Alrehaili, Middle East Managing Director at HPE.
HPE-built supercomputers are collectively No 1 in performance share, accounting for more than 5.75 exaflops among the TOP500 list of the world’s fastest systems. Among the top 10 supercomputers, seven run on HPE’s leadership-class HPE Cray Supercomputing EX systems based on the industry’s first 100% fan less direct liquid cooling system architecture.
El Capitan was designed to help the United States maintain its competitive edge in national security using high-fidelity modelling and simulation to solve challenging problems that are only within reach using exascale computing. It will also further the country’s investment in artificial intelligence, AI to advance scientific research in areas such as nuclear science, materials discovery and fusion energy. More information on El Capitan’s achievements can be found here.
Frontier, ranking as the second-fastest supercomputer at 1.353 exaflops, was the first system in the world to break the exascale barrier in 2022. Built in collaboration with ORNL and AMD, Frontier ushered in the modern era of AI-enabled supercomputers worldwide. Researchers using Frontier leverage AI to advance progress on a variety of subjects including cancer research, drug discovery, nuclear fusion, exotic materials science, superefficient engine design and supernovae.
At number three is Aurora, co-built by Intel and HPE, which leverages AI technologies to advance data-intensive research on aircraft design, cancer treatments and sustainable energy. Researchers in the Aurora Early Science Programme are making important strides in connectomics by 3D-mapping neuron connections in the brain and were nominated for a Gordon Bell Prize for their work developing a computing framework to accelerate the design of new protein sequences that can be used in vaccines or materials science.
HPE’s supercomputing leadership also extends into Europe with HPE Cray Supercomputing EX powering the region’s three most powerful AI-enabled systems.
HPE’s 100% fan less direct liquid-cooled solutions featured among most energy-efficient supercomputers on the planet
Supercomputers are designed for both precision and speed and this makes it ideal for scientists trying to create new medication or researchers advancing the world’s understanding of climate or space. However, the growing energy demands of large HPC, AI systems make energy efficiency more important than ever.
With 50 years of experience deploying reliable direct liquid cooling systems to customers, HPE supercomputers feature eight elements of cooling, which reduce energy consumption of the cooling infrastructure by up to 94% as compared to computer room air handler-based air-cooling.
Additional optimisations built throughout the HPE Cray Supercomputing EX solutions include the liquid-cooled HPE Slingshot interconnect, which is built with Dragonfly topology, enabling supercomputers to complete workloads faster and minimise overall energy requirements.
