Three of the world’s fastest exascale supercomputers have been developed and deployed by HPE, according to the latest TOP500 list, reflecting the company’s global leadership in providing high-performance computing (HPC) systems designed for the era of artificial intelligence.
HPE also maintains industry-leading energy efficiency thanks to its innovative direct liquid cooling technology, found in 10 of the world’s 20 most efficient supercomputers, according to the Green500 list.
The company is driving the future of AI and high-performance computing with a new portfolio and open source advancements set to have a profound impact across the industry.
To reinforce its leadership in supercomputing, HPE has introduced the second generation of exascale supercomputers and a new range of end-to-end solutions spanning architecture, storage, compute and acceleration blades, networking, software and direct liquid cooling.
Additionally, HPE reaffirms its commitment to open source technologies that accelerate innovation in HPC and AI, with the addition of the Chapel project to a new open source software foundation.
Strengthens its leadership in performance
For the eighth consecutive time, a system developed and implemented by HPE has been recognized as the world’s fastest supercomputer, as certified by the independent TOP500 list, which is updated twice a year.
In addition, it is the third time that HPE exascale supercomputers occupy the top three positions. This milestone highlights HPE’s unique expertise in designing, building and supporting large-scale infrastructure that accelerates scientific discovery, innovation and business transformation.
Systems developed by HPE that appear in the top 10 of the TOP500 list
#1 – El Capitan
Built for Lawrence Livermore National Laboratory (LLNL), El Capitan maintains its position as the world’s fastest supercomputer and has reached 1,809 exascaleflops, a 4% advance over its previous benchmarks. The system also continues to rank #1 in the High-Performance Conjugate Gradient (HPCG) benchmark, which measures performance in scientific applications, and in the HPL-MxP benchmark, which evaluates the convergence of HPC and AI in mixed-precision calculations.
#2 – Frontier
Built for Oak Ridge National Laboratory (ORNL) and equipped with AMD Instinct MI250 GPUs and AMD EPYC CPUs, Frontier debuted in 2022 as the first verified exascale system. With a performance of 1,353 exascale flops, Frontier continues to drive advances in quantum physics, renewable energy and precision engineering.
#3 – Aurora
Built in collaboration with Intel for Argonne National Laboratory (Argonne), Aurora delivers 1,021 exascale flops and is the second most powerful system in the HPL-MxP benchmark, where it achieves 11.6 exascale flops in mixed-precision calculations, critical for AI and data-driven science research.
#6 – HPC6
Developed for Eni, one of the largest technology companies in the energy sector, HPC6 is the world’s most powerful on-premise supercomputer for industrial use. Eni takes advantage of its 477.9 petaflops to, among other applications, optimize industrial processes, improve fluid dynamics studies for CO₂ storage and simulate the behavior of plasma in fusion research by magnetic confinement.
#8 – Alps
Created for the Swiss National Supercomputing Center (CSCS), Alps offers 434.9 petaflops to a broad community of researchers in Switzerland and around the world to study medicine, quantum chemistry, climatology and AI/machine learning.
#9 – LUMI
Developed for EuroHPC JU, LUMI is a pan-European supercomputer with 379.7 petaflops of performance that supports climate modeling, neural network training in European languages and a wide range of scientific disciplines across the continent.
In addition to the equipment mentioned above, over the last year, HPE has developed and delivered several new AI and HPC systems, supported by services that offer customers a turnkey implementation.
Industry-leading supercomputers and AI clusters are massively powerful systems designed to address some of the world’s most complex challenges. HPE plays a critical role in helping organizations and researchers achieve scientific breakthroughs, solve engineering problems, and transform the way we live. Some recent notable projects include the following ACM Gordon Bell Award finalists:
6 of the 10 fastest systems in the world and 10 of the 20 most energy-efficient supercomputers have been developed by HPE
Boosting innovation in HPC
The High Performance Software Foundation (HPSF) has announced that the Chapel project is joining the Foundation as its newest project. Initially created by Cray, Chapel evolved into an open source project focused on improving programming for parallel computers, making it easier to write clear and concise programs with a minimal amount of code.
Since then, the project has been promoted by HPE with the aim of bridging the gap between HPC users and desktop environment programmers. As a project of the Linux Foundation, HPSF’s community governance model will expand Chapel’s reach with the goal of attracting new users and contributors. HPE will continue to participate in Chapel as a collaborator and founding member.
