Lawrence Livermore National Laboratory’s El Capitan supercomputer has achieved an unprecedented triple crown in high-performance computing, claiming the top spot on three major global benchmarks simultaneously.
The exascale machine, powered by more than 11 million cores and over 44,000 AMD processing units, delivered 1.742 exaFLOPS on the industry-standard benchmark while setting new records in AI-assisted computing and memory-intensive workloads.
The achievement marks a significant milestone for American supercomputing leadership. El Capitan maintained its position as the world’s fastest supercomputer on the prestigious TOP500 list while debuting at number one on two additional rankings that better reflect real-world scientific applications.
Beyond Raw Speed
What sets this achievement apart isn’t just the raw computational power—it’s the versatility. For the first time, El Capitan topped the High-Performance Conjugate Gradient (HPCG) benchmark with 17.41 petaFLOPS, a metric that measures performance on the complex, memory-intensive calculations typical in actual scientific research.
The system also debuted at number one on the HPL-MxP benchmark, delivering a staggering 16.7 exaFLOPS using mixed-precision AI techniques. This performance demonstrates the machine’s ability to handle the hybrid workloads that increasingly define modern scientific computing.
“This isn’t just a win for Livermore — it’s a win for national security, the NNSA enterprise and the future of AI-assisted scientific discovery,” said Bronis R. de Supinski, chief technology officer for Livermore Computing. “El Capitan is delivering exactly as designed: fast, flexible and optimized for the world’s most demanding workloads.”
The Exascale Elite
El Capitan joins an exclusive club of only three verified exascale supercomputers worldwide, all operated by the U.S. Department of Energy. The rankings reaffirm American dominance in high-performance computing, with Oak Ridge National Laboratory’s Frontier holding second place and Argonne National Laboratory’s Aurora taking third.
But what makes exascale computing so significant? These machines can perform more than a quintillion calculations per second—a capability that opens entirely new frontiers in scientific simulation, from modeling nuclear weapons to accelerating drug discovery.
El Capitan’s Technical Specifications:
- More than 11 million processing cores
- Over 44,000 AMD Instinct MI300A APUs
- 1.742 exaFLOPS peak performance
- 58.9 gigaFLOPS per watt energy efficiency
Energy Efficiency Matters
Beyond pure performance, El Capitan demonstrates that enormous computational power doesn’t have to come at an environmental cost. The system earned 26th place on the GREEN500 list of most energy-efficient supercomputers, achieving 58.9 gigaFLOPS per watt.
This efficiency becomes critical as supercomputers grow in size and capability. The ability to balance computational power with energy consumption will determine the sustainability of future exascale systems.
The Livermore Ecosystem
Perhaps most remarkably, Livermore now operates more TOP500-ranked systems than any other site globally. Fourteen different machines at the laboratory earned spots on the latest rankings, creating an unmatched ecosystem of computational resources.
The roster includes not just El Capitan but also Tuolumne, an unclassified companion system ranked 12th globally. Built with identical AMD processing units, Tuolumne serves as a crucial platform for open science applications, from AI-assisted fusion research to earthquake modeling and drug discovery.
The depth extends beyond the headline machines. Sierra holds 20th place, while systems like rzAdams, Lassen, and various computing clusters fill out the rankings, each serving specialized roles in the laboratory’s research mission.
National Security Computing
El Capitan’s primary mission involves supporting the National Nuclear Security Administration’s Stockpile Stewardship Program, ensuring America’s nuclear deterrent remains safe, secure, and reliable without underground testing. The system performs critical calculations and modeling tasks for NNSA’s three national laboratories: Livermore, Los Alamos, and Sandia.
This work requires not just speed but reliability and security—factors that influence supercomputer design far beyond benchmark performance. The system must handle classified workloads while maintaining the flexibility to support diverse scientific applications.
As scientific computing increasingly relies on AI-assisted methods and complex simulations, El Capitan’s triple-benchmark dominance suggests that future supercomputing leadership will depend on versatility as much as raw computational power.
If our reporting has informed or inspired you, please consider making a donation. Every contribution, no matter the size, empowers us to continue delivering accurate, engaging, and trustworthy science and medical news. Independent journalism requires time, effort, and resources—your support ensures we can keep uncovering the stories that matter most to you.
Join us in making knowledge accessible and impactful. Thank you for standing with us!