|
Getting your Trinity Audio player ready...
|
The planet’s fastest supercomputer, El Capitan, located at the Lawrence Livermore National Laboratory, received a warm welcome from National Nuclear Security Administration officials and industry partners at a grand dedication ceremony this month.
The $600 million system is NNSA’s first exascale supercomputer and will be used to support its advanced simulation and computing program, which works to predict the behavior of nuclear weapons through comprehensive, science-based simulations, according to the LLNL website.
Operating at a peak performance of 2.79 exaFLOPs or 2.79 quintillion calculations per second, El Capitan has equal processing power to one million of today’s fastest smartphones simultaneously working on a calculation.
In effect, researchers from across the Department of Energy’s NNSA labs — LLNL, and Los Alamos and Sandia National Laboratories — will have unprecedented modeling and simulation abilities to ensure the safety, security and reliability of the nation’s nuclear deterrent without underground nuclear testing.
Since going online in 2024, the system is currently undergoing usability testing and is expected for full deployment as a classified supercomputer in March 2025.
Also spotlighted at the Jan. 9 ceremony was El Capitan’s unclassified, companion system Tuolumne, which is set for open science research like climate and earthquake modeling, advanced manufacturing, energy sustainability and computational biology.
“What we will do with this machine will far outserve anything we imagine today,” LLNL Director Kim Budil said of El Capitan.

The computing power El Capitan offers is unprecedented, lab officials repeated at the dedication event.
At peak performance El Capitan offers over a 22-fold increase in computing performance over Sierra, previously the fastest system at the lab, according to documents from the event. This means high-resolution simulations that would have taken weeks or months will only require hours or days on El Capitan.
Researchers will use the mighty system for complex modeling and simulations to better understand weapons performance, reliability and safety as materials age, components are upgraded and new designs are utilized, according to documents from the event. Increased fidelity of simulation will also improve researcher’s understanding of physical phenomena like fluid flow, radiation transport and material behavior under extreme conditions.
In addition to the sheer computational power of El Capitan, it will also support AI-based workflows for NNSA-relevant work like material discovery, design optimization, advanced manufacturing, digital twins and intelligent AI assistants trained on classified data, according to a press release from Jeremy Thomas, public information officer at LLNL.

“This story will continue to unfold in the coming months and years, as this system is applied to our national security mission,” said Rob Neely, LLNL’s associate director of weapon simulation and computing, strategic deterrence. “This is just a waypoint on a long journey.”
Already, El Capitan has been used to understand more about fusion ignition — achieved by the lab in 2022 — during the machine’s open science era.
To create such a powerful machine required a collaborative effort between LLNL, NNSA, Hewlett Packard Enterprise and AMD.
The speed of El Capitan can be attributed to its AMD Instinct MI300A accelerated processing units, which combine CPU cores, GPU cores and high bandwidth memory.
Also important to the supercomputer are contributions from HPE, including its Cray Supercomputing EX solution, direct liquid-cooling system and Slingshot interconnect.

Not only does the El Capitan offer unprecedented computing power, it also serves as a symbol of the nation’s capabilities, according to Neely.
“This is a signal to the rest of the world, both our adversaries and our allies, that we’re confident that we can continue to maintain our nuclear deterrent without returning to underground nuclear testing — that’s something that we don’t want to have to do,” he said. “And because of these tools and capabilities, we don’t believe that we will have to and we can continue to remain at the forefront of deploying an effective deterrent.”
Marvin Adams, deputy administrator for defense programs at NNSA emphasizes the importance of that stockpile to deterrence.
“I do believe that our nuclear deterrent is playing a key role in keeping us out of major wars while maintaining our way of life in this country and the countries our partners and allies are in,” Adams said.
In addition to El Capitan, LLNL celebrated the companion system Tuolumne.
Despite being about 1/10th the size of El Capitan, Tuolumne ranks as the 10th fastest supercomputer in the world thanks to its identical architecture and components to El Capitan.
“I think calling Tuolumne number ten doesn’t do it justice,” Budil said. “Our open science machine is bigger than our current, last generation national security machine. So this step is truly revolutionary for us in terms of capability.”
Researchers at LLNL, Los Alamos and Sandia as well as other agencies through Strategic Partnership Programs will be able to use the open science system.










Notable, AMD Instinct accelerated M1300A processing units used in this Quantum Computer
Quantum computers can simulate the behavior of materials at the atom level, helping researchers discover new materials with desired properties for applications in electronics, energy storage, and more.
Quantum computers can analyze molecular structures and interactions more efficiently, accelerating the process of discovering new drugs and treatments for diseases.
Quantum computers can process vast amounts of data and simulate complex climate systems, improving our understanding of climate change and helping to develop more accurate models for predicting further climate patterns.
Quantum computing has the potential to revolutionize cryptography by developing new encryption methods that are more secure against potential quantum attacks.
Quantum algorithms can solve optimization problems more efficiently, which have applications in logistics, finance, and supply chain management.
Quantum computers can play a crucial role in advancing fusion research by optimizing the design and operation of fusion devices.
Quantum computers’ ability to perform complex simulations much faster than classical computers can help scientists better understand and control the plasma conditions necessary for fusion.