The most powerful supercomputer, El Capitan, launched in the U.S. to aid weapons development

The Lawrence Livermore National Laboratory (LLNL) in California has officially put into operation El Capitan, the world’s most powerful supercomputer. Its primary purpose is to carry out classified tasks related to U.S. national security, including the development and analysis of nuclear weapons. The project cost $600 million, according to Space.

What Tasks Will El Capitan Perform?

El Capitan is designed to simulate processes that were previously tested through underground nuclear experiments, which have been banned in the U.S. since 1992. Additionally, it will focus on:

• Materials science and high-energy-density physics research
• Nuclear data analysis
• Advanced weapons development
• Other classified operations

El Capitan’s Computing Power

During the High-Performance Linpack (HPL) test, used to evaluate supercomputers, El Capitan achieved 1.742 exaflops, making it the fastest supercomputer in the world. Its peak performance could reach 2.746 exaflops.

For comparison:

• 1 exaflop equals one quintillion (10¹⁸) operations per second.
• Top-tier consumer laptops only achieve a few hundred gigaflops at most.

El Capitan is the third supercomputer in history to surpass the 1-exaflop threshold. Its closest competitor, Frontier, from the Oak Ridge National Laboratory in Tennessee, has a standard performance of 1.353 exaflops and a peak of 2.056 exaflops.

What Makes El Capitan So Powerful?

El Capitan is equipped with 44,544 AMD MI300A processors, housing over 11 million computing cores. The system is based on:

• AMD EPYC Genoa (central processors)
• AMD CDNA3 (graphics accelerators)
• 128 GB of high-bandwidth memory (HBM) per processor

El Capitan was commissioned under the U.S. Department of Energy’s CORAL-2 program as a successor to the Sierra supercomputer, which was deployed in 2018. Although Sierra remains operational, it currently ranks only 14th among the world’s top supercomputers.

The launch of El Capitan marks a new milestone in computational technology, which some experts believe could have a significant impact on nuclear security, materials science, and weapons development.