Hewlett Packard Enterprise (HPE) has presented its “El Capitan” supercomputer to the Lawrence Livermore National Laboratory (LLNL) of the United States Department of Energy (DOE). Thus positioning itself as the only company to develop the three existing exascale systems globally. With a capacity of 1,742 exaflops, a performance of 58.89 gigaflops per watt and using fanless direct liquid cooling technology, El Capitan is also among the top 20 most energy efficient systems in the world.
“El Capitan represents a significant advance in exascale supercomputing, combining exceptional performance with power efficiency and innovative capabilities to accelerate AI-powered scientific discovery. In addition, it will enable important advances in national security and renewable energy,” says Trish Damkroger, senior vice president and general manager of HPC & AI Infrastructure Solutions at HPE.
As the world’s most powerful supercomputer, El Capitan will enable the United States to maintain its leadership in national security and strengthen the ability of the National Nuclear Security Administration’s (NNSA) Tri-Labs to meet growing challenges related to safety, security, and reliability of the country’s nuclear arsenal, supporting both current and future modernization efforts.
HPE presents “El Capitan”: the world’s fastest exascale supercomputer
The Captain will also address secondary missions related to national security, such as nuclear nonproliferation and counterterrorism. Additionally, it will be used in advanced research in materials discovery, nuclear data, and high-energy-density science, such as inertial confinement fusion research conducted at the National Ignition Facility. The advances made thanks to El Capitan will also have applications in unclassified projects, covering areas such as energy security, climate change, modernization of the electrical grid, discovery of new drugs and other strategic fields.
El Capitan reaches 1,742 exaflops, becoming the most powerful supercomputer in the world
The innovative platform underlying El Capitan’s performance is based on cutting-edge technology integrated into the category-leading HPE Cray EX supercomputing solution. This solution combines AMD Instinct MI300A APUs, which integrate CPU and GPU cores along with high-bandwidth memory in a single package, the high-speed HPE Slingshot interconnect, and a custom storage solution.
HPE Slingshot, an Ethernet-based fabric designed for speed and efficiency, acts as the backbone of the system, enabling massive computations on more than 11,000 nodes. As a result of public-private collaboration, LLNL and HPE have developed a custom near-node local storage solution designed to reduce latency. This dynamically configurable system organizes data into tiers within a Luster-based global file system, shared efficiently among all compute nodes.
