Hewlett Packard Enterprise El Capitan, is an upcoming exascale supercomputer, hosted at the Lawrence Livermore National Laboratory in Livermore, United States and projected to become operational in 2024. It is based on the Cray EX Shasta architecture. When deployed, El Capitan is projected to displace Frontier as the world's fastest supercomputer.
| Active |
|
|---|---|
| Sponsors | U.S. Department of Energy |
| Operators | Lawrence Livermore National Laboratory and U.S. Department of Energy |
| Location | Livermore Computing Complex |
| Architecture | HPE Cray Shasta |
| Power | 40 MW (Proj) |
| Space | TBA |
| Memory | TBA |
| Storage | TBA |
| Speed | 2 exaFLOPS (Rmax) (Proj) |
| Cost | US$600 million (estimated cost) |
| Purpose | Scientific research and development, stockpile stewardship[1] |
Design edit
El Capitan will use an unknown number of AMD Instinct MI300A accelerated computing units (APUs).[2] The MI300A consists of 24 AMD Zen AMD64-based CPU cores, and a CDNA 3-based GPU integrated onto a single organic package, along with 128GB of HBM3 RAM.[3]
The floor space and number of racks for El Capitan have not yet been announced.
Blades are interconnected by an HPE Slingshot 64-port switch that provides 12.8 terabits/second of bandwidth. Groups of blades are linked in a dragonfly topology with at most three hops between any two nodes. Cabling is either optical or copper, customized to minimize cable length. Total cabling runs 145 km (90 mi).
El Capitan uses an APU architecture, where the CPU and GPU share an internal on-chip coherent interconnect.
History edit
El Capitan was ordered as a part of the Department of Energy's CORAL-2 initiative, intended to replace Sierra (supercomputer), an IBM/NVIDIA machine deployed in 2018. The original design envisioned hundreds of thousands of GPUs and 40 MW of power.[citation needed] LLNL partnered with HPE Cray and AMD to build the system.[4]
Three El Capitan prototypes – named rzVernal, Tioga, and Tenaya – themselves were powerful enough to be listed on the TOP200 supercomputer list in June, 2023.[5] rzVernal reached 4.1 petaflops.[6] In early July, the first components of El Capitan were installed at Lawrence Livermore, with complete installation expected by mid 2024.[7]
References edit
- ^ "Fiscal Year 2023 Stockpile Stewardship and Management Plan – Biennial Plan Summary Report to Congress" (PDF). United States Department of Energy. pp. 3–17. Retrieved May 27, 2023.
- ^ Shilov, Anton (July 6, 2023). "el-capitan-installation-begins-first-apu-exascale-system-shaping-up-for-2024". Anandtech. Top500.org. Retrieved July 15, 2023.
- ^ Smith, Ryan (January 25, 2023). "ces-2023-amd-instinct-mi300-data-center-apu-silicon-in-hand-146b-transistors-shipping-h223". Anandtech. Retrieved February 13, 2023.
- ^ Trader, Tiffany (August 13, 2019). "Cray Wins NNSA-Livermore "El Capitan" Exascale Contract". hpcwire.com. Retrieved February 13, 2023.
- ^ "June 2023 list". TOP500.org. Retrieved October 10, 2023.
- ^ Aaron Klotz (June 3, 2022). "Trio of Prototype AMD-Based El Capitan Supercomputers Already Rank in Top 200". Tom's Hardware.
- ^ Anton Shilov (July 6, 2023). "El Capitan Installation Begins: First APU-based Exascale System Shaping Up For 2024". Anandtech.