China's LineShine Claims Top Spot on Global Supercomputer List

The LineShine supercomputer, installed at the National Supercomputing Centre in Shenzhen, achieved 2.198 exaflops on the High‑Performance Linpack benchmark, surpassing the United States’ El Capitan by more than 20 %. This marks the first time since 2017 that a Chinese system has led the TOP500 ranking, which is published bi‑annually at the International Supercomputing Conference in Hamburg.

LineShine’s architecture relies solely on standard central‑processing units (CPUs) rather than the graphics‑processing units (GPUs) used by most leading machines. The system uses custom 304‑core LX2 processors based on the Armv9 instruction set, an in‑house LingQi interconnect and the Kylin operating system, and consumes about 42.2 MW of power. Analysts note that the CPU‑only design demonstrates a path for China to achieve exascale performance while circumventing U.S. export controls on advanced AI chips.

The achievement intensifies the ongoing technology rivalry between China and the United States, where American firms dominate AI‑chip markets and U.S. government policies have restricted advanced semiconductor exports to China. While LineShine dominates the traditional scientific‑computing benchmark, experts say it does not yet lead on AI‑oriented mixed‑precision tests.

The supercomputer will be used for climate modelling, material science, nuclear‑stockpile stewardship and other complex simulations, underscoring its strategic importance for both scientific research and national security.