China Prepares New Technology That Could Disrupt America
China has once again made a significant breakthrough in technological innovation development. The National Supercomputing Centre in China has announced plans to develop a machine that will achieve a speed of 2 exaFLOPS using only processors.
The Lingsheng system, unveiled at the April 2026 conference in Shenzhen, will house 47,000 processors in 92 computing cabinets.
Lu Yutong, director of the National Supercomputing Centre in Shenzhen and the system’s chief designer, explained that the full hardware-software stack can be independently controlled.
The development of this new machine could eliminate dependence on GPU accelerators or specialised hardware. This CPU-only approach represents a completely different architectural strategy from current trends.
The system utilises high-performance domestically produced CPUs, along with on-chip high-bandwidth memory (HBM) and high-speed interconnection networks.
It also incorporates 3D orthogonal floating-point computing and full liquid cooling to manage thermal output.
According to the announcement, the Lingsheng platform achieves breakthroughs in six key technical areas: architecture, performance, energy consumption, programming, scalability, and reliability, as cited from TechRadar on Tuesday (5/5/2026).
The system supports exascale computing power with exascale storage and petascale communication, and employs the world’s largest-scale centralised liquid cooling technology, according to officials.
The pilot verification phase uses 100 Huawei Kunpeng servers built on Arm-based Taishan cores, with a total of 12,800 cores.
Upon full production, the same system design will include 1,580 blade servers using x86 CPUs with 101,120 cores and a theoretical peak above 10 petaflops.
The complete infrastructure also features 36 network cabinets supporting one million port interconnections.
Additionally, there will be 650 PB of planned storage distributed across 428 nodes and 67 liquid-cooled storage cabinets providing 10 TB/s bandwidth.
America at Risk of Decisive Defeat?
The world’s current fastest computer, El Capitan from the US Department of Energy, runs on 44,544 AMD MI300A APUs, which integrate CPU and GPU silicon in a single package.
If the Lingsheng system’s sustained performance of 2 exaFLOPS is achieved, it would surpass El Capitan’s measured Linpack score of 1.809 exaFLOPS.
On the other hand, the 2 exaFLOPS figure for the Lingsheng system is theoretical, while El Capitan already has a theoretical value of 2.79 exaFLOPS.
Therefore, the claim of surpassing the world’s fastest computer appears unachievable when comparing theoretical values against each other.
There are still several questions that need to be answered regarding the Lingsheng system. Primarily, as there is no benchmark data available as a reference for evaluating the machine.
Although China asserts that the system will not rely on non-Chinese vendors, the country’s domestic x86 options remain limited to Zhaoxin and Hygon.
None of these domestic alternatives have demonstrated processors that can compete with current-generation components from US firms Intel or AMD.
The announcement also failed to mention specific suppliers for the production system and did not provide an operational timeline for its completion.
From the perspective of potential applications, the technology covers nine fields, including remote sensing, materials science, bioinformatics, meteorology, pharmaceuticals, oil exploration, artificial intelligence, life sciences, and electromagnetic simulation.
One research team reported achieving 81% parallel scalability for first-principles calculations involving 100 million atoms.
Another group claimed that virtual screening of trillion-scale compounds could improve efficiency up to 1,000 times through a combination of AI and reinforcement learning.
However, these remain theoretical claims until a functioning machine produces verifiable benchmark results.