- Lingsheng system targets two exaFLOPS using only central processing units
- CPU-only architecture challenges GPU-dominated supercomputing industry standards
- System design incorporates high-bandwidth memory and high-speed interconnect networks
A Chinese computing center has announced plans to create a machine capable of achieving two exaFLOPS using only central processors.
The Lingsheng system, unveiled at a conference in April 2026 in Shenzhen, would group 47,000 processors into just 92 computing cabinets.
Lu Yutong, director of the National Supercomputing Center in Shenzhen and chief designer of the system, explained that the hardware and software stack is “completely independently controllable.”
Article continues below
A fundamentally different architectural strategy
Exascale machines in the industry currently rely heavily on GPU accelerators or specialized hardware.
This makes the CPU-only approach a major departure from established global trends.
The system leverages domestically produced high-performance processors, as well as high-bandwidth on-chip memory and high-speed interconnection networks.
It also incorporates 3D floating orthogonal calculation and full liquid cooling to manage thermal production.
According to the announcement, the Lingsheng platform achieves breakthroughs in six major technical areas: architecture, performance, power consumption, scheduling, scalability and reliability.
The system supports exascale computing power with exascale storage and petascale communication, and uses what officials described as the world’s largest-scale centralized liquid cooling technology.
A verification pilot phase uses 100 Huawei Kunpeng servers built on Arm-based Taishan cores, totaling 12,800 cores.
When scaled to full production, the same system design would integrate 1,580 blade servers using x86 processors with 101,120 cores and a theoretical peak greater than 10 petaflops.
The complete infrastructure also includes 36 network cabinets supporting interconnection of one million ports.
It will also feature 650 PB of planned storage distributed across 428 nodes and 67 liquid-cooled storage cabinets offering 10 TB/s of bandwidth.
The world’s fastest computer, the US Department of Energy’s El Capitan, runs on 44,544 AMD MI300A APUs, integrating CPU and GPU silicon in a single package.
If Lingsheng’s sustained performance of 2 exaFLOPS is achieved, it will surpass the El Capitan-measured Linpack score of 1,809 exaFLOPS.
On the other hand, the figure of 2 exaFLOPS for the Lingsheng system is a theoretical number, but El Capitan already has a theoretical value of 2.79 exaFLOPS.
Therefore, the claim of surpassing the world’s fastest computer does not seem feasible when comparing theoretical values with each other.
Unanswered Questions and Unproven Abilities
Several critical questions remain unanswered regarding the Lingsheng system, mainly because no benchmark data exists for the machine.
Although China says this system will not rely on any non-Chinese providers, the country’s domestic x86 options remain limited to Zhaoxin and Hygon.
None of these domestic alternatives have demonstrated processors capable of competing with current generation components from Intel or AMD.
The announcement also did not name specific suppliers for the production system and provided no operational timeline for its completion.
In terms of potential applications, the technology covers nine areas, including remote sensing, materials science, bioinformatics, meteorology, pharmaceuticals, oil exploration, artificial intelligence, life sciences and electromagnetic simulation.
A research team reported achieving 81% parallel scalability for first-principles calculations involving 100 million atoms.
Another group claimed that virtual screening of compounds on the scale of a trillion could improve efficiency by 1,000 times through a combination of AI and reinforcement learning.
However, these claims remain theoretical until a working machine produces verifiable benchmark results.
Via Tom’s material
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds.
