- The Lingsheng system targets two exaFLOPS using only central processing units
- CPU-only architecture challenges GPU-dominated supercomputing industry standards
- System design integrates high-bandwidth memory and high-speed connectivity networks
A Chinese supercomputing center has announced plans for a machine that would reach two exaFLOPS using only central processors.
The Lingsheng system, unveiled at an April 2026 conference in Shenzhen, would pack 47,000 processors into just 92 computer cases.
Lu Yutong, director of the National Supercomputing Center in Shenzhen and the system’s chief designer, explained that the hardware and software stack is “completely independently controllable.”
The article continues below
A fundamentally different architectural strategy
The exascale machines in the industry currently rely heavily on GPU accelerators or specialized hardware.
This makes the CPU-only approach a significant departure from established global trends.
The system utilizes domestically produced high-performance CPUs together with on-chip high-bandwidth memory and high-speed connection networks.
It also incorporates 3D liquid orthogonal computing and full liquid cooling to control thermal output.
According to the announcement, the Lingsheng platform achieves breakthroughs in six major technical areas: architecture, performance, energy consumption, programming, scalability and reliability.
The system supports exascale computing power with exascale storage and petascale communications and uses what officials described as the world’s largest centralized liquid cooling technology.
A pilot verification 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 will incorporate 1,580 blade servers using x86 CPUs with 101,120 cores and a theoretical peak above 10 petaflops.
The complete infrastructure also has 36 network cabinets that support a million-port interconnection.
It will also contain 650 PB of planned storage across 428 nodes and 67 liquid-cooled storage cabinets delivering 10 TB/s of bandwidth.
The current fastest computer in the world, the US Department of Energy’s El Capitan, runs on 44,544 AMD MI300A APUs that integrate CPU and GPU silicon into a single package.
If Lingsheng’s sustained performance of 2 exaFLOPS is achieved, it will surpass El Capitan’s measured Linpack score of 1,809 exaFLOPS.
On the flip side, the 2 exaFLOPS figure for the Lingsheng system is a theoretical number, but the El Capitan already has a theoretical value of 2.79 exaFLOPS.
Therefore, the claim of surpassing the fastest computer in the world does not seem to be achievable when comparing theoretical values with each other.
Unanswered questions and unproven abilities
Several critical questions remain unanswered regarding the Lingsheng system, primarily because no benchmark data exists for the machine.
Although China claims that this system will not rely on any 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 parts from either Intel or AMD.
The announcement also failed to name specific suppliers for the production system and provided no operational timetable for completion.
On the potential application side, the technology spans nine areas, including remote sensing, materials science, bioinformatics, meteorology, pharmaceuticals, oil exploration, artificial intelligence, life sciences, and electromagnetic simulation.
A research team reported achieving parallel scalability of 81% for first-principles calculations involving 100 million atoms.
Another group claimed that virtual screening of trillion-scale compounds could improve efficiency by 1,000-fold through a combination of AI and reinforcement learning.
However, these remain theoretical claims until a working machine produces verifiable benchmark results.
Via Tom’s hardware
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds.
