China has unveiled its new supercomputer, called LineShine, in Shenzen, which will deliver 2 ExaFlops of compute performance.
The Chinese LineShine Supercomputer Is All Set To Become The Fastest In The World, Exceeding 2 ExaFlops
During a conference at the National Supercomputing Center in Shenzhen today, the fastest domestic supercomputer project was announced. Known as LineShine, the Supercomputer will be built in two phases and will deliver over 2 ExaFlops of compute. Currently, the world's fastest supercomputer is the AMD-based El Capitan, which has a peak compute of 2.8 Exaflops.
Coming to the specifications, the LineShine Supercomputer is purely a CPU-only project with a sustained performance of over 2 ExaFlops. It will house an advanced architecture that leverages highly-efficient CPUs, high-bandwidth memory, and high-speed interconnect.
All of the tech inside the supercomputer will be domestically produced with zero reliance on international vendors. Besides this, the platform is going to offer an easy-to-use and practical ecosystem with a complete toolchain comprising a compiler, debugger, and various tuning tools.
The LineShine Supercomputer will have two phases. The pilot verification system will be based on 100 Huawei Kunpeng servers, totalling 12,800 cores. The industrial complex will house 1580 blade servers featuring x86 CPUs. This will total 101,120 cores with a theoretical peak of 10+ Petaflops. The system is going to feature a 16 4-Way server layout with a total of 2048 cores and four 8-way servers with a total of 1280 cores.
The entire system will house a total of 92 compute cabinets with a total of 47,000 CPUs, 36 network cabinets with large-scale expansion to hundreds of thousands of nodes, a million-port interconnect, and will all be assembled as China's largest supercomputing storage base & the largest liquid cooling solution in the world with 67 liquid-cooled storage cabinets, 428 storage nodes, and 10 TB/s of bandwidth.
Liquid cooling capabilities of the system are on a grand tier as the entire unit will house secondary pipes measuring 3214.7 meters and will have a net weight of 243.9 tons.
A lot of emphasis is being made on the storage capabilities of the supercomputer, with 650 PB of capacity planned for the entire system. Besides this, the Fusion Architecture will integrate SMT accelerators, which dramatically improve matrix and math operations, offering full-stack mixed precision that supports FP64, FP32, FP16, and INT8 capabilities for a large variety of AI models.
LineShine has already tested the per-CPU throughput in DeepSeek, reaching 578 Tokens Per Second. The overall throughput of the system will be 100 times higher. Besides supporting mainstream models such as Qwen, various domestically produced AI models, the LineShine supercomputer will also support general-purpose and HPC tasks such as remote sensing, materials, bioinformatics, meteorology, pharmaceuticals, oil exploration, life sciences, and electromagnetic simulation.
There's no specific operational timeframe, but given China's growing needs, we can see the LineShine supercomputer coming up around 2029-2030.
News Source: OGAWA, Tadashi
About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.
Follow Wccftech on Google to get more of our news coverage in your feeds.
