TOP500 list<\/a> has unveiled a remarkable shift in the supercomputing landscape as of June 2026. China’s LineShine system, located at the Shenzhen National Supercomputing Center, has made a tremendous debut by securing the top position. This monumental achievement displaces the American supercomputer El Capitan, which had maintained dominance for several years. It marks a significant turn, as China has not held this title since 2017, effectively ending a decade-long North American supremacy.<\/p>\nUnprecedented Raw Power<\/h3>\n
LineShine’s performance is nothing short of extraordinary. According to benchmark tests, it has achieved a staggering 2,198 exaflops, surpassing El Capitan, which stands at 1,809 exaflops. This impressive result makes China’s supercomputer 20% more powerful than its American counterpart, establishing a new benchmark in the sphere of global supercomputing.<\/p>\n
A Surprising Architecture: Zero GPUs<\/h3>\n
What sets LineShine apart from its competitors is not just its performance but its architecture. Unlike most modern supercomputers that heavily employ GPUs from Nvidia or AMD for processing, LineShine operates solely on CPUs. This innovative approach makes the performance even more astonishing and highlights the unique engineering prowess behind this machine.<\/p>\n
The Heart of the Supercomputer: LX2 CPU<\/h3>\n
LineShine\u2019s core is built around the LX2 CPU, designed by Huawei. Each CPU comprises two dies equipped with high-bandwidth memory (HBM). Each die contains 152 ARMv9 cores, which enable exceptional processing of vectors and matrices, even without the aid of GPUs. In total, LineShine houses 304 processors, resulting in a staggering 13,789,440 cores that power its immense computational capabilities.<\/p>\n
Navigating Trade Challenges<\/h3>\n
One significant factor contributing to the design of LineShine is the ongoing U.S.-China trade conflict. Stricter tariffs and export restrictions on hardware and software have complicated the acquisition of Nvidia\u2019s GPUs for AI processing. Despite these challenges, China continues to demonstrate its remarkable technological advancement, achieving this feat without relying on public funding from the Chinese government.<\/p>\n
AI Clusters vs. Traditional Supercomputers<\/h3>\n
As impressive as LineShine’s capabilities are, they highlight an evolving landscape in high-performance computing (HPC). While supercomputers like LineShine dominate the TOP500 rankings, AI clusters created by major tech companies may eclipse them in sheer computational power. Jimmy Goodrich from the University of California remarks that if hyperscalers entered the competition, faster systems would likely not even place in the top five.<\/p>\n
A Comparison of Purpose: Precision vs. Speed<\/h3>\n
It’s essential to understand the distinctions between traditional supercomputers and AI clusters. The latter typically uses reduced precision formats (like FP16 or FP8) to maximize speed and efficiency, processing large volumes of data quickly. Conversely, supercomputers like LineShine operate on the FP64 standard, which prioritizes precision for scientific computations, making them less suitable for rapid data processing.<\/p>\n
The European Scene in Supercomputing<\/h3>\n
The latest TOP500 list also reveals promising developments for European supercomputing. The HPC7 system, developed by Eni, ranks directly at number six, while Germany\u2019s Jupiter Booster holds the fifth position. Altogether, Europe boasts four systems in the top 10, with eight appearing in the top 20, showing a robust presence in the global digital arena.<\/p>\n
\n
General News – 2<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"China Leads the Supercomputing Race with LineShine The biannual TOP500 list has unveiled a remarkable shift in the supercomputing landscape as of June 2026. China’s LineShine system, located at the Shenzhen National Supercomputing Center, has made a tremendous debut by securing the top position. This monumental achievement displaces the American supercomputer El Capitan, which had […]<\/p>\n","protected":false},"author":1,"featured_media":233701,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[36399],"tags":[2397,18866,3477,1250,11301,49985,110],"class_list":["post-233700","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology","tag-china","tag-interesting","tag-part","tag-powerful","tag-story","tag-supercomputer","tag-world"],"_links":{"self":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/233700","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/comments?post=233700"}],"version-history":[{"count":1,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/233700\/revisions"}],"predecessor-version":[{"id":233702,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/233700\/revisions\/233702"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/media\/233701"}],"wp:attachment":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/media?parent=233700"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/categories?post=233700"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/tags?post=233700"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}