home.social

#top500 — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #top500, aggregated by home.social.

fetched live
  1. China's LineShine Supercomputer: A Sense of Scale ⚡🖥️

    China's LineShine 灵晟 supercomputer is the most powerful supercomputer in the world 🌍, delivering a verified 2.198 exaFLOPS on the TOP500 benchmark. That means it can perform more than 2 quintillion (2×10¹⁸) calculations every second 🚀. The system consumes 42.2 megawatts of electricity ⚡ and is powered by more than 13.7 million conventional CPU cores 🧠.

    The Scale of Its Computing Power 📊

    Compared to Humanity 👥: If every person on Earth performed one mathematical calculation every second, without stopping ⏱️, it would take the entire global population about 4 years 📅 to equal the amount of computation LineShine completes in a single second.

    Consumer Hardware 💻: Matching LineShine's sustained performance would require more than 20 million of today's fastest consumer graphics cards 🖥️, such as an RTX 5090, working together in perfect synchronization 🔗.

    LineShine demonstrates the extraordinary scale of modern high-performance computing (HPC), enabling scientific calculations that would be practically impossible using conventional computers 🚀.

    #Supercomputer #LineShine #Exascale #Exaflop #HPC #HighPerformanceComputing #Technology #Innovation #Science #Engineering #Computing #CPU #ParallelComputing #ClimateScience #WeatherForecasting #Neuroscience #QuantumChemistry #Physics #MolecularScience #DataScience #ArtificialIntelligence #AI #Research #STEM #FutureTech #DigitalEarth #ComputerScience #TechFacts #NextGenComputing #TOP500

  2. China's LineShine Supercomputer: A Sense of Scale ⚡🖥️

    China's LineShine 灵晟 supercomputer is the most powerful supercomputer in the world 🌍, delivering a verified 2.198 exaFLOPS on the TOP500 benchmark. That means it can perform more than 2 quintillion (2×10¹⁸) calculations every second 🚀. The system consumes 42.2 megawatts of electricity ⚡ and is powered by more than 13.7 million conventional CPU cores 🧠.

    The Scale of Its Computing Power 📊

    Compared to Humanity 👥: If every person on Earth performed one mathematical calculation every second, without stopping ⏱️, it would take the entire global population about 4 years 📅 to equal the amount of computation LineShine completes in a single second.

    Consumer Hardware 💻: Matching LineShine's sustained performance would require more than 20 million of today's fastest consumer graphics cards 🖥️, such as an RTX 5090, working together in perfect synchronization 🔗.

    LineShine demonstrates the extraordinary scale of modern high-performance computing (HPC), enabling scientific calculations that would be practically impossible using conventional computers 🚀.

    #Supercomputer #LineShine #Exascale #Exaflop #HPC #HighPerformanceComputing #Technology #Innovation #Science #Engineering #Computing #CPU #ParallelComputing #ClimateScience #WeatherForecasting #Neuroscience #QuantumChemistry #Physics #MolecularScience #DataScience #ArtificialIntelligence #AI #Research #STEM #FutureTech #DigitalEarth #ComputerScience #TechFacts #NextGenComputing #TOP500

  3. The #TOP500 list, updated twice a year, ranks the world's most powerful non-distributed computer systems based on their performance executing the LINPACK benchmark.

    knowledgezone.co.in/posts/Meas

  4. The #TOP500 list, updated twice a year, ranks the world's most powerful non-distributed computer systems based on their performance executing the LINPACK benchmark.

    knowledgezone.co.in/posts/Meas

  5. #Arm CPUs Take Number 1 in Latest #Top500 List with #China's #LineShine
    At just under 2.2EF measured, it is roughly 22% faster than El Capitan.
    The fundamental building block of LineShine is the #LX2 #CPU. Thought to be designed by #Huawei, the LX2 CPU comprises two compute dies and has onboard HBM. Each die has 152 #Armv9-architecture CPU cores with both SVE and SME – so even without GPUs, the system is still making heavy use of vector and matrix accelerator blocks.
    servethehome.com/arm-cpus-take

  6. #Arm CPUs Take Number 1 in Latest #Top500 List with #China's #LineShine
    At just under 2.2EF measured, it is roughly 22% faster than El Capitan.
    The fundamental building block of LineShine is the #LX2 #CPU. Thought to be designed by #Huawei, the LX2 CPU comprises two compute dies and has onboard HBM. Each die has 152 #Armv9-architecture CPU cores with both SVE and SME – so even without GPUs, the system is still making heavy use of vector and matrix accelerator blocks.
    servethehome.com/arm-cpus-take

  7. #China's #supercomputer using local processors heads #TOP500 list
    #LineShine comprises 20,480 nodes, and each #LX2 #CPU “integrates two compute dies (304 cores total) and eight on-package HBM stacks (32 GB, 4 TB/s aggregate bandwidth).” Machine also runs #KylinOS – a #Linux distribution that features contributions from around the world. Nodes are interconnected via the #LingQi high-speed network with a dual-plane multi-rail fat-tree topology, offering 1.6 Tb/s bandwidth."
    theregister.com/hpc/2026/06/24

  8. #China's #supercomputer using local processors heads #TOP500 list
    #LineShine comprises 20,480 nodes, and each #LX2 #CPU “integrates two compute dies (304 cores total) and eight on-package HBM stacks (32 GB, 4 TB/s aggregate bandwidth).” Machine also runs #KylinOS – a #Linux distribution that features contributions from around the world. Nodes are interconnected via the #LingQi high-speed network with a dual-plane multi-rail fat-tree topology, offering 1.6 Tb/s bandwidth."
    theregister.com/hpc/2026/06/24

  9. JUPITER⚡remains one of the world’s most powerful supercomputers on the #TOP500 list!

    🥇 No. 1 in Europe
    🌎 No. 5 worldwide
    🌱 The most energy-efficient computer in the exascale class

    Read more: fz-juelich.de/en/news/archive/

  10. JUPITER⚡remains one of the world’s most powerful supercomputers on the #TOP500 list!

    🥇 No. 1 in Europe
    🌎 No. 5 worldwide
    🌱 The most energy-efficient computer in the exascale class

    Read more: fz-juelich.de/en/news/archive/

  11. On the 67th edition of the TOP500, #LineShine debuts as the new No. 1 system, ending El Capitan's run atop the list and becoming the fifth Exascale system overall. It is the first China-based system to lead the #TOP500 since Sunway TaihuLight in 2017 top500.org/lists/top500/2026/0

  12. On the 67th edition of the TOP500, #LineShine debuts as the new No. 1 system, ending El Capitan's run atop the list and becoming the fifth Exascale system overall. It is the first China-based system to lead the #TOP500 since Sunway TaihuLight in 2017 top500.org/lists/top500/2026/0

  13. Ein starkes Signal für europäisches Supercomputing: Unser Exascale-Rechner #exa_JUPITER ⚡ zählt weiter zur Weltspitze. In der aktuellen #TOP500-Liste belegt JUPITER Platz 5 weltweit – und ist der energieeffizienteste Supercomputer der Exascale-Klasse. 🏆

    fz-juelich.de/en/news/archive/

    #HPC #Supercomputing

  14. Ein starkes Signal für europäisches Supercomputing: Unser Exascale-Rechner #exa_JUPITER ⚡ zählt weiter zur Weltspitze. In der aktuellen #TOP500-Liste belegt JUPITER Platz 5 weltweit – und ist der energieeffizienteste Supercomputer der Exascale-Klasse. 🏆

    fz-juelich.de/en/news/archive/

    #HPC #Supercomputing

  15. Yesterday was #Top500 day at #ISC26, today is #HPC ingenuity day - where #supercomputer centres & vendors alike exercise imaginations to claim how they are leaders in the world of #supercomputing - let each enjoy their moment of limelight 👏

    linkedin.com/posts/andrewjones

  16. Yesterday was #Top500 day at #ISC26, today is #HPC ingenuity day - where #supercomputer centres & vendors alike exercise imaginations to claim how they are leaders in the world of #supercomputing - let each enjoy their moment of limelight 👏

    linkedin.com/posts/andrewjones

  17. Surprise! #China's #LineShine Takes Number 1 on #TOP500
    The #supercomputer in Shenzhen was declared the world’s fastest. It uses only standard microprocessors (#CPU) and not the special-purpose chips called graphics processing units. These chips are an original design based on a set of instructions licensed by #Arm Holdings. The Shenzhen scientists have also sought recognition for the new machine through 14 submissions for the #GordonBellPrize.
    hpcwire.com/2026/06/23/surpris

  18. Surprise! #China's #LineShine Takes Number 1 on #TOP500
    The #supercomputer in Shenzhen was declared the world’s fastest. It uses only standard microprocessors (#CPU) and not the special-purpose chips called graphics processing units. These chips are an original design based on a set of instructions licensed by #Arm Holdings. The Shenzhen scientists have also sought recognition for the new machine through 14 submissions for the #GordonBellPrize.
    hpcwire.com/2026/06/23/surpris

  19. #China's #LineShine #supercomputer dethrones US' El Capitan, secures first place in #Top500 list — first machine in rankings to sustain more than 2 #ExaFLOPS using only #CPU
    Deployed at National Supercomputing Centre and built by Shenzhen Cloud Computing Center using semi-custom 304-core LX2 processors based on #Armv9 instruction set and running at 1.55GHz. Machine employs 13.79 million cores in total, uses proprietary LingQi interconnect, and consumes 42.2 MW of power.
    tomshardware.com/tech-industry

  20. #China's #LineShine #supercomputer dethrones US' El Capitan, secures first place in #Top500 list — first machine in rankings to sustain more than 2 #ExaFLOPS using only #CPU
    Deployed at National Supercomputing Centre and built by Shenzhen Cloud Computing Center using semi-custom 304-core LX2 processors based on #Armv9 instruction set and running at 1.55GHz. Machine employs 13.79 million cores in total, uses proprietary LingQi interconnect, and consumes 42.2 MW of power.
    tomshardware.com/tech-industry

  21. A lot of the focus on today’s latest #Top500 list will obviously focus on the new #1 system from China.

    But the other big story is that a single commercial user site (ENI) now has 2 #supercomputers in the top 10. I don’t think that has ever happened before?

    #HPc #ISC26

  22. A lot of the focus on today’s latest #Top500 list will obviously focus on the new #1 system from China.

    But the other big story is that a single commercial user site (ENI) now has 2 #supercomputers in the top 10. I don’t think that has ever happened before?

    #HPc #ISC26

  23. The new #TOP500 is out with Erich Strohmaier presenting its highlights

    China Retakes #1 (HPL & HPCG)
    - With LineShine an All-CPU System

    The Exascale Club
    - Five Systems Now Break the Exaflop Barrier

    Industrial Systems by Eni S.p.A. rank 6th & 8th in TOP500

    #ISC26 #Connecting_the_Dots #HPC #AI

  24. #Top500 FUD:

    Meaningless because not all #supercomputers are listed. - This has always been true.

    HPL doesn’t represent range of real applications. - No single benchmark ever will.

    #FP64 isn’t critical anymore. - It’s still heavily used in many/most production #HPC apps.

  25. #Top500 FUD:

    Meaningless because not all #supercomputers are listed. - This has always been true.

    HPL doesn’t represent range of real applications. - No single benchmark ever will.

    #FP64 isn’t critical anymore. - It’s still heavily used in many/most production #HPC apps.

  26. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 02/06/2026

    Carte :
    framacarte.org/m/157544/

    Source :
    rncmobile.net/top/500mbps

  27. CPU не умер, он просто ждал. Китай строит двухэксафлопсный суперкомпьютер без единого GPU — прорыв, необходимость, фейк?

    El Capitan, Frontier, Aurora, JUPITER Booster — четыре нынешние эксафлопсные системы из рейтинга Top500, первые строчки суперкомпьютерной табели о рангах. Все четыре используют GPU-ускорение. Это архитектурный консенсус, который формировался около десяти лет и к 2024 году стал самоочевидным: хочешь эксамасштаб — используй GPU. На этом фоне Китай объявляет о суперкомпьютере исключительно на CPU, без единого ускорителя. Это либо достаточно нетривиальное архитектурное решение, либо красивая история про независимость и импортозамещение. Скорее всего, оба варианта одновременно. Но чтобы это понять, нужно посмотреть назад и выяснить, как GPU вообще оказался на вершине суперкомпьютинга и почему это происходило не один год. Да и вообще неплохо бы разобраться, как GPU добрался до вершины суперкомпьютинга, что в этом противостоянии осталось от CPU и почему именно Китай решил поставить на него. Разобраться, в чём там дело

    habr.com/ru/companies/ru_mts/a

    #суперкомпьютер #top500 #железо #китай #производительность #itинфраструктура #рейтинг

  28. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 08/04/2026

    Carte : framacarte.org/m/157544/

    Source : rncmobile.net/top/500mbps

  29. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 09/03/2026

    Source : rncmobile.net/top/500mbps

  30. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 05/02/2026

    Source : rncmobile.net/top/500mbps

  31. #NASA Launches Its Most Powerful, Efficient #Supercomputer bringing #Athena Online at #AmesResearchCenter
    Composed of 4 #HPE #Cray #EX4000 racks, with 264,144 cores, 768TB of memory, and a 20.132 Pflop/s theoretical peak! Likely placing it in the top 100 of the #Top500 list.
    nasa.gov/technology/computing/

  32. #NASA Launches Its Most Powerful, Efficient #Supercomputer bringing #Athena Online at #AmesResearchCenter
    Composed of 4 #HPE #Cray #EX4000 racks, with 264,144 cores, 768TB of memory, and a 20.132 Pflop/s theoretical peak! Likely placing it in the top 100 of the #Top500 list.
    nasa.gov/technology/computing/

  33. Recap: SC25 was a huge success for JSC! 🚀
    • JUPITER: 1.0 ExaFLOP/s ⚡, #4 worldwide, #1 Europe on #TOP500 & most energy-efficient in class
    • HPCwire Readers’ Choice Award 🏆 “Top Supercomputing Achievement”
    • ICON climate team wins Gordon Bell Prize 🌍 & HPCwire Top HPC-Enabled Scientific Achievement
    🇪🇺 European HPC Passport:
    13 organisations, lots of engagement — thanks to everyone who joined the stamp collection!
    👉 fz-juelich.de/en/jsc/news/news
    #SC25 #FZJ #HPC #JUPITER #QuantumComputing #EuroHPC

  34. Recap: SC25 was a huge success for JSC! 🚀
    • JUPITER: 1.0 ExaFLOP/s ⚡, #4 worldwide, #1 Europe on #TOP500 & most energy-efficient in class
    • HPCwire Readers’ Choice Award 🏆 “Top Supercomputing Achievement”
    • ICON climate team wins Gordon Bell Prize 🌍 & HPCwire Top HPC-Enabled Scientific Achievement
    🇪🇺 European HPC Passport:
    13 organisations, lots of engagement — thanks to everyone who joined the stamp collection!
    👉 fz-juelich.de/en/jsc/news/news
    #SC25 #FZJ #HPC #JUPITER #QuantumComputing #EuroHPC

  35. Europejski przełom w erze eksaskali. Superkomputer JUPITER dołącza do elity TOP500

    Podczas konferencji SC25 ogłoszono nową edycję prestiżowego rankingu superkomputerów TOP500.

    Najnowszy ranking potwierdza dominację USA w dziedzinie mocy obliczeniowej, ale przynosi historyczny moment: Europa zyskała swój pierwszy system eksaskalowy (o mocy co najmniej 1 Eflops), wprowadzając się do elitarnego grona technologicznych gigantów. Jednak to co daje się zauważyć, to nie tylko walka o surową moc, ale też o efektywność energetyczną.

    Globalni giganci: 4 systemy eksaskalowe

    Na szczycie rankingu bezdyskusyjnie prowadzi Ameryka. Globalny lider to superkomputer El Capitan, który po raz kolejny umocnił swoją pozycję z mocą 1,809 eksaflops/s. Jednak największym wydarzeniem jest historyczny awans Europy do eksaskali.

    Pierwszym superkomputerem Starego Kontynentu o mocy eksaskalowej jest zainstalowane w Niemczech JUPITER Booster. Według najnowszych notowań Top500 sprzęt zainstalowany w Centrum EuroHPC w niemieckim Jülich, stał się czwartym i pierwszym poza USA superkomputerem eksaskalowym. Osiąga 1000 Eflops/s i bazuje na architekturze Eviden BullSequana XH3000 i superchipach NVIDIA Grace Hopper.

    Podium cały czas okupują Stany Zjednoczone ze swoimi superkomputerami, kolejno: El Capitan, Frontier i Aurora.

    Green500: zwycięstwo efektywności

    Równolegle opublikowany, najnowszy ranking Green500 (najbardziej energooszczędnych systemów) pokazuje, że topowe miejsca zdominowały maszyny oparte na układach NVIDIA Grace Hopper. Najlepsze wyniki osiągnęły francuskie systemy KAIROS (73,28 GFlops/W) i ROMEO-2025.

    Jest też polski akcent na liście Green500. Zainstalowany w Krakowie superkomputer Helios znalazł się wśród 20 najbardziej energooszczędnych systemów na świecie.

    Podsumowując, najświeższe zestawienie TOP500 przedstawia cztery działające systemy eksaskalowe (trzy zlokalizowane w USA i jeden w Europie) oraz potwierdza, że wojna na rynku HPC toczy się między AMD (siła surowa) a Nvidią (efektywność i AI).

    Ranking jest niestety niepełny

    Choć ranking TOP500 jest najważniejszym miernikiem mocy obliczeniowej, jeżeli chodzi o superkomputery, to w najnowszej edycji nie jest on pełny. USA umacnia globalną dominację, posiadając trzy z czterech systemów eksaskalowych, a sam tylko superkomputer El Capitan jest warty więcej niż cała reszta maszyn z rankingu, które znajdują się poza pierwszą dwudziestką. Jednakże Chiny, które historycznie dominowały na liście pod względem ilości (liczby systemów), celowo nie zgłaszają swoich najpotężniejszych, nowych maszyn eksaskalowych, takich jak Tianhe-3. W efekcie na oficjalnej liście liderem Azji jest Japonia (system Fugaku), a Chińczycy, choć z pewnością dysponują dużo większą mocą, są na papierze strategicznie nieobecni w walce o czołówkę. Ale tylko na papierze, warto o tym pamiętać.

    Niemcy uruchomiły Jupitera – najpotężniejszy superkomputer Europy. Kanclerz Merz: „Chcemy stać się krajem AI”

    #eurohpc #exaflops #green500 #heliosKrakow #jupiterBooster #news #nvidiaGraceHopper #superkomputery #top500

  36. #Europe joins US as #exascale superpower after #Jupiter clinches #Top500 Number 4 Slot!
    #EuroHPC's biggest iron still has more to give with Universal Cluster expansion expected next year
    Jupiter became first public system outside the US to exceed one #exaFLOPS of double precision performance on High-Performance Linpack (#HPL) benchmark.. We'll note that China is believed to possess several exascale-class systems, but has become increasingly secretive about their existence
    theregister.com/2025/11/17/eur

  37. #Europe joins US as #exascale superpower after #Jupiter clinches #Top500 Number 4 Slot!
    #EuroHPC's biggest iron still has more to give with Universal Cluster expansion expected next year
    Jupiter became first public system outside the US to exceed one #exaFLOPS of double precision performance on High-Performance Linpack (#HPL) benchmark.. We'll note that China is believed to possess several exascale-class systems, but has become increasingly secretive about their existence
    theregister.com/2025/11/17/eur

  38. JUPITER ⚡ is the first Exascale Supercomputer in Europe 🎉🥂🥳

    The newly published #Top500 list – set to be officially presented tomorrow during the #SC25 BoF "TOP500 Supercomputers" – shows #exa_jupiter as the 4️⃣th fastest supercomputer worldwide 🌐 and the most energy-efficient 🌱 among Exascale systems. ⚡

    ➡️ Press release: fz-juelich.de/en/news/archive/

  39. JUPITER ⚡ is the first Exascale Supercomputer in Europe 🎉🥂🥳

    The newly published #Top500 list – set to be officially presented tomorrow during the #SC25 BoF "TOP500 Supercomputers" – shows #exa_jupiter as the 4️⃣th fastest supercomputer worldwide 🌐 and the most energy-efficient 🌱 among Exascale systems. ⚡

    ➡️ Press release: fz-juelich.de/en/news/archive/

  40. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 26/10/2025

    Source : rncmobile.net/top/500mbps

  41. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 06/10/2025

    Source : rncmobile.net/top/500mbps

  42. #RNCMobile #Freemobile #Top500

    Débits descendants moyens supérieurs à 1250/1000/750/500 Mbps.

    Nombre de sites par département au 22/09/2025

    Source : rncmobile.net/top/500mbps

  43. Внутри лидеров TOP500 суперкомпьютеров планеты: El Capitan, Frontier и Aurora

    Привет! На связи команда mClouds. В июне 2025 года вышло новое издание рейтинга TOP500, который ранжирует самые производительные суперкомпьютеры мира. Тройка лидеров сохранила свои позиции: El Capitan, Frontier и Aurora — все три системы перешагнули показатель в 1 экзафлопс и принадлежат Министерству энергетики США. Мы решили разобраться, что скрывается за цифрами производительности: какие технологии позволили El Capitan достичь невероятных 1,742 экзафлопса, что помогло Frontier и Aurora удержаться в тройке лидеров и какие проекты могут бросить им вызов в ближайшем будущем.

    habr.com/ru/companies/mclouds/

    #itинфраструктура #суперкомпьютеры #экзафлопс #el_capitan #frontier #aurora #top500

  44. An intruiging article by IEEE Spectrum about our new exascale supercomputer ⚡JUPITER shows why it is high time to “have some science done on the machine,” as our director Thomas Lippert puts it. 😊👌
    Read the article here: spectrum.ieee.org/jupiter-exas
    #exa_JUPITER #HPC #Top500 #FZJ

  45. An intruiging article by IEEE Spectrum about our new exascale supercomputer ⚡JUPITER shows why it is high time to “have some science done on the machine,” as our director Thomas Lippert puts it. 😊👌
    Read the article here: spectrum.ieee.org/jupiter-exas
    #exa_JUPITER #HPC #Top500 #FZJ

  46. By becoming the fastest #supercomputer in Europe JUPITER propels European Computing Power. #exa_JUPITER, developed by @fzj_jsc and owned by #EuroHPC JU, ranks 4th on the June 2025 #TOP500 list of the world’s fastest supercomputers and stands as the most energy-efficient system of the top 5 of the listing, which was presented today at the International Supercomputing Conference #ISC25. fz-juelich.de/en/news/archive/