home.social

#supercomputing — Public Fediverse posts

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

  1. The "AI" phenomenon and craze has been discussed endlessly, including the vast computational and energy requirements of dedicated #datacenters

    There is an important split betwen model building and inference of statistical models (and hence also #llm ) which means there is not one but two problems to solve.

    The first one is *highly* non-trivial. It bumps against #mooreslaw and requires #supercomputing networking tricks. The datacenter must operate as one giant machine

    openai.com/index/mrc-supercomp

  2. The "AI" phenomenon and craze has been discussed endlessly, including the vast computational and energy requirements of dedicated #datacenters

    There is an important split betwen model building and inference of statistical models (and hence also #llm ) which means there is not one but two problems to solve.

    The first one is *highly* non-trivial. It bumps against #mooreslaw and requires #supercomputing networking tricks. The datacenter must operate as one giant machine

    openai.com/index/mrc-supercomp

  3. The "AI" phenomenon and craze has been discussed endlessly, including the vast computational and energy requirements of dedicated #datacenters

    There is an important split betwen model building and inference of statistical models (and hence also #llm ) which means there is not one but two problems to solve.

    The first one is *highly* non-trivial. It bumps against #mooreslaw and requires #supercomputing networking tricks. The datacenter must operate as one giant machine

    openai.com/index/mrc-supercomp

  4. The "AI" phenomenon and craze has been discussed endlessly, including the vast computational and energy requirements of dedicated #datacenters

    There is an important split betwen model building and inference of statistical models (and hence also #llm ) which means there is not one but two problems to solve.

    The first one is *highly* non-trivial. It bumps against #mooreslaw and requires #supercomputing networking tricks. The datacenter must operate as one giant machine

    openai.com/index/mrc-supercomp

  5. The "AI" phenomenon and craze has been discussed endlessly, including the vast computational and energy requirements of dedicated #datacenters

    There is an important split betwen model building and inference of statistical models (and hence also #llm ) which means there is not one but two problems to solve.

    The first one is *highly* non-trivial. It bumps against #mooreslaw and requires #supercomputing networking tricks. The datacenter must operate as one giant machine

    openai.com/index/mrc-supercomp

  6. #hpc #supercomputing #linux.
    🚨🚨🚨🚨🚨
    The first NVL4 shelves is provisioning over the network! This is the beginning of the new system!

  7. #hpc #supercomputing #linux.
    🚨🚨🚨🚨🚨
    The first NVL4 shelves is provisioning over the network! This is the beginning of the new system!

  8. #hpc #supercomputing #linux.
    🚨🚨🚨🚨🚨
    The first NVL4 shelves is provisioning over the network! This is the beginning of the new system!

  9. Wall Street Journal: The quest to use AI to help find new drugs. “Lilly first announced a partnership with chip-maker Nvidia in October to build what it called the industry’s most powerful supercomputer, and expanded that in January with a $1 billion, five-year collaboration mixing their scientists and engineers in a new Bay Area lab aimed at discovering new medicines with AI tools. They […]

    https://rbfirehose.com/2026/05/05/wall-street-journal-the-quest-to-use-ai-to-help-find-new-drugs/
  10. Wall Street Journal: The quest to use AI to help find new drugs. “Lilly first announced a partnership with chip-maker Nvidia in October to build what it called the industry’s most powerful supercomputer, and expanded that in January with a $1 billion, five-year collaboration mixing their scientists and engineers in a new Bay Area lab aimed at discovering new medicines with AI tools. They […]

    https://rbfirehose.com/2026/05/05/wall-street-journal-the-quest-to-use-ai-to-help-find-new-drugs/
  11. Wall Street Journal: The quest to use AI to help find new drugs. “Lilly first announced a partnership with chip-maker Nvidia in October to build what it called the industry’s most powerful supercomputer, and expanded that in January with a $1 billion, five-year collaboration mixing their scientists and engineers in a new Bay Area lab aimed at discovering new medicines with AI tools. They […]

    https://rbfirehose.com/2026/05/05/wall-street-journal-the-quest-to-use-ai-to-help-find-new-drugs/
  12. Wall Street Journal: The quest to use AI to help find new drugs. “Lilly first announced a partnership with chip-maker Nvidia in October to build what it called the industry’s most powerful supercomputer, and expanded that in January with a $1 billion, five-year collaboration mixing their scientists and engineers in a new Bay Area lab aimed at discovering new medicines with AI tools. They […]

    https://rbfirehose.com/2026/05/05/wall-street-journal-the-quest-to-use-ai-to-help-find-new-drugs/
  13. #Linux #HPC #Supercomputing All, if you're affected by copy fail (CVE-2026-31431), and cannot patch your system, apply the mitigations ASAP and reboot. It's a PITA for cluster admins, but it's better than the alternative. Even if someone chooses _NOT_ to run the exploit, any system binary they target will become at the minimum, corrupted.

  14. An international team of physicists, including researchers from Jülich, used supercomputers to deliver a new high-precision calculation of the muon’s magnetic moment ⚡🧬

    The result reduces theoretical uncertainty and strongly validates the Standard Model of particle physics.

    The paper appeared in Nature.

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

    #FZJ #Supercomputing #HPC #ParticlePhysics #exa_JUPITER

    Image credit: Peter Allen, Ryan Allen, Universität Wuppertal

  15. The #UK government has launched #SovereignAI, a £500 million initiative to support homegrown #AI companies. The programme offers #funding, access to #supercomputing resources, and support for #talentacquisition. The first #investment is in #Callosum with six other companies receiving access to supercomputing capacity. gov.uk/government/news/ai-firm #tech #media #news

  16. Am heutigen World #Quantum Day dreht sich alles ums #Quantencomputing . HPE hat bei renommierten #Supercomputing - und Forschungszentren (auch beim LRZ) nachfragen lassen, wie die Technologie in der #Forschung eingesetzt wird. Sie können unterschiedliche Erfahrungen kennenlernen– heute ab 18 Uhr beim #HPECube. Hier können Sie sich registrieren: media.now.hpe.com/program.aspx
    Und hier findet sich das #Programm (zu beachten: die Zeitzonen): thecube.net/events/hpe/world-q

  17. Stephan Hachinger, LRZ researcher and expert in #research #data #management explains what the LRZ is developing for the European EXA4MIND project. This holds exciting prospects for #supercomputing and #HPC . Have a look: youtube.com/watch?v=HSbOZ00HASo

    #researchdata #science #IT4science
    @openwebsearcheu
    @opendata

  18. #Quantencomputing im Verein mit #Supercomputing: Bis zum 15. Mai bleibt noch Zeit, Paper und Vortrag für Q-Works in Pisa einzureichen.
    Das QEX Quantum Excellence Centre verbessert das Zusammenspiel von #HPC und Quantencomputing, sucht daher Ideen und Erfahrungen für neue Workloads, wenn beide Technologien gleichzeitig arbeiten sollen. Außerdem sind mehr Anwendungen fürs hybride Quanten- oder beschleunigte Supercomputing gefragt: q-works.github.io/cfp/

    #quantum #computing #IT
    @MunichQuantum

  19. #HPE adds #Blackwell, #Rubin systems to #Nvidia-backed #AI push
    HPE has expanded its Nvidia-based AI portfolio with new systems built on Blackwell and upcoming Rubin #GPU, alongside updates to its #Alletra Storage MP X10000, which it claims is the first object storage platform to achieve Nvidia-Certified Storage validation.
    Also announcing new Nvidia-powered #A Factory and #Supercomputing offerings, which include AI grids and enable #sovereignAI in Europe and US.
    blocksandfiles.com/ai-ml/2026/
    #GTC2026

  20. Thermodynamic Computing Advances with Design and Training

    BYLINE: Elizabeth Ball Newswise — What if the thermal noise that hinders the efficiency of both classical and…
    #NewsBeep #News #Science #AllJournalNews #AU #Australia #DOEScienceNewsSource #Energy #LawrenceBerkeleyNationalLaboratory #naturejournal #Newswise #QuantumMechanics #Supercomputing #Thermodynamics;Computing;ThermalNoise;Supercomputing;HighPerformaceComputing;Microelectronics
    newsbeep.com/au/522336/

  21. If you care about the hardware behind LLMs, agents, and frontier AI research, Huawei’s Atlas 950 SuperPoD is worth a closer look.
    #ArtificialIntelligence #Huawei #Ascend #Cloud #SuperComputing

  22. New Interfaces article!! (link below) Insightful article on the history of Indian supercomputing by Cornell U. STS scholar Andra Sonia Petrutiu "Disrupting Flattening Universal Ambitions: The Case of Supercomputing in India."

    #histsci #supercomputing #india #highperformancecomputing #software #technology #tech #science #sociology #anthropology #mediastudies #computerscience #ai #artificialintelligence #datascience #history

    cse.umn.edu/cbi/interfaces

  23. #JMU:
    "
    Umfassende Karte des Radiohimmels veröffentlicht

    Das Radioteleskop LOFAR hat die bislang detaillierteste Radiokarte des Nordhimmels geliefert. Sie eröffnet neue Einblicke in aktive Galaxien, Sternentstehung und seltene Himmelsobjekte.
    "
    uni-wuerzburg.de/aktuelles/pre

    19.2.2026

    #ASTRON #Astronomie #FZJülich #Himmelsdurchmusterung #Himmelskarte #JUWELS #LOFAR #LoTSS #Nordhimmel #Radioantronomie #Radioteleskop #Supercomputing

  24. Looking for a #job in High-Performance Computing? Our friends at University of Innsbruck are looking for a 𝗦𝗲𝗻𝗶𝗼𝗿 𝗦𝗰𝗶𝗲𝗻𝘁𝗶𝘀𝘁 (𝗗𝗼𝗰𝘁𝗼𝗿𝗮𝘁𝗲/𝗣𝗵𝗗) – 𝗛𝗣𝗖 𝗧𝗲𝗮𝗺𝗹𝗲𝗮𝗱 Applications open until 15 February: lfuonline.uibk.ac.at/public/karri... #hpc #highperformancecomputing #supercomputing

    Studienorganisation

  25. Tackling the #energy demands of #supercomputers through technology and experience: Satoshi Matsuoka, Director of the @riken Center for Computational Science, and Dieter Kranzlmüller, head of LRZ, signed a Memorandum of Understanding to establish close cooperation in the field of #energyefficiency Both #supercomputing centers rely on water-based cooling and aim to advance technology for thermal optimization as well as for the recovery and reuse of #wasteheat: lrz.de/en/news/detail/riken-r-

    #IT4Science

  26. Engadget: Musk claims Tesla will restart work on its Dojo supercomputer. “Elon Musk posted on X that Tesla will be restarting work on Dojo3, the third generation of its in-house supercomputer project. The Dojo team had been disbanded last year as the company prioritized the AI chips that run on board Tesla vehicles. Musk said the company is returning to the project ‘now that the AI5 chip design […]

    https://rbfirehose.com/2026/01/21/engadget-musk-claims-tesla-will-restart-work-on-its-dojo-supercomputer/
  27. 𝗔𝘂𝘀𝘁𝗿𝗶𝗮𝗻-𝗦𝗹𝗼𝘃𝗲𝗻𝗶𝗮𝗻 𝗛𝗣𝗖 𝗠𝗲𝗲𝘁𝗶𝗻𝗴 – 𝗿𝗲𝗴𝗶𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻 𝗼𝗽𝗲𝗻!

    Join the #supercomputing community at #ASHPC26, where HPC users and providers come together to share latest research results, challenges, solutions, and success stories. Call for abstracts and registration are open! Hurry up to secure a spot with an early bird fee 👉 ashpc.eu/event/27/page/30-regi

    ASHPC26
    🗓️ 8-10 April 2026
    📍 Vienna, Austria

    #EuroCC #hpc #ASHPC26 #ASC

  28. " Datenmanagement für verteilte Rechenabläufe: Eine iRODS-basierte #Konfiguration und ihre Leistungsfähigkeit", heißt ein Paper, das das europäische #Projekt EXA4MIND veröffentlicht hat: tiny.badw.de/ZpPR8
    Die Studie zeigt, wie sich eine #Dateninfrastruktur auf verteilten Systemen aufbauen lässt. Diese ergänzt nun die LEXIS-Plattform, die mehrere #Supercomputing-Zentren verbindet, lokal verteilte Datenverarbeitung und nun auch Speicherung ermöglicht.

  29. #AMD touts Instinct #MI430X, #MI440X, and #MI455X #AI accelerators and #Helios #rackscale AI architecture at CES — full MI400-series family fulfills a broad range of infrastructure and customer requirements
    The newly disclosed MI440X and MI455X are set to be optimized for low-precision workloads, such as FP4, FP8, & BF16. The previously disclosed MI430X targets both sovereign AI and #HPC thus fully supports FP32 and FP64 technical computing and traditional #supercomputing
    tomshardware.com/tech-industry

  30. This is me after with some of the attendees after successfully delivering my first workshop at the #Supercomputing India 2025 in Bengaluru.
    Taught how to perform an "Efficient Halo Exchange for stencil-based #scientific codes on the #Cerebras Wafer-Scale Engine 3"

    #cdac #hpc #parallelprogramming #cerebras #scientificcomputing #sci2025

  31. This is me after with some of the attendees after successfully delivering my first workshop at the India 2025 in Bengaluru.
    Taught how to perform an "Efficient Halo Exchange for stencil-based codes on the Wafer-Scale Engine 3"

  32. This is me after with some of the attendees after successfully delivering my first workshop at the #Supercomputing India 2025 in Bengaluru.
    Taught how to perform an "Efficient Halo Exchange for stencil-based #scientific codes on the #Cerebras Wafer-Scale Engine 3"

    #cdac #hpc #parallelprogramming #cerebras #scientificcomputing #sci2025

  33. This is me after with some of the attendees after successfully delivering my first workshop at the #Supercomputing India 2025 in Bengaluru.
    Taught how to perform an "Efficient Halo Exchange for stencil-based #scientific codes on the #Cerebras Wafer-Scale Engine 3"

    #cdac #hpc #parallelprogramming #cerebras #scientificcomputing #sci2025

  34. This is me after with some of the attendees after successfully delivering my first workshop at the #Supercomputing India 2025 in Bengaluru.
    Taught how to perform an "Efficient Halo Exchange for stencil-based #scientific codes on the #Cerebras Wafer-Scale Engine 3"

    #cdac #hpc #parallelprogramming #cerebras #scientificcomputing #sci2025

  35. Today is your last chance to register for the 3-day online course “Node-Level Performance Engineering”! 📣

    Organised in cooperation with the Erlangen National High Performance Computing Center (NHR@FAU), this course covers performance engineering approaches at the compute node level.

    Registration: app1.edoobox.com/en/LRZ/Online

    #HPC #Supercomputing #performanceengineering

  36. Can you claim to be a real #HPC software engineer if you've never coded with at least 2 of #Fortran, #MPI, #OpenMP, or #CUDA?

    Can you claim to be a modern #supercomputing scientist if you've never worked with at least 2 of #cloud, #AI/#ML, #FP16, or #RSEs?

  37. Can you claim to be a real #HPC software engineer if you've never coded with at least 2 of #Fortran, #MPI, #OpenMP, or #CUDA?

    Can you claim to be a modern #supercomputing scientist if you've never worked with at least 2 of #cloud, #AI/#ML, #FP16, or #RSEs?

  38. Can you claim to be a real #HPC software engineer if you've never coded with at least 2 of #Fortran, #MPI, #OpenMP, or #CUDA?

    Can you claim to be a modern #supercomputing scientist if you've never worked with at least 2 of #cloud, #AI/#ML, #FP16, or #RSEs?

  39. Can you claim to be a real #HPC software engineer if you've never coded with at least 2 of #Fortran, #MPI, #OpenMP, or #CUDA?

    Can you claim to be a modern #supercomputing scientist if you've never worked with at least 2 of #cloud, #AI/#ML, #FP16, or #RSEs?

  40. Can you claim to be a real #HPC software engineer if you've never coded with at least 2 of #Fortran, #MPI, #OpenMP, or #CUDA?

    Can you claim to be a modern #supercomputing scientist if you've never worked with at least 2 of #cloud, #AI/#ML, #FP16, or #RSEs?

  41. #AWS announced a $50 billion investment to build #AI and #supercomputing #infrastructure for #USgovernmentagencies. This initiative aims to accelerate #discovery and #decisionmaking across government missions, enabling agencies to accomplish tasks in hours that previously took weeks or months. The investment supports the White House’s #AIActionPlan. foxbusiness.com/markets/amazon #tech #media #news

  42. A team led by #MPIM & #DKRZ has received the prestigious HPCwire Readers’ Choice Award🏆, presented at the #SC25 conference in St. Louis, MO, US. It recognizes a simulation of the full Earth system with the #ICON model at 1.25-km resolution 🌐, which was carried out on the Alps & #exa_JUPITER supercomputers & achieved an unprecedented time compression of 145.7 simulated days per day. Congratulations on this outstanding achievement! 🎉 #HPC #Supercomputing #ClimateModeling
    🔗 hpcwire.com/2025-readers-edito

  43. Angekommen in St. Louis und auf der #sc25: Das LRZ-Team hat einige Impressionen geschickt und sich eingerichtet. Am Stand gibt es neue Technologien aus Projekten und Partnerschaften, vom @MunichQuantum oder photonische #Computer on Q.ANT. Und natürlich Anwendungen in #virtuelleRealität Schon gibt es auch eine erste Ehrung: HPCwire zeichnet das LRZ wird als Enabler von exzellenter computergestützter #Forschung aus. Wir berichten weiter.

    #it4Science #HPC #hpc4Germany #BavarianHPC #supercomputing