BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Running and Scaling Performant Deep Learning Models on JSC Superco
 mputers
DTSTART:20260601T110000Z
DTEND:20260603T150000Z
DTSTAMP:20260528T162000Z
UID:indico-event-477@events.hpc-portal.eu
CONTACT:ai-courses-jsc@fz-juelich.de
DESCRIPTION:This course will take place as an online event. The link to th
 e streaming platform will be provided to the registrants only. The course 
 will be held in English. \nCourse Content:\nIn this workshop\, we will gu
 ide you through the first steps of using the JSC supercomputer machines fo
 r your own AI application. This workshop should be tailored to your needs
  – and our team will guide you through questions like:\n\nHow do I get 
 access to the JSC machines?  \nHow do I use the pre-installed\, optimis
 ed software? \nHow can I run my own code? \nHow can I store data so 
 I can access it fast in training? \nHow can I parallelise my training a
 nd use more than one GPU? \nHow can I profile AI code to identify bottl
 enecks and improve efficiency?  \nHow can I evaluate whether my code can
  scale beyond a single node?\n\nIn this workshop\, we will try to get yo
 ur code and your workflow running and make your start on the JSC supercomp
 uters as smooth as possible. \nYou will leave with a working setup and 
 a clearer understanding of how to iterate efficiently – whether that m
 eans speeding up a single GPU run\, scaling to multiple GPUs\, or moving t
 o multi-node training.\nAfter this course\, you are not only ready to use 
 Helmholtz AI COmpute REssources (HAICORE) but you have made your first ste
 ps into unlocking compute resources even on the largest scale with a comp
 ute time application at the Gauss Supercomputing Center.\nThis workshop
  will be held in a small group setting with enough space to address your q
 uestions. Please give us an indication on what topics you are interested 
 in and we will try to adjust.\nPrerequisites:\nFor this course\, you shoul
 d have some initial experience with running Python and machine learning fr
 ameworks like TensorFlow and PyTorch.\nA personal institutional email addr
 ess (university/research institution\, government agency\, organisation\, 
 or company) is required to register for JSC training courses. If you don't
  have an institutional email address\, please get in touch with the contac
 t person for this course.\nLanguage:\nThis course is given in English.\nDu
 ration:\n3 half days\nDates:\n1-3 June 2026\, 13:00-17:00 each day\nVenue:
 \nOnline\nNumber of Participants:\nMinimum 5\, maximum 25\nInstructors:\nS
 abrina Benassou\, Ismail Khalfaoui\, Javad Kasravi (JSC)\n\nhttps://events
 .hpc-portal.eu/event/477/
LOCATION:Online
URL:https://events.hpc-portal.eu/event/477/
END:VEVENT
END:VCALENDAR