BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Fundamentals of Accelerated Computing with CUDA for CERN Summer St
 udents
DTSTART:20260731T063000Z
DTEND:20260731T153000Z
DTSTAMP:20260904T052800Z
UID:indico-event-1692411@indico.cern.ch
DESCRIPTION:Speakers: Simone Balducci (Universita e INFN\, Bologna (IT)\, 
 CERN)\n\nAbstract\nThis workshop teaches the fundamental tools and techniq
 ues for accelerating C/C++ applications to run on massively parallel GPUs
  with CUDA®. You’ll learn how to write code\, configure code paralleliz
 ation with CUDA\, optimize memory migration between the CPU and GPU accel
 erator\, and implement the workflow that you’ve learned on a new task
 —accelerating a fully functional\, but CPU-only\, particle simulator for
  observable massive performance gains. \nThe CUDA computing platform ena
 bles the acceleration of CPU-only applications to run on the world’s fas
 test massively parallel GPUs. Experience C/C++ application acceleration by
 :\n\nAccelerating CPU-only applications to run their latent parallelism on
  GPUs\nUtilizing essential CUDA memory management techniques to optimize a
 ccelerated applications\nExposing accelerated application potential for co
 ncurrency and exploiting it with CUDA streams\nLeveraging command line and
  visual profiling to guide and check your work\n\nFind more information on
  this workshop here\nLearning Objectives\nAt the conclusion of the worksho
 p\, you’ll have an understanding of the fundamental tools and techniques
  for GPU-accelerating C/C++ applications with CUDA and be able to:\n\nWri
 te code to be executed by a GPU accelerator\nExpose and express data and i
 nstruction-level parallelism in C/C++ applications using CUDA\nUtilize CUD
 A-managed memory and optimize memory migration using asynchronous prefetch
 ing\nLeverage command line and visual profilers to guide your work\nUtiliz
 e concurrent streams for instruction-level parallelism\nWrite GPU-accelera
 ted CUDA C/C++ applications\, or refactor existing CPU-only applications\,
  using a profile-driven approach \n\nPrerequisites\nBasic C/C++ competenc
 y including familiarity with variable types\, loops\, conditional statemen
 ts\, functions\, and array manipulations.\nPlease note that pictures and v
 ideos might be taken during the event. The pictures and videos might be us
 ed for communication about the event. By joining the lecture\, you are agr
 eeing to being featured in these communication actions. \n\nhttps://indic
 o.cern.ch/event/1692411/
LOCATION:31/3-004 - IT Amphitheatre (CERN)
URL:https://indico.cern.ch/event/1692411/
END:VEVENT
END:VCALENDAR