BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T100136Z
UID:61b119fd-8386-4c57-954a-867c88607bd4
DTSTART:20200415T070000Z
DTEND:20200417T103000Z
DESCRIPTION:This course gives an overview of the most relevant GPGPU comput
 ing techniques to accelerate computationally demanding tasks on HPC hetero
 geneous architectures based on GPUs.\n\nThe course will start with an arch
 itectural overview of modern GPU based heterogeneous architectures\, focus
 ing on its computing power versus data movement needs. The course will cov
 er a high level (pragma-based) programming approach with OpenACC for a fas
 t porting startup\, and lower level approaches based on nVIDIA CUDA  pro
 gramming languages for finer grained computational intensive tasks. . A pa
 rticular attention will be given on performance tuning and techniques to o
 vercome common data movement bottlenecks and patterns.\n\nSkills:\nBy the 
 end of the course\, students will be able to:\n\n\n	understand the strengt
 hs and weaknesses of GPUs as accelerators\n	program GPU accelerated applic
 ations using both higher and lower level programming approaches\n	overcome
  problems and bottlenecks regarding data movement between host and device 
 memories\n	make best use of independent execution queues for concurrent co
 mputing/data-movement operations\n\n\nTarget Audience: \nResearchers and p
 rogrammers interested in porting scientific applications or use efficient 
 post-process and data-analysis techniques in modern heterogeneous HPC arch
 itectures.\n\nPrerequisites: \n\nA basic knowledge of C or Fortran is mand
 atory. Programming and Linux or Unix. A basic knowledge of any parallel pr
 ogramming technique/paradigm is recommended.\n\nTeachers: Dr. L.Ferraro\, 
 Dr. S.Orlandini\n\n \nhttps://events.prace-ri.eu/event/972/
SUMMARY:[ONLINE] Programming paradigms for GPU devices @ CINECA
URL;VALUE=URI:https://events.prace-ri.eu/event/972/
END:VEVENT
END:VCALENDAR