BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260702T030557Z
UID:4425f753-4fc8-416b-a939-813680463931
DTSTART:20160621T060000Z
DTEND:20160623T160000Z
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 both a high level (pragma-based) programming approach with OpenACC for 
 a fast porting startup\, and lower level approaches based on nVIDIA CUDA a
 nd OpenCL programming languages for finer grained computational intensive 
 tasks. A particular attention will be given on performance tuning and tech
 niques to overcome common data movement bottlenecks and patterns.\n\nLearn
 ing outcomes:\n\nThe students will learn the strengths and weaknesses of G
 PUs as accelerators\, the problem of data movement between host and device
  memories and how to overcome bottlenecks through concurrent computing/dat
 a-movement operations. The students will learn the basics of GPGPU program
 ming using both higher and lower level programming approaches. \n\nTarget
  Audience: \nResearchers and programmers interested in porting scientific 
 applications or use efficient post-process and data-analysis techniques in
  modern heterogeneous HPC architectures.\n\nTopics:\n\nOverview of archite
 ctural trends of GPUs in HPC. GPGPU parallel programming in heterogeneous 
 architectures. Basis of OpenACC\, CUDA and OpenCL programming.\n\nPrerequi
 sites: \n\nA basic knowledge of C or Fortran is mandatory. Programming and
  Linux or Unix. A basic knowledge of any parallel programming technique/pa
 radigm is recommended.\n\nGrant:\nThe lunch for the three days will be off
 ered to all the participants and some grants are available. The only requi
 rement to be eligible is to be not funded by your institution to attend th
 e course and to work or live in an institute outside the Rome area. The g
 rant  will be 300 euros for students working and living outside Italy and
  150 euros for students working and living in Italy (outside Rome). Some d
 ocumentation will be required and the grant will be paid only after a cert
 ified presence of minimum 80% of the lectures.\n\nFurther information abou
 t how to request the grant\, will be provided at the confirmation of the c
 ourse: about 3 weeks before the starting date.\n\nSUBSCRIPTIONS WILL BE OP
 EN ON MARCH 2016\n\nhttps://events.prace-ri.eu/event/475/
SUMMARY:Programming paradigms for new hybrid architectures (GPU) @Cineca
URL;VALUE=URI:https://events.prace-ri.eu/event/475/
END:VEVENT
END:VCALENDAR