BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260705T221215Z
UID:709002b0-d0e1-41c5-9985-ae16f1dac77c
DTSTART:20201104T080000Z
DTEND:20201105T160000Z
DESCRIPTION:Modern high core count CPUs and accelerators\, such as GPUs\, h
 ave been shown to provide significant performance benefits in many differe
 nt applications. However\, for a novice\, or even for a moderately experie
 nced scientist or programmer\, it is not always clear which applications c
 ould potentially benefit from these resources and which do not. For exampl
 e\, a Nvidia V100 GPU can perform artificial intelligence (AI) related com
 putations in a fraction of the time it takes a regular CPU to perform the 
 same computations but ill-informed OpenACC compiler pragmas can actually m
 ake a code run slower. Why is this? When should one invest time in paralle
 lization and/or GPU-acceleration? How much speedup can be expected with a 
 given application?\n\nThe main goal of this two-day course is to start an
 swering these questions. The course also covers the basics of GPU progra
 mming and aims to provide the necessary information for avoiding the most
  common pitfalls.\n\nMore information\, schedule\, and registration can be
  found on the course webpage at HPC2N\, Umeå University.\nhttps://events.
 prace-ri.eu/event/1064/
SUMMARY:[ONLINE] Heterogeneous computing with performance modelling @ SNIC
URL;VALUE=URI:https://events.prace-ri.eu/event/1064/
END:VEVENT
END:VCALENDAR