BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260621T021326Z
UID:8205016d-fa00-4b20-91f0-3d8143afd743
DTSTART:20121012T071000Z
DTEND:20121012T163000Z
DESCRIPTION:This course give an overview of MPI+OpenMP hybrid programming t
 echniques for HPC applications. OpenMP directives and a distributed memory
  programming paradigm can be mixed to exploit at best on a modern HPC clus
 ter. The course aims at providing the student with the fundamental techniq
 ues for developing an hybrid code both for C/C++ and Fortran programming l
 anguages\, giving also an introduction to the architecture of NUMA machine
 s. The course summarize the performance tuning techniques\, by also introd
 ucing the most common performance bottlenecks of an hybrid code.Topics \n\
 n		Hybrid models for HPC\n	\n		Memory usage\n	\n		Thread safety\n	\n		Fals
 e sharing\n	\n		Mixing modes\n	\n		NUMA machines\n	\n		Thread affinity\nLe
 arning outcomes\nAbility to develop a simple MPI+OpenMP code. Understandin
 g on NUMA architectures and their connection with the hybrid paradigm. Und
 erstanding on the common problems with hybrid codes.Prerequisites\nBasic k
 nowledge of MPI and OpenMP.Price: free of charge\n\nhttps://events.prace-r
 i.eu/event/57/
SUMMARY:Introduction to hybrid programming MPI+OpenMP @ CINECA
URL;VALUE=URI:https://events.prace-ri.eu/event/57/
END:VEVENT
END:VCALENDAR