BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260820T192507Z
UID:2d9ba583-d6eb-48c3-a965-5884bbb9e30f
DTSTART:20170425T070000Z
DTEND:20170427T150000Z
DESCRIPTION:The aim of this course is to give users the best practices to g
 o further using CURIE system and to give hints to prepare their codes for 
 future architectures.\n\n \n\nTopics covered: \n\n architecture (hardwa
 re\, interconnect\, file-system)\,\n	 code optimisation (vectorisation\, 
 profiling)\,\n	 parallelisation (MPI\, openMP\, mixte)\,\n	 IO improveme
 nt.\n \n\nPrerequisites\n\ncode developpers\, knowledge of C or F90\, MPI
 \, OpenMP\n\n \n\nDay 1\, tuesday 25\n\nIntroduction\, Allinea tools\n\nC
 urrent &amp\; future CPU architecture\n\nVectorisation &amp\; threads\, Op
 enMP4\n\n \n\nDay 2\, wednesday 26\n\nIntra nodes\, profiling &amp\; tuni
 ng\n\nWI4MPI\n\nAdvanced MPI\, point to point\, MPI &amp\; threads\, tunin
 g librairies\n\n \n\nDay 3\, thursday 27\n\nAdvanced MPI\, communicators\
 , collectives\, non blocking collectives\,\n\nMPI3\n\nAdvanced IO\, Lustre
 \, meta data\, Posix\, MPI-IO\, HDF5\n\nhttps://events.prace-ri.eu/event/5
 79/
SUMMARY:Advanced usage on Curie supercomputer : Best practice for current a
 nd future HPC architectures @TGCC
URL;VALUE=URI:https://events.prace-ri.eu/event/579/
END:VEVENT
END:VCALENDAR