BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260629T141843Z
UID:de6efa7e-cfcd-404b-ba44-4aa5c7898bdc
DTSTART:20140121T070000Z
DTEND:20140123T143000Z
DESCRIPTION:Course description\nAccelerators such as GPUs have broken throu
 gh in today’s HPC architectures. This course covers the more advanced to
 pics in GPU programming with the CUDA language and the new OpenACC standar
 d for directive-based accelerator programming. The course will focus on im
 proving the performance and the scalability of GPU codes by analyzing them
  with performance tools\, identifying bottlenecks\, and making algorithm a
 djustments.Schedule\nTuesday\, January 21\n				09:00 – 09:20\n			\n				C
 ourse introduction\n		\n				09:20 - 10:00\n			\n				CUDA recap I\n		\n				
 10:00 - 10:15\n			\n				Coffee/tea break\n		\n				10:15 – 10:45\n			\n		
 		CUDA recap II\n		\n				10:45 – 11:15\n			\n				Introduction to CSC clu
 ster\n		\n				11:15 – 12:00\n			\n				Cluster hands-on\n		\n				12:00 
 – 13:00\n			\n				Lunch break\n		\n				13:00 – 14:30\n			\n				Overvie
 w of CUDA on Kepler\n		\n				14:30 – 14:45\n			\n				Coffee/tea break\n	
 	\n				14:45 – 15:30\n			\n				Tools and libraries\n		\n				15:30 – 16
 :00\n			\n				CUDA 6 overview\n		\n				16:00 – 16:30\n			\n				Debugging
 \n		\n				16:30 – 17:00\n			\n				Debugging hands-on\n		\n				18:00 -\n	
 		\n				Dinner sponsored by NVIDIA (Restaurant Cantina\, Otakaari 24\, Esp
 oo\n		Wednesday\, January 22\n				09:00 – 09:45\n			\n				Profiling over
 view\n		\n				09:45 – 10:15\n			\n				Profiling hands-on\n		\n				10:15 
 – 10:30\n			\n				Coffee/tea break\n		\n				10:30 – 11:30\n			\n				Op
 timization 1\n		\n				11:30 – 12:00\n			\n				Optimization hands-on\n		\
 n				12:00 – 13:00\n			\n				Lunch break\n		\n				13:00 – 14:30\n			\n
 				Optimization 2\n		\n				14:30 – 14:45\n			\n				Coffee/tea break\n		
 \n				14:45 – 15:45\n			\n				Optimization hands-on\n		\n				15:45 – 1
 6:15\n			\n				CUDA aware MPI\n		\n				16:15 – 17:00\n			\n				Optimizat
 ion hands-on\n		Thursday\, January 23\n				09:00 – 10:15\n			\n				Intro
 duction to OpenACC\n		\n				10:15 – 10:30\n			\n				Coffee/tea break\n		
 \n				10:30 – 11:30\n			\n				Tuning OpenACC programs\n		\n				11:30 –
  12:00\n			\n				Exercise session\n		\n				12:00 – 13:00\n			\n				Lunch
  break\n		\n				13:00 – 14:15\n			\n				Exercise session continues\n		\n
 				14:15 – 14:30\n			\n				Coffee/tea break\n		\n				14:30 – 15:30\n	
 		\n				Advanced OpenACC topics\n		\n				15:30 – 16:15\n			\n				Exercis
 e session\n		\n				16:15 – 16:30\n			\n				Course wrap-up\n		Learning ou
 tcome\nUnderstanding the GPU execution model and best practices in multi-G
 PU programming. Ability to parallelize and write scientific applications o
 n GPUs.Pre-requisites\nC/C++\, Fortran\, MPI\, working knowledge in CUDAPr
 ice: Free of chargeLecturers: Sami Ilvonen (CSC)\, Sami Saarinen (CSC)\,
  Peter Messmer (NVIDIA) and Axel Koehler (NVIDIA)\n\nhttps://events.prace
 -ri.eu/event/216/
SUMMARY:Advanced CUDA & OpenACC @ CSC
URL;VALUE=URI:https://events.prace-ri.eu/event/216/
END:VEVENT
END:VCALENDAR