BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T184607Z
UID:91303784-e70b-42fd-a046-0554683f25a0
DTSTART:20190401T070000Z
DTEND:20190403T143000Z
DESCRIPTION:GPU-accelerated computing drives current scientific research. W
 riting fast numeric algorithms for GPUs offers high application performanc
 e by offloading compute-intensive portions of the code to an NVIDIA GPU. T
 he course will cover basic aspects of GPU architectures and programming. F
 ocus is on the usage of the parallel programming language CUDA-C which all
 ows maximum control of NVIDIA GPU hardware. Examples of increasing complex
 ity will be used to demonstrate optimization and tuning of scientific appl
 ications.\n\nTopics covered will include:\n\n\n    Introduction to GPU/Par
 allel computing\n    Programming model CUDA\n    GPU libraries like CuBLAS
  and CuFFT\n    Tools for debugging and profiling\n    Performance optimiz
 ations\n\n\nPrerequisites: Some knowledge about Linux\, e.g. make\, comman
 d line editor\, Linux shell\, experience in C/C++\n\nApplication \nRegistr
 ations are only considered until 28 February 2018 due to available space\,
  the maximal number of participants is limited. Applicants will be notifie
 d\, whether they are accepted for participitation.\n\nInstructors: Dr. Jan
  Meinke\, Jochen Kreutz\, Dr. Andreas Herten\, JSC\; Jiri Kraus\, NVIDIA\n
 \nContact \nFor any questions concerning the course please send an e-mail 
 to j.meinke@fz-juelich.de\nhttps://events.prace-ri.eu/event/823/
SUMMARY:GPU Programming with CUDA @ JSC
URL;VALUE=URI:https://events.prace-ri.eu/event/823/
END:VEVENT
END:VCALENDAR