BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T123739Z
UID:3adf8e16-ef98-4184-83f9-79a8562a7213
DTSTART:20181126T090000Z
DTEND:20181127T173000Z
DESCRIPTION:Graphics Processing Units (GPUs) were originally developed for 
 computer gaming and other graphical tasks\, but for many years have been e
 xploited for general purpose computing in a number of areas. They offer ad
 vantages over traditional CPUs because they have greater computational cap
 ability\, and use high-bandwidth memory systems (memory bandwidth is the m
 ain bottleneck for many scientific applications).\n\n \n\nThis introducto
 ry course will describe GPUs\, and the advantages they offer.\n\nIt will t
 each participants how to start to program GPUs\, which cannot be used in i
 solation\, but are usually used in conjunction with CPUs.\n\nImportant iss
 ues affecting performance will be covered.\n\nThe course focuses on NVIDIA
  GPUs\, and the CUDA programming language (an extension to C/C++ or Fortra
 n). Please note the course is aimed at application programmers\; it does n
 ot consider machine learning or any of the packages available in the machi
 ne learning arena.\n\nHands-on practical sessions are included.\n\nYou wil
 l require your laptop\, and your institutional credentials to connect to e
 duroam.  The training parctical exercises will be run on a web-based syst
 em so all you will need is a relatively recent web browser (Firefox\, Chro
 me and Safari are known to work).\n\nTimetable\n\nDay 1\n\n10:00 Introduct
 ion\n	10:20 GPU Concepts/Architectures\n	11:00 Break\n	11:20 CUDA Programm
 ing\n	12:00 A first CUDA exercise\n	13:00 Lunch\n	14:00 CUDA Optimisations
 \n	14:20 Optimisation Exercise\n	15:00 Break\n	15:20 Constant and Shared M
 emory\n	16:00 Guest Lecture Alan Gray (NVIDiA) Overview of NVIDIA Volta\n	
 17:00 Close\nDay 2\n\n10:00 Constant and Shared Memory\n	10:10 Exercise\n	
 10:30 OpenCL and Directives\n	11:00 Break\n	11:30 OpenCL and / or Directiv
 es Exercises\n	13:00 Lunch\n	14:00 Performance portability and Kokkos\n	14
 :30 Exercise: Getting started with Kokkos patterns\n	15:00 Break\n	15:10 K
 okkos memory management\n	15:30 Memory management exercises\n	16:00 Close\
 n \n\nCourse Materials\n\nCourse materials page\n\n \n\nhttps://epcced.g
 ithub.io/archer-gpu-course/\n\nhttps://github.com/EPCCed/archer-gpu-course
 \n\n \n\nhttps://events.prace-ri.eu/event/784/
SUMMARY:GPU  Programming with CUDA @ EPCC at Imperial College London
URL;VALUE=URI:https://events.prace-ri.eu/event/784/
END:VEVENT
END:VCALENDAR