PPAM 2013 GPU Tutorial

Scientific Computing on GPUs

Welcome to the course notes for the full-day  PPAM 2013 GPU Computing Tutorial!

The tutorial was held at PPAM 2013 in Warsaw, Poland, on Sunday, September 8, 2013.

Course Organizers

Course Material

  1. Introduction [PDF]
  2. GPU libraries  [PDF]
  3. OpenACC  [PDF]
  4. GPU hardware [PDF]
  5. CUDA primer [PDF]
  6. Memory performance optimisations [PDF]
  7. Host-device performance optimisations and concurrency [PDF]
  8. Shared memory and instruction optimisations, dynamic parallelism [PDF]
  9. The future of CUDA [PDF]
  10. Demo codes used throughout the tutorial [ZIP]