Home | english  | Impressum | Sitemap | KIT

Praktikum GPU-Computing

Praktikum GPU-Computing
Typ: Praktikum (P)
Lehrstuhl: Fakultät für Informatik
Semester: SS 2014
Zeit: 16.04.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten


23.04.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

30.04.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

07.05.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

14.05.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

21.05.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

28.05.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

04.06.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

11.06.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

18.06.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

25.06.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

02.07.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

09.07.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten

16.07.2014
14:00 - 15:30 wöchentlich
50.34 Raum 148 50.34 INFORMATIK, Kollegiengebäude am Fasanengarten


Dozent:

Gabor Liktor
Hauke Rehfeld
Christoph Schied
Prof.Dr.Ing. Carsten Dachsbacher

SWS: 4
LVNr.: 24909

Exploiting the vast horse power of contemporary GPUs for general purpose applications has become a must for any realtime or interactive application nowadays. Current computer games use the GPUs not only for rendering graphics, but also for collision detection, physics, or artificial intelligence. General purpose computing on GPUs (GPGPU) has also penetrated the field of scientific computing enabling realtime experience of large scale fluid simulations, medical visualization, or signal processing.

This lecture introduces the concepts of programming graphics cards for non-graphical applications, such as data sorting, image filtering (e.g. denoising, sharpening), or physically based simulations. During the course, students familiarize themselves with the architecture of the current GPUs and develop parallel programming skills using the OpenCL and CUDA language.

The practical assignments will cover the following topics:

  • The architecture of contemporary GPUs (execution model, memory model, etc.)
  • Introduction to the OpenCL programming language
  • Reduction, scanning, and sorting parallel algorithms
  • Image filtering via separable convolution kernels
  • Physics simulation (cloth simulation using particles and springs)
  • Introduction to optimization and profiling on the GPU

Course Materials

    Kick-off Slides
    Materials: 
Assignment 1     Introduction
    Deadline:  April 30th, 2014, 14:00
    Topics:  Introduction into OpenCL, parallel vector summation and matrix transposition.
    Materials: 
Assignment 2     Data-Parallel Algorithms
    Deadline:  May 21st, 2014, 14:00
    Topics:  Parallel programming primitives in GPU computing: reduction, scan
    Materials: 
Assignment 3     Image Convolution
    Deadline:  June 11th, 2014, 14:00
    Topics:  Convolution, separable kernels, bilateral filter
    Materials: 
Assignment 4     Particle Simulation
    Deadline:  July 2nd, 2014, 14:00
    Topics:  Particle systems, Verlet-integration, cloth simulation
    Materials: 
Assignment 5     Freestyle
    Specifications due: July 2nd, 2014 (in email)
    Solutions due:  July 23rd, 2014, 14:00
    Topics: You are free to choose this time :)
    Materials: