a national resource for computational science education

HOME BWPEP Shodor Blue Waters

Introduction to GPU programming using CUDA

By Paul Gray
University of Northern Iowa, Cedar Falls, IA

This module provides:

  • An introduction to the GPU architecture
  • An introduction to the CUDA development environment
  • Instructions on interfacing with the GPU hardware
  • An emphasis on debugging C and CUDA codes with cuda-gdb

Upon completion of this module students should be able to:

  • Develop, compile, and execute programs on the GPU architecture
  • Troubleshoot issues with GPU codes using cuda-gdb
  • Translate basic algorithms into programs suitable for execution on the GPU architecture
  • Understand the many-thread model for parallelism present on the GPU architecture

Instructors wishing to incorporate this module into their curriculum have a few options:

  • They can have a clone of the course all to themselves, hosted on the current moodle site, with class accounts and formal deadlines for a semester-long course
  • They can use the content housed at UNI as a skeleton for their own moodle site. A copy of the moodle content housed at UNI would be provided upon request which would be suitable for a moodle course import.

This module is set up as a Moodle course. To access the Moodle course, please click here.