This is a collection of tutorials, blogs, articles and other resources for CUDA C that I hope you'll find useful. Feel free to contribute with a comment what you think can help people learn CUDA and optimise their code. I am planning to keep this post continuously updated, so stay tuned. Most of the links hosted here point to free resources.1. GENERAL LEARNING RESOURCES
1.1. Official resources and APIs
- CUDA C Programming Guide (pdf)
- CUDA Runtime API (pdf)
- CUDA Best Practices Guide (pdf)
- CUDA Driver API
1.2. Blogs
- CUDA Programming
- Parallel for all and a list of posts on disqus
- Solarian programmer
1.3. Slides and courseware
- Official slides
- Qwiklab's CUDA lab (among other interesting tutorials)
- An excellent collection of course material by the university of Illinois
- Yong Cao's (Virginia Tech) presentation of the CUDA programming model
- A very good collection of lectures notes on the website of the university of Standford
- More slides with a very good tutorial on reduction and scan
- Rice university lecture notes
1.4. Books
- CUDA Application Design and Development by Rob Farberg which is an excellent book
- CUDA by Example by Jason Sanders and Edward Kandrot
- The CUDA Handbook by Nicholas Wilt
- More suggestions can be found here by NVIDIA
2. CUDA ACCELERATED LIBRARIES
A collection of general-purpose libraries and links to the official websites:
3. SPECIAL TOPICS
3.1. Thrust
- First steps
- Thurst wiki on github - a good reference
3.2. Matrix multiplication
3.3. Advanced Topics
- Optimisation for your CUDA code: Performance assessment, Memory, Instructions, CUDA and concurrent execution, and more from this web page of the Penn State University, Institute for CyberScience.
- M. Harris, S. Sengupta and J.D. Owens, "Parallel Prefix Sum (Scan) with CUDA," GPU Gems 3, by NVIDIA.
No comments:
Post a Comment