The release of CUDA has triggered a big wave in parallel computing. More and more people are getting involved in porting various applications to CUDA. However, the application field is still limited, mainly due to the lack of libraries in CUDA. Currently, the libraries for CUDA kernel code mainly focus on scientific computation. Though stdcuda provides some STL containers, the library can only be used in host code, not kernel code.
cudaSTL is an effort to port STL to CUDA kernel code, which provides STL-like interface to CUDA kernel code. We plan to port most of the containers and algorithms into CUDA. Meanwhile, we will strive to make the implementation as fast as possible.
Using cudaSTL, you will be able to use STL containers and algorithms in CUDA kernel code. We believe that this will be a major step towards real "general purpose" GPU.
If you have any comment or suggestions, send me a mail at: Chuntao.HONG@gmail.com.
Apr. 22, 2009: HashMap ver0.1 is available
30 Day Summary not available
12 Month Summary not available
Copyright © 2013 Black Duck Software, Inc. and its contributors, Some Rights Reserved. Unless otherwise marked, this work is licensed under a Creative Commons Attribution 3.0 Unported License . Ohloh ® and the Ohloh logo are trademarks of Black Duck Software, Inc. in the United States and/or other jurisdictions. All other trademarks are the property of their respective holders.