Implementation of heavy computing tasks on GPU's faces two limits:
1) number of parameters/variables in each kernel;
2) computation time.
This very simple C++ header allows to easily create CUDA kernels that overcome both limitations. The library adds a small overhead to the GPU kernels, but in carefully designed applications the slowdown should hardly be noticeable.
License
Creative Commons Attribution Non-Commercial License V2.0Follow GPU Interruptible Kernels
You Might Also Like
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of GPU Interruptible Kernels!