Stephen Jones from Nvidia writes that Dynamic Parallelism in CUDA can speed up a wide range of tasks.
Dynamic Parallelism allows the GPU to operate more autonomously from the CPU by generating new work for itself at run-time, from inside a kernel. The concept is simple, but the impact is powerful: it can make GPU programming easier, particularly for algorithms traditionally considered difficult for GPUs such as divide-and-conquer problems.
Read the Full Story.
Leave your own comment