CUDA: reduction or atomic operations?

Question

I\'m writing a CUDA kernel which involves calculating the maximum value on a given matrix and I\'m evaluating possibilities. The best way I could find is:

Forcing ev

Answer 1

If you have K20 or Titan, I suggest dynamic parallelism: lunching a single thread kernel, which lunches #items worker kernel threads to produce data, then lunches #items/first-round-reduction-factor threads for first round reduction, and keep lunching till result coming out.