![]() ![]() ![]() I know, that the bottleneck of CUDA is the channel bandwidth between GPU and CPU. the kernel returns to Mathematica a very large array of reals (100 000).Result = 0 (*count for transmitted particles*)įor[j = 0, j can I call this function once for all threads? My approach is writing a kernel for CUDA and execute it as a Mathematica function.Įxtern "C" _global_ void monteCarlo(Real_t *transmission, mint seed, mint pathN), 256, I'm writing a simple monte carlo simulation for particle transport.
0 Comments
Leave a Reply. |