multigpu

Accelerated Computing CUDA CUDA Programming and Performance
1

suppose there are 2 device;

while(someCond())

{

      for (dev = 0; dev < 2; dev++)

      {

          cudaSetDevice(dev); 

          launch_kernel<<<grid, block, 0, stream[dev]>>>();

      }

      cudeDeviceSyncronize();

      someCommunicationBtwDevice0andDevice1(); //after every iteration i have to do this communication  

}

now the question is at every iteration i have to call “cudaSetDevice()” its a very slow takes around 320ms. is there a way, so i need not call “cudaSetDevice()” at every iteration.

2

I assume this is CUDA 4.0?

3

ya this is cuda 4.0…