Obviously, there is not an easy way to do that..... The code needs to be implemented for multiGPU using OpenMP or MPI that communicate the CPU cores that host the different GPUs.... Versions 3 and v4.0 (this March) only include singleGPU. A multiGPU version of the code will be release in v5.0
Comments
Versions 3 and v4.0 (this March) only include singleGPU. A multiGPU version of the code will be release in v5.0
Regards