Open jjdredd opened 2 years ago
Hi @jjdredd ,
Currently the simulator supports computations on a CPU (numpy backend) or computations on a GPU (torch backend). However, further parallelization (for example using MPI) is not supported.
That said, it should be fairly trivial to create a dask backend (see backend.py). I would happily accept a PR for that :)
Best, Floris
Thanks. I'll take a look. didn't know about dask.
Hello, Thanks for your work. Is there a way to run this code in parallel on cpu? I assume that it's automatically parallelized if the cuda backend is used. Is this correct?