Hi Claire, You might want to have a look at the function called mpi_bcast in yt/utilities/parallel_tools/parallel_analysis_interface.py. This should do a standard mpi broadcast from the root process to all of the subprocesses. Alternatively, if the dictionary is something that can be loaded in in parts by each of the processors and then combined, you might also want to check par_combine_object in the same file. Britton On Mon, May 5, 2014 at 3:56 PM, Claire Kopenhafer <clairekope@gmail.com>wrote:

Hello all,

I've got a code that reads in a dictionary from file and then operates on it, and I'd like to run it in parallel. My best thought is to have the root process read in the dict and then broadcast it to all the rest, but I am not sure how best to do this within yt. Any suggestions? Thanks!

-Claire

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org