I have some code that used to work with ipyparallel using a load_balanced_view. I used to be able to:
from ipyparallel import Client
rcAll = Client()
lbvAll = rcAll.load_balanced_view()
for anInpt in allInpt:
lbvAll.apply(doAll, anInpt)
lbvAll.wait()
lbvAll.get_result()
and then lbvAll.results.values() would be a list of the results
however, now lbvAll.apply() does not work for me
I can do result = lbvAll.map_sync(doAll, allInpt) and result is returned as a list of results.
Using 2-4 cores/engines, there is not much improvement using 4 cores over 2 cores.
My feeling is that ipyparallel has changed but I am not sure and I do not seem to using it correctly. Thanks in advance for any help.
I am using Python 3.4.5, IPython3 5.1.0 and ipyparallel 5.2.0 on linux