Sorry for write again, It's solved now. I found it's really a load balance problem. Now the computation is more balanced with utilizing a random sequence function.
On Tue, Jul 23, 2013 at 12:09 AM, Kai Li <kai.li...@gmail.com> wrote: > Hello, > > Does anyone has the experience of using shapely with mpi4py? > I just used intersects() function with mpi4py. Although I assigned points > to each process and broadcast the polygon to be passed as parameter of > intersects(), the program only take advantage of one single cpu (rank 1). > It's weired that function like numpy.lib.bench() can take advantage of > multiple cpu with the same program. > > Regards, > > -- > Kai Li > -- Kai Li
_______________________________________________ Community mailing list Community@lists.gispython.org http://lists.gispython.org/mailman/listinfo/community
