471,342 Members | 1,867 Online
Bytes | Software Development & Data Engineering Community
Post +

Home Posts Topics Members FAQ

Join Bytes to post your question to a community of 471,342 software developers and data experts.

Scientifi Python+MPI

Hello everyone

The point of my work is to run a python program at a cluster with MPI communication, so I use Python Scientific module 2.4.11 . I have installed the Scientific and the MPI module according to instructions at each node of cluster. When I execute mpi.py at the cluster the following message appears :

nimitsis@centaurus:~> mpipython mpi.py
Traceback (most recent call last):
File "mpi.py", line 11, in ?
communicator.send(data, 1, 0)
Scientific.MPI.core.MPIError: invalid MPI destination

According to core.py file at 299 line which is the following :

def send(self, data, destination, tag):
if destination != 0:
raise MPIError, "invalid MPI destination"
self.messages.append((tag, Numeric.array(data, copy=1).flat))

the above message is right, but it seems that the send function can not send message to any destination except 0. Is that desirable ?

In addition, I have 2 questions :
1. If i want to execute the program at several processors , where do I declare that? for example at C I use for execution at 4 processors the command : mpirun -np=4 "name of executable" , what must I do for that ?
2. Where does happens the initialization of the MPI_world with the appropriate arguments of the processor number ?

I will appreciate very each answer.
Thank you in advance
Sep 11 '07 #1
0 1099

Post your reply

Sign in to post your reply or Sign up for a free account.

Similar topics

reply views Thread by Natsu Mizutani | last post: by
reply views Thread by Barry Rountree | last post: by
reply views Thread by Marc Poinot | last post: by
2 posts views Thread by uli | last post: by
53 posts views Thread by Michael Tobis | last post: by
2 posts views Thread by dfj225 | last post: by
43 posts views Thread by parallelpython | last post: by
7 posts views Thread by Spectrum | last post: by

By using Bytes.com and it's services, you agree to our Privacy Policy and Terms of Use.

To disable or enable advertisements and analytics tracking please visit the manage ads & tracking page.