Archives


- Beowulf
- Beowulf Announce
- Scyld-users
- Beowulf on Debian

[Beowulf] MPI - time for packing, unpacking, creating a message...

Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.

Search

tribur at vision.ee.ethz.ch tribur at vision.ee.ethz.ch
Mon May 18 07:49:04 PDT 2009


Hi all,

is there anyone who can tell me if A) or B) is probably faster?

A)
process 0 sends 3x500 elements, e.g. doubles, to 3 different  
processors using something like
if(rank==0){
MPI_Send(sendbuf, 500, MPI_DOUBLE, 1, 1, MPI_COMM_WORLD);
MPI_Send(sendbuf, 500, MPI_DOUBLE, 2, 2, MPI_COMM_WORLD);
MPI_Send(sendbuf, 500, MPI_DOUBLE, 3, 3, MPI_COMM_WORLD);
}
else
MPI_Recv(recvbuf, 500, MPI_DOUBLE, 0, rank, MPI_COMM_WORLD, status);


B)
process 0 sends 2000 elements to process 1 using
if(rank==0)
MPI_Send(sendbuf, 2000, MPI_DOUBLE, 1, 1, MPI_COMM_WORLD);
else
MPI_Recv(recvbuf, 2000, MPI_DOUBLE, 0, rank, MPI_COMM_WORLD, status);





More information about the Beowulf mailing list