This operation performs a tree-wise data reduction operation (here:
bit-wise or) on all participating processes and distributes the result
to all participating nodes. This result distribution to all
paricipating nodes is the difference to the normal MPI_Reduce
operation, where the result is stored in a single root processor. So
it is interesting to compare this operation to the normal MPI_Reduce and to a MPI_Reduce follwed by an MPI_Bcast
operation (our measurement MPI_Reduce_Bcast), which would also
distribute the result to all nodes. We vary over the number of nodes
with a message length of 256 Bytes for each node.
Pattern: Collective varied over number of nodes.
default values: 8 nodes, message length 256 units, max. / act. time for suite disabled/0.00 min.
