User`s guide

Intel® MPI Library User’s Guide for Windows* OS
10
Suppose that the geometry is <#ranks> = 4 and <#nodes> = 2, where adjacent pairs of
ranks are assigned to each node (for example, for two-way SMP nodes). To see the
cluster nodes, enter the command:
type machines.Windows
The results should look as follows:
clusternode1
clusternode2
To equally distribute four processes of the application on two-way SMP clusters, enter
the following command:
mpiexec -n 2 -host clusternode1 .\testf : -n 2 -host clusternode2 .\testf
The output for the testf executable file may look as follows:
Hello world: rank 0 of 4 running on clusternode1
Hello world: rank 1 of 4 running on clusternode1
Hello world: rank 2 of 4 running on clusternode2
Hello world: rank 3 of 4 running on clusternode2
In general, if a cluster has i nodes and each node is a j-way SMP system, the mpiexec
command-line syntax for distributing the i*j processes amongst the i*j processors
within the cluster is:
mpiexec -n j -host <nodename-1> .\mpi_example : \
-n j -host <nodename-2> .\mpi_example : \
-n j -host <nodename-3> .\mpi_example : \
...
-n j -host <nodename-i> .\mpi_example
NOTE: You have to fill in appropriate host names from <nodename-1> through
<nodename-i> with respect to your cluster system.
See Also
Local Options topic of the Intel® MPI Library Reference Manual for Windows OS.
You can get more information about controlling MPI process placement at Controlling Process
Placement with the Intel® MPI Library.