Note that with the op=MPI.SUM option, we apply the sum operation to all the elements of the column array. To better understand how the reduction operates, let's look at the following diagram:

The sending operation is as follows:

• The P0 process sends the [0 1 2] data array.
• The P1 process sends the [0 2 4] data array.
• The P2 process sends the [0 3 6] data array.

The reduction operation sums the i^th elements of each task and then puts the result in the i^th element of the array in the P0 root process. For the receiving operation, the P0 process receives the [0 6 12] data array.

Some of the reduction operations defined by MPI are as follows:

• MPI.MAX: This returns the maximum element.
• MPI.MIN: This returns the minimum element.
• MPI.SUM: This sums up the elements.
• MPI.PROD: This multiplies all elements.
• MPI.LAND: This performs the AND logical operation across the elements.
• MPI.MAXLOC: This returns the maximum value and the rank of the process that owns it.
• MPI.MINLOC: This returns the minimum value and the rank of the process that owns it.