In the memory-distributed (MPI) running modes, the data input matrix is assumed to be distributed in between the MPI processes (nodes, processors, etc). The library routines allow for essentially any distribution of the data with a single constraint that a number of data points assigned to any process taking part in the calculation is not smaller than the half band-width of the Toeplitz matrix.

In all the cases, the layout of the output coincides with that of the input.