Refering to the iparm (0-indexed) parameter: https://software.intel.com/en-us/mkl-developer-reference-c-pardiso-iparm-parameter#IPARM24
Is it really the case that the fwd/backward steps are not parallel as default (=0) ?
Here is the correct version of description:
0* -- Intel ® MKL PARDISO uses the following strategy for parallelizing the solving step: In the case of the one right-hand side, the parallelization will be performed by partitioning the matrix. Otherwise, the parallelization will be over the right-hand sides.
1 -- Intel ® MKL PARDISO uses the sequential forward and backward solve.
2 -- Independent from the number of the right-hand sides, Intel ® MKL PARDISO uses the parallel algorithm based on the matrix partitioning.
This feature is available only for in-core Intel MKL PARDISO (see iparm).