I have been running the parallel version of the MKL with Pardiso as a solver in finite element applications. I have run on both dual and quad core intel machines. It seems that the performance gain in parallel mode is not muchfaster than sequential mode. I was just doing some reading in the Pardiso manual (university of Basel) and read the following,
(o) Reproducibility of exact numerical results staon multi-core architectures. The solver is now able to compute the exact bit identical solution independent on the number of cores without effecting the scalability.
Here are some results for a nonlinear FE model with 500'000 elements.
Intel MKL PARDISO 10.2
1 core - factor: 17.980 sec., solve: 1.13 sec.
2 cores - factor: 9.790 sec., solve: 1.13 sec.
4 cores - factor: 6.120 sec., solve: 1.05 sec.
8 cores - factor: 3.830 sec., solve: 1.05 sec.
U Basel PARDISO 4.0.0:
1 core - factor: 16.820 sec., solve: 1.09 sec.
2 cores - factor: 9.021 sec., solve: 0.67 sec.
4 cores - factor: 5.186 sec., solve: 0.53 sec.
8 cores - factor: 3.170 sec., solve: 0.43 sec.
This method is currently only working for symmetric indefinite matrices.
This seems to be consistent with what I am experiencing using the MKL 10.2. Do we get updated versions of Pardiso from Basel ?