On Fri, 20 Oct 2017, Sven Schreiber wrote:
Am 20.10.2017 um 02:58 schrieb Allin Cottrell:
> By default gretl expects to use OpenMP on a multi-core machine. You may
> find that it helps to limit the number of threads spawned by each instance
> of gretl, as in
Do I remember correctly that the parallel-computing approach in gretl is
roughly as follows: (?)
- for some built-in functionality gretl uses OpenMP "automagically"
- for hansl scripting and thus user-designed parallelism it's MPI
Yes, that's right -- with this addition: if gretl is linked against
OpenBLAS (or another threaded BLAS/LAPACK library) then there's a
further layer of automatic parallelism (either OpenMP or plain
pthreads) associated with that library. (That is, beyond any explicit
use of OpenMP within libgretl itself.)
One other point: our experimental implementation of machine learning
via libsvm involves "automatic" use of MPI if available.
Allin