[aspect-devel] Direct Solver in Aspect

[Wolfgang Bangerth]

>> I have no idea how difficult it would be to support any of this (I don't
>> even recall who wrote the current interfaces...)
>
> 1. configure Trilinos with it
> 2. change one line in SolverDirect. :-)

Interesting. Maybe we should change the installation instructions for Trilinos 
to make this possible in the future.


>> Yes, that occurred to me as well after I sent the last email. We currently
>> drive the MPI_InitFinalize class with only a single thread, but maybe we
>> should be using
>>    n_cores/n_mpi_jobs_on_this_host
>> instead? This would give us exactly the behavior you describe, while
>> reverting back to the current state on machines that we fully load with MPI
>> processes...
>
> Yes. I was going to try things out first though. Maybe this could be a
> runtime parameter...

That's too late -- we need to know at the time we call the MPI_InitFinalize 
constructor. At least I haven't found a way yet to change the number of 
threads TBB creates at a later time...

Best
  W.


-- 
------------------------------------------------------------------------
Wolfgang Bangerth               email:            bangerth at math.tamu.edu
                                 www: http://www.math.tamu.edu/~bangerth/