|
[Sponsors] |
February 21, 2014, 03:59 |
MPI Communication within Linear solver
|
#1 |
Member
|
Hi,
I would like to know if there are MPI communications within a BCGSTAB iteration ? Seeing through the code, I don't seem to notice that. I can see that all partitions do individual solves and then communicate. For e.g. in solver_direct_mean.cpp void CEulerSolver::ImplicitEuler_Iteration(), the Set_MPI_Solution is called at the end, after the solution is computed. May be your convergence would improve if the communication is placed inside the BCGSTAB routine as well ? but many increase your total communication time. - Is this why it was done this way ? Dominic |
|
March 2, 2014, 19:09 |
|
#2 | |
Super Moderator
Francisco Palacios
Join Date: Jan 2013
Location: Long Beach, CA
Posts: 404
Rep Power: 15 |
Quote:
For example the mat_vec is done in void CSysMatrix::MatrixVectorProduct(const CSysVector & vec, CSysVector & prod, CGeometry *geometry, CConfig *config) { Cheers, Francisco |
||
March 2, 2014, 23:05 |
|
#3 |
Member
|
Thank you Francisco, I missed that part ! I can see that there is a
SendReceive_Solution() inside the function you've mentioned. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Linear Solver for Particle Methods | prapanj | Main CFD Forum | 1 | December 2, 2013 20:03 |
please help me which solver is better for my application | Ger_US | OpenFOAM | 8 | January 8, 2013 12:29 |
MPI issues with my own solver | davide_c | OpenFOAM Running, Solving & CFD | 1 | March 23, 2012 08:57 |
solution diverges when linear upwind interpolation scheme is used | subash | OpenFOAM | 0 | May 29, 2010 01:23 |
Getting too many iterations by velocity solving (aborting). Changing U - Solver? | suitup | OpenFOAM Running, Solving & CFD | 0 | January 20, 2010 07:45 |