CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Pre-Processing (https://www.cfd-online.com/Forums/openfoam-pre-processing/)
-   -   Decomposition of a case with multiple cyclic patches (https://www.cfd-online.com/Forums/openfoam-pre-processing/97254-decomposition-case-multiple-cyclic-patches.html)

mrs206 February 13, 2012 12:19

Decomposition of a case with multiple cyclic patches
 
1 Attachment(s)
Dear All,

I have a problem using OF 2.0.1 when i try to decompose a case with multiple cyclic patches using scotch. I have included the preservePatches option in the decomposeParDict to ensure that owner and neighbour patches stay on the same processor. In my case this has the undesired effect of keeping a large part of my mesh on the same processor. This is the output from running decomposePar:


/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 2.0.1 |
| \\ / A nd | Web: www.OpenFOAM.com |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : 2.0.1
Exec : decomposePar
Date : Feb 09 2012
Time : 17:20:28
Host : blue32
PID : 29828
Case : /temp/mrs206/OFLAM2
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Time = 0
Create mesh

Calculating distribution of cells
Keeping owner of faces in patches
44
(
top1-8
leftwheels1back-9
leftstrut1back-10
frontwheelsback-11
frontstrutback-12
leftwheels1-13
frontwheels-14
top-wall-15
frontstrut-16
leftstrut1-17
backwheels1back-18
backstrut1back-19
leftwheels2back-20
leftwheels2-21
leftstrut2back-22
leftstrut2-23
backwheels1-24
top2-25
backstrut1-26
rightwheels1back-28
rightstrut1back-29
rightwheels1-30
rightstrut1-31
rightwheels2back-32
rightwheels2-33
rightstrut2back-34
rightstrut2-35
backwheels2back-38
backstrut2back-39
leftwheels3back-40
leftwheels3-41
leftstrut3back-42
leftstrut3-43
backwheels2-44
top3-45
backstrut2-46
rightwheels3back-47
rightwheels3-48
rightstrut3back-49
rightstrut3-50
backwheels3back-51
backstrut3back-52
backwheels3-53
backstrut3-54
)
on same processor. This only makes sense for cyclics.
Selecting decompositionMethod scotch
Selected 652584 faces whose owner and neighbour cell should be kept on the same processor


I have been able to run the case in serial without any problems. The purpose of all of these cyclics is to create an invisible or completely porous surface on which i can collect data around my geometry (p, U) for aeroacoustic analysis during postprocessing.

I was wondering if there is a way that i can specify each cyclic patch pair to be kept on the same processor, without having to keep everyone of my cyclic patches on just one processor. Or whether these is a better way of sampling on an internal surface around my geometry.

Many thanks in advance,

Michael

Mirage October 7, 2016 14:02

Dear All,

I am running a case with multiple cyclic BC parallel and I 'd would like to know, if each cyclic patch pair must be in the same processor.

I appreciate any help :)


All times are GMT -4. The time now is 10:46.