What is the "best" decompostition method?

Attesz · March 1, 2012, 07:33

Hi,

I'm dealing with an axial fan of 3 regions, and I'm currently using the scotch decomposition method. In your opinion, what is the best decomp. method for such cases (long domain, 2 interface). I'm thinking about the structured method...

olivierG · March 1, 2012, 09:56

hello, just my 2 cents

From some test i made some time ago, ... it depend of your geometry.
- With "simple" one like pipe and / or very elongated in one or two direction, try the simple/ hierarchical method in the corresponding direction.
- with no clear direction, scotch will do a very good job.

regards,
olivier

Attesz · March 1, 2012, 10:09

Thanks Olivier. My geometry is elongated in y direction, but I'll run on 16cores or more. Decomposing the geometry as (2 4 2) could be a good start in your opinion?

olivierG · March 1, 2012, 10:32

No, if your geometry is on the y axis, i would try (1 16 1).
But make a small bench, and you would get your answer ...

regards,
olivier

Attesz · March 1, 2012, 10:34

Quote:

But make a small bench

I've 20millions of cells and transient simulation, so doing a small benchmark is not so easy

but thank you, I'll try it!

best,
Attila

olivierG · March 1, 2012, 10:48

then decompose using different setting, and sum the total of exchange face: the less = the best.

March 1, 2012, 07:33	What is the "best" decompostition method?	#1
Attesz Senior Member Attesz Join Date: Mar 2009 Location: Munich Posts: 368 Rep Power: 17	Hi, I'm dealing with an axial fan of 3 regions, and I'm currently using the scotch decomposition method. In your opinion, what is the best decomp. method for such cases (long domain, 2 interface). I'm thinking about the structured method... __________________ I am doing CFD Consulting Services. Ich biete CFD Strömungssimulationen an.

March 1, 2012, 10:09		#3
Attesz Senior Member Attesz Join Date: Mar 2009 Location: Munich Posts: 368 Rep Power: 17	Thanks Olivier. My geometry is elongated in y direction, but I'll run on 16cores or more. Decomposing the geometry as (2 4 2) could be a good start in your opinion? __________________ I am doing CFD Consulting Services. Ich biete CFD Strömungssimulationen an.

March 1, 2012, 10:48		#6
olivierG Senior Member Olivier Join Date: Jun 2009 Location: France, grenoble Posts: 272 Rep Power: 18	then decompose using different setting, and sum the total of exchange face: the less = the best. Attesz likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Gmsh] discretizer - gmshToFoam	Andyjoe	OpenFOAM Meshing & Mesh Conversion	13	March 14, 2012 05:35
Fluent 6.3.26 vs 12.1 and partition method	Anorky	FLUENT	0	April 27, 2010 11:55
Comparison: Finite Volume Method vs. Analytic Method	m-fry	Main CFD Forum	1	April 20, 2010 15:40
Code for most powerfull FDV Method	D.S.Nasan	Main CFD Forum	6	September 4, 2008 03:08
Help about the preconditioner in gmres method	Dan Gao	Main CFD Forum	3	July 5, 2008 02:18

March 1, 2012, 09:56		#2
olivierG Senior Member Olivier Join Date: Jun 2009 Location: France, grenoble Posts: 272 Rep Power: 18	hello, just my 2 cents From some test i made some time ago, ... it depend of your geometry. - With "simple" one like pipe and / or very elongated in one or two direction, try the simple/ hierarchical method in the corresponding direction. - with no clear direction, scotch will do a very good job. regards, olivier

March 1, 2012, 10:32		#4
olivierG Senior Member Olivier Join Date: Jun 2009 Location: France, grenoble Posts: 272 Rep Power: 18	No, if your geometry is on the y axis, i would try (1 16 1). But make a small bench, and you would get your answer ... regards, olivier