Numerical spurious modes at processors interface in parallel run

yeharav · April 21, 2015, 04:35

Hi all,

I simulate s cyclic 2D domain using a solver I wrote (based on buoyantBoussinesqPimpleFoam) - that solver was tested on analytical results.

The parallel simulation is divided on 32 processors (8 horizontal, 4 vertical).

As you can see (first picture), I get some spurious modes between the processors interfaces.

However, If I take this timestep as initial conditions for a single processor,
the spurious modes disappear (very quickly), as can be seen in the second picture.

Any idea? Did somebody see this before? Is it related to the discretization schemes?
the linear solver?

hjasak · April 22, 2015, 03:35

Did you reconstruct before post-processing or are you using Paraview native reader. Paraview is not aware of parallel decomposition, so it does not do the correct interpolation to points at processor boundaries.

Hrv

yeharav · April 22, 2015, 04:06

Hi,
Thank you for your reply.

The result's you see are after the reconstruction.

However, I tested the same case with the OF solver it is OK.
So presently I assume that it has to do with my solver.

I will update further developments

April 22, 2015, 03:35		#2
hjasak Senior Member Hrvoje Jasak Join Date: Mar 2009 Location: London, England Posts: 1,905 Rep Power: 33	Did you reconstruct before post-processing or are you using Paraview native reader. Paraview is not aware of parallel decomposition, so it does not do the correct interpolation to points at processor boundaries. Hrv __________________ Hrvoje Jasak Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Error running simpleFoam in parallel	Yuby	OpenFOAM Running, Solving & CFD	14	October 7, 2021 04:38
Can not run OpenFOAM in parallel in clusters, help!	ripperjack	OpenFOAM Running, Solving & CFD	5	May 6, 2014 15:25
The results difference between parallel and serial run.	Hkp	OpenFOAM Running, Solving & CFD	2	April 17, 2014 02:26
parallel Grief: BoundaryFields ok in single CPU but NOT in Parallel	JR22	OpenFOAM Running, Solving & CFD	2	April 19, 2013 16:49
Replace periodic by inlet-outlet pair	lego	CFX	3	November 5, 2002 20:09

April 22, 2015, 04:06		#3
yeharav Member Join Date: Feb 2014 Posts: 32 Rep Power: 12	Hi, Thank you for your reply. The result's you see are after the reconstruction. However, I tested the same case with the OF solver it is OK. So presently I assume that it has to do with my solver. I will update further developments