SU2 parallel run results visualization

ddvdc · February 12, 2019, 16:48

I ran the Caradonna and Tung rotor test case included with the source code using the command

Code:

mpiexec -np 16 SU2_CFD rot_caradonna_tung.cfg

, which completed successfully. Then I tried to post-process the restart file using

Code:

SU2_SOL rot_caradonna_tung.cfg

First, this gave me the following error:

Code:

Unable to open SU2 restart file solution_flow.dat

I managed to get around this by changing line 209 of the .cfg file to point to the restart_flow.dat file instead of solution_flow.dat like this:

Code:

SOLUTION_FLOW_FILENAME= restart_flow.dat

Is that what I am supposed to do?

Regardless, after the changes the SU2_SOL command ran successfully and generated the desired surface_flow.vtk Paraview file for me. However, it seems something went wrong, because this is what the rotor geometry looks like when I opened the file in Paraview:
https://ibb.co/nzLHGyz

Any help figuring this out would be greatly appreciated.

ddvdc · February 12, 2019, 17:14

It turned out that I was trying to post-process the solution using the sequential version of SU2_SOL, whereas I should have used the parallel version instead, because I ran SU2_CFD in parallel too.

February 12, 2019, 16:48	SU2 parallel run results visualization	#1
ddvdc New Member D. de Vries Join Date: Feb 2019 Posts: 3 Rep Power: 7	I ran the Caradonna and Tung rotor test case included with the source code using the command Code: mpiexec -np 16 SU2_CFD rot_caradonna_tung.cfg , which completed successfully. Then I tried to post-process the restart file using Code: SU2_SOL rot_caradonna_tung.cfg First, this gave me the following error: Code: Unable to open SU2 restart file solution_flow.dat I managed to get around this by changing line 209 of the .cfg file to point to the restart_flow.dat file instead of solution_flow.dat like this: Code: SOLUTION_FLOW_FILENAME= restart_flow.dat Is that what I am supposed to do? Regardless, after the changes the SU2_SOL command ran successfully and generated the desired surface_flow.vtk Paraview file for me. However, it seems something went wrong, because this is what the rotor geometry looks like when I opened the file in Paraview: https://ibb.co/nzLHGyz Any help figuring this out would be greatly appreciated.

February 12, 2019, 17:14	Solution found	#2
ddvdc New Member D. de Vries Join Date: Feb 2019 Posts: 3 Rep Power: 7	It turned out that I was trying to post-process the solution using the sequential version of SU2_SOL, whereas I should have used the parallel version instead, because I ran SU2_CFD in parallel too.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Slightly different results between parallel and serial run. Which results to trust?	h0rst	FLUENT	10	April 12, 2018 02:41
Ahmed body simulation gives unexpected results in su2 6.0	anas651	SU2	0	March 28, 2018 04:42
User fortran to input/output file in a parallel run	doublestrong	CFX	5	March 31, 2017 09:15
[mesh manipulation] Cannot get refineMesh to run in parallel	smschnob	OpenFOAM Meshing & Mesh Conversion	2	June 3, 2014 12:20
Transient Run - Output "Time" in partial results?	evcelica	CFX	2	May 16, 2012 22:36