This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
howto:su2 [2019/05/27 12:33] ccrosby [Sample PBS script] |
howto:su2 [2022/11/02 16:10] (current) ccrosby [Sample PBS script] |
||
|---|---|---|---|
| Line 15: | Line 15: | ||
| #PBS -q normal | #PBS -q normal | ||
| #PBS -l walltime=01: | #PBS -l walltime=01: | ||
| - | #PBS -o /home/ | + | #PBS -o /mnt/lustre/ |
| - | #PBS -e /home/ | + | #PBS -e /mnt/lustre/ |
| #PBS -m abe | #PBS -m abe | ||
| #PBS -M user@wherever.co.za | #PBS -M user@wherever.co.za | ||
| Line 22: | Line 22: | ||
| module add chpc/ | module add chpc/ | ||
| # Set the working directory environment variable explicitly. | # Set the working directory environment variable explicitly. | ||
| - | export PBS_JOBDIR=/ | + | export PBS_JOBDIR=/ |
| # Change to the job directory | # Change to the job directory | ||
| cd $PBS_JOBDIR | cd $PBS_JOBDIR | ||
| Line 28: | Line 28: | ||
| ### Run the SU2_CFD solver | ### Run the SU2_CFD solver | ||
| ### For the gcc/mpich and PGI/mpich versions, use mpirun -iface ib0 .... | ### For the gcc/mpich and PGI/mpich versions, use mpirun -iface ib0 .... | ||
| - | mpirun -np $nproc | + | mpirun -np $nproc SU2_CFD inv_ONERAM6_JST.cfg > SU2_CFD.out |
| ### Check to see if the solution_flow.dat file exists. | ### Check to see if the solution_flow.dat file exists. | ||
| ### use it to generate the necessary output files. | ### use it to generate the necessary output files. | ||
| Line 42: | Line 42: | ||
| </ | </ | ||
| - | ===== Postprocesing | + | ===== Postprocessing |
| By default, the SU2 example cases write Tecplot format output files. | By default, the SU2 example cases write Tecplot format output files. | ||
| for Tecplot, although it is possible to read Tecplot files into Paraview. | for Tecplot, although it is possible to read Tecplot files into Paraview. | ||
| it is easy to specify Paraview format output files (VTK) directly. | it is easy to specify Paraview format output files (VTK) directly. | ||
| + | |||
| + | ===== Scaling ===== | ||
| + | SU2 scales reasonably well, down to about 50 000 cells per core. Performance with the Intel compiler is better than GCC or PGI. Refer to the following graph: | ||
| + | |||
| + | {{: | ||