Restart v2.0.5
Posted: Wed Jul 24, 2013 2:36 pm
Hello,
I'm using version 2.0.5 and I have a doubt on the use of RESTART files. I'm sorry for the basic nature of my question, but I haven't been able to find an explicit answer in previous posts. I read in a tutorial for version 2.0.0 that to perform a restart, the RESTART directory should be copied from RESU to DATA, but I have not been doing this. I set up the calculation through the GUI as much as I can and, on a workstation, when choosing a RESTART from Calculation Management/Restart it automatically takes me to the RESU directory. I then save the .xml file and launch the runcase script. This has worked for me in the past when restarting an LES from a preliminary RANS or from another LES and my results seemed consistent.
I have recently tried to restart a steady state RANS on a cluster and the results are not what I expected. I followed the same procedure and selected the RESTART file from the RESU directory. In all cases, I did not alter RESTART_IN=$DATA/RESTART in the runcase script. I never questioned my method until now, as the restart time step or iteration number has always been correct.
Could you please tell me if I have been wrong to select the RESTART folder from the RESU directory as opposed to copying it in the DATA directory? Or should I follow a different procedure when using a cluster?
Kindest regards,
Kristin
I'm using version 2.0.5 and I have a doubt on the use of RESTART files. I'm sorry for the basic nature of my question, but I haven't been able to find an explicit answer in previous posts. I read in a tutorial for version 2.0.0 that to perform a restart, the RESTART directory should be copied from RESU to DATA, but I have not been doing this. I set up the calculation through the GUI as much as I can and, on a workstation, when choosing a RESTART from Calculation Management/Restart it automatically takes me to the RESU directory. I then save the .xml file and launch the runcase script. This has worked for me in the past when restarting an LES from a preliminary RANS or from another LES and my results seemed consistent.
I have recently tried to restart a steady state RANS on a cluster and the results are not what I expected. I followed the same procedure and selected the RESTART file from the RESU directory. In all cases, I did not alter RESTART_IN=$DATA/RESTART in the runcase script. I never questioned my method until now, as the restart time step or iteration number has always been correct.
Could you please tell me if I have been wrong to select the RESTART folder from the RESU directory as opposed to copying it in the DATA directory? Or should I follow a different procedure when using a cluster?
Kindest regards,
Kristin