Bug during running with Larger Homogeneous Modes
Hi, I am trying to run 3D simulations in parallel with 1D homogeneous in the span-wise direction. For a certain number of Fourier Modes the simulation runs smoothly, but increasing them beyond some critical value causes an error to pop up, and this error is not related to memory shortage. The cluster in which simulation is being run has the capacity of running 192 processors with 2GB Ram/core in parallel environment. Just to give you an idea, in my case with homogeneous modes 64 things are perfect and increasing them causes simulation to crash giving errors like: Fatal : Level 0 assertion violation Exceeded maximum number of iterations srun: error: s02r2b71: tasks 0-47: Exited with exit code 1 CG iterations made = 5001 using tolerance of 1e-09 (error = -nan, rhs_mag = 1) I would appreciate if someone could suggest something. Best Regards
Hi, This is likely due to the run becoming unstable, likely you have a relatively high Reynolds number. To stabilise these types or runs we typically turn on Dealiasing and SpectralVanishingViscosity. Are you using these options? We have a more advanced SpectralVanishing option but that is currently in a developmental branch. Berst regards, Spencer. On 6 Jul 2017, at 17:46, wasim sarwar <waseemsarwar121@hotmail.com<mailto:waseemsarwar121@hotmail.com>> wrote: Hi, I am trying to run 3D simulations in parallel with 1D homogeneous in the span-wise direction. For a certain number of Fourier Modes the simulation runs smoothly, but increasing them beyond some critical value causes an error to pop up, and this error is not related to memory shortage. The cluster in which simulation is being run has the capacity of running 192 processors with 2GB Ram/core in parallel environment. Just to give you an idea, in my case with homogeneous modes 64 things are perfect and increasing them causes simulation to crash giving errors like: Fatal : Level 0 assertion violation Exceeded maximum number of iterations srun: error: s02r2b71: tasks 0-47: Exited with exit code 1 CG iterations made = 5001 using tolerance of 1e-09 (error = -nan, rhs_mag = 1) I would appreciate if someone could suggest something. Best Regards _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk<mailto:Nektar-users@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/nektar-users Spencer Sherwin McLaren Racing/Royal Academy of Engineering Research Chair, Professor of Computational Fluid Mechanics, Department of Aeronautics, Imperial College London South Kensington Campus London SW7 2AZ s.sherwin@imperial.ac.uk<mailto:s.sherwin@imperial.ac.uk> +44 (0) 20 759 45052
participants (2)
- 
                
                Sherwin, Spencer J
- 
                
                wasim sarwar