Residuals batch mode
I am running FLUENT 6 parallel in batch mode (GMPI) on a Linux cluster with the unsteady solver. I noticed that my residuals (x- and y-velocity, continuity, and energy) do not converge between each time step. They just fluctuate. However, when I am running exactly the same case and data files interactively with FLUENT parallel (smpi or net) or even with the serial version of FLUENT, my residuals behave well. They decrease further by 2 orders of magnitude and converge between each time step. Has anyone ever had this problem? Do you know what could be the problem?
|All times are GMT -4. The time now is 11:40.|