Hello
I followed these steps to create a regional high resolution simulation: Setting up (high-res sparse) regional-grid CTSM simulations · ESCOMP/CTSM · Discussion #1919.
Unfortunately, for large domains (e.g., all California, ~28,000 active grid cells) the turnaround is extremely slow. For a smaller domain (e.g., a river basin in California, ~4,000 grid cells), however, the turnaround is normal.
Throwing more cores at this does not seem to work. Attached, could please find the log files of the simulation. Could please let me know your thoughts on this? I have tried a myriad of changes without success: different hist_* namelists, different mesh files, different NTASKs. I think this might related the model stuck at closing history files.
Thanks so much!
I followed these steps to create a regional high resolution simulation: Setting up (high-res sparse) regional-grid CTSM simulations · ESCOMP/CTSM · Discussion #1919.
Unfortunately, for large domains (e.g., all California, ~28,000 active grid cells) the turnaround is extremely slow. For a smaller domain (e.g., a river basin in California, ~4,000 grid cells), however, the turnaround is normal.
Throwing more cores at this does not seem to work. Attached, could please find the log files of the simulation. Could please let me know your thoughts on this? I have tried a myriad of changes without success: different hist_* namelists, different mesh files, different NTASKs. I think this might related the model stuck at closing history files.
Thanks so much!