Hi,
I have an application where I want to run MOM5 model using payu, but I’m only interested in the restart files and don’t care about diagnostic outputs. Thus, I would like to find a way to make payu completely bypass the collate step, because I just need to run a postscript that will extract restart information and then delete the output folder and the restart folder (then modify the original restart and resubmit as if it’s the “same” run).
Is there a way to tell payu not to do collate? Please help
On top of this, if you don’t need the diagnostic output at all, you could choose to turn off diagnostics completely. This will make your model run even faster. Please let me know if this is something that you are interested.
For MOM5 only experiment, we can remove most of the diag_table and only leave the first two lines (i.e., experiment string and the base date). Hopefully this can help fasten your runs