Conversation
|
I am going to a real pain in the neck for you ... The script you have submitted seems to be very dependent on Atos@ECMWF. I can work with you to make this more general. I have some bash logic that we can re-use to make this more user-friendly. |
|
Hi @ewhelan, Sure, I will be happy to work with you in this. Please provide the bash logic. |
|
I have definitely broken what you have developed but it is now a little more user-friendly. Take a look and let me know what you think. |
| #SBATCH --time=02:00:00 | ||
| # | ||
|
|
||
| EXP=SURF_CARRA2_CY46 |
There was a problem hiding this comment.
I do not believe we should have a dependency on a Harmonie experiement to run this.
|
|
||
| EXP=SURF_CARRA2_CY46 | ||
| ENV=${HOME}/hm_home/$EXP | ||
| source ${ENV}/Env_system |
There was a problem hiding this comment.
You should extract what is environment is needed. i.e. the correct modules at ECMWF.
| source ${ENV}/Env_system | ||
|
|
||
| export MPPEXEC="srun" | ||
| export NPROC=999 |
There was a problem hiding this comment.
Hard-codings for NPROC should be replaced by command-line settings
| &NAMFESTAT | ||
| NPRINTLEV=0, | ||
| NCASES=$NPROC, | ||
| NFRGRP=9, |
There was a problem hiding this comment.
Should be derived, I guess, rather than hard-coded.
| LUVBAL=.FALSE., | ||
| OUTBAL='stab.bal' | ||
| OUTCVT='stab.cvt', | ||
| OUTCVU='stab.cv', |
There was a problem hiding this comment.
Would it be useful to add the NetCDF swithc here (LNETCDF_IO) as an option?
2 participants
No description provided.