-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Make atmos_fcst job work using GFSv17 and new COM structure #120
Make atmos_fcst job work using GFSv17 and new COM structure #120
Conversation
On branch feature/v13_atmos_fcst renamed: parm/gefs_fcst.parm -> parm/parm_gefs/gefs_fcst.parm Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: rocoto/bin/wcoss2/common.sh modified: rocoto/bin/wcoss2/forecast_hr.sh modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
Also modify parm files On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: parm/.gitignore new file: parm/config/config.ufs modified: parm/parm_gefs/gefs.parm modified: parm/parm_gefs/gefs_fcst.parm new file: parm/parm_gefs/gefs_fcst_resources.parm modified: sorc/link_gefs.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: parm/.gitignore modified: parm/config/config.ufs modified: parm/parm_gefs/gefs.parm modified: parm/parm_gefs/gefs_fcst.parm modified: rocoto/parm/setbase modified: rocoto/py/GEFS_XML_For_Tasks.py modified: rocoto/user_full.conf modified: scripts/exgefs_forecast.sh modified: sorc/link_gefs.sh modified: ush/.gitignore Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: rocoto/user_full.conf Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: parm/parm_gefs/gefs_fcst.parm modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: rocoto/bin/hera/forecast_hr.sh modified: rocoto/bin/wcoss2/forecast_hr.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: rocoto/bin/hera/forecast_hr.sh modified: rocoto/parm/setbase Refs: NOAA-EMC#94
#export jchunk2d=$((2*RESTILE)) | ||
#export ichunk3d=$((4*RESTILE)) | ||
#export jchunk3d=$((2*RESTILE)) | ||
#export kchunk3d=1 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Lines 210~252 are for GEFS fcst job, especially for d16_35 and chem
parm/parm_gefs/gefs_fcst.parm
Outdated
elif [[ "$imp_physics" -eq 11 ]]; then # GFDL | ||
export ncld=5 | ||
if [[ $cplchm = ".true." ]]; then | ||
export FIELD_TABLE="$HOMEgfs/parm/parm_fv3diag/chm_field_table_gfdl" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is for chem, maybe we don't use it in gefs.v13
parm/parm_gefs/gefs_fcst.parm
Outdated
# Disable the use of coupler.res; get model start time from model_configure | ||
export USE_COUPLER_RES="NO" | ||
|
||
# Write more variables to output |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In original config.fcst, there is no condition for gefs
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@WalterKolczynski-NOAA @aerorahul Could you please review my changes for fcst job? I have a few comments for the config files, could you also give me some suggestions? I have ran c00, p01 and p02 (C48) succeeded on Hera, but failed for p02 on WCOSS2 because of instability of the model. |
On branch feature/v13_atmos_fcst modified: parm/parm_gefs/gefs_fcst.parm modified: rocoto/user_full.conf Refs: NOAA-EMC#94
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Instead of duplicating the contents of GFS config files to make a few changes, you should be making changes to the config files in global workflow to support GEFS and/or adding config files that just change the things that are different. We want GEFS to run more more like GFS.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Much of this should be in a config file.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
in config.fcst in g-w? If it is, you and @aerorahul should determine can I use cdump=gefs
? or only use RUN=gefs
? If you want to use run=gefs, then we should modify the config* in g-w to use RUN instead of cdump
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Per the slack and in-person conversations, make GEFS versions of the config files and then make sure these settings are defined in there.
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
@WalterKolczynski-NOAA I have submitted a few commits based on your comment. When you have time, please review it. |
On branch feature/v13_atmos_fcst renamed: parm/parm_gefs/gefs_fcst_resources.parm -> parm/config/config.resources modified: parm/parm_gefs/gefs_fcst.parm Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/parm_gefs/gefs_fcst.parm modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_ATMOS_PREP modified: jobs/JGEFS_FORECAST renamed: parm/parm_gefs/gefs.parm -> parm/config/config.base renamed: parm/parm_gefs/gefs_fcst.parm -> parm/config/config.fcst renamed: parm/parm_gefs/gefs_atmos_prep.parm -> parm/config/config.prep Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_ATMOS_PREP Refs: NOAA-EMC#94
The current g-w can not be used as flat structure for GEFS so I change the default structure is vertical structure On branch feature/v13_atmos_fcst modified: rocoto/compile_install_all.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: rocoto/compile_install_all.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.fcst Refs: NOAA-EMC#94
after delete them On branch feature/v13_atmos_fcst Changes to be committed: modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst Changes to be committed: modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.base modified: parm/config/config.fcst modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.fcst Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.fcst modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
@WalterKolczynski-NOAA @aerorahul I have moved some vars from ex-script of fcst to config.fcst and also did other modification. Please review my changes. Thanks |
@WalterKolczynski-NOAA @aerorahul When you have time, could you review my changes for this PR? Thanks. |
On branch feature/v13_atmos_fcst modified: sorc/checkout.sh Refs: NOAA-EMC#94
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Didn't complete the review of the config files. It seems like a lot of GEFS settings were tacked on to GFS settings without considering whether they are still needed or to update them to match the GFS style (for instance, using $cplchm
for branching instead of $DO_AERO
). Many settings have been updated for GFS v17/GEFS v13 and we should defer to the new settings as given in GFS unless there is a good reason to deviate (like turning on stochastic physics for some members). Please carefully review all of the settings and determine whether it is conflicting with an existing setting elsewhere. If you have any trouble determining whether a setting should be overridden for GEFS, I'll help and if necessary we'll ask MDAB.
Also, note the changes that were made in the NOAA-EMC/global-workflow#1509. You've updated the commit hash, but there are changes to the config files that will need to be duplicated here.
export GESIN=${GESIN:-$(compath.py $envir/com/$NET/${ver})/${RUN}.${PDY}/$cyc/nwges} | ||
export GESOUT=${GESOUT:-$(compath.py -o $NET/${ver})/${RUN}.${PDY}/$cyc/nwges} | ||
ver=${gefs_ver:0:5} | ||
export ROTDIR=${ROTDIR:-$(compath.py $envir/com/$NET/${ver})} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This should be defined before the j-job is called, and definitely should not use compath.py
. compath.py
should only be used for external packages (GFS, tracker, etc.).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is it the NCO new requirements? Because all the compath.py are in j-job in the current production.
fi | ||
|
||
# Choose coupling with wave | ||
if [[ "$DO_WAVE" = "YES" && "$WAVE_CDUMP" != "gdas" ]]; then |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
if [[ "$DO_WAVE" = "YES" && "$WAVE_CDUMP" != "gdas" ]]; then | |
if [[ "$DO_WAVE" = "YES" ]]; then |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is the copy of code segments from config.fcst in g-w, I will follow the the config.fcst of g-w changes
|
||
export ROTDIR=${COMROOT}/gefs/${ver} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For development, I don't think we need to add the additional unnecessary directory levels that NCO is.
export ROTDIR=${COMROOT}/gefs/${ver} | |
# Assuming $EXPDIR is already built into $COMROOT | |
ROTDIR=${COMROOT} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
GEFS will not use EXPDIR, This line is to set ROTDIR because I call j-job. This is only be used on Hera.
|
||
export ROTDIR=${COMROOT}/gefs/${ver} | ||
export ROTDIR_GFS=${HOMEdata}/gfs/${gfs_ver} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Same here. For GFS development we don't include the extra directory levels NCO does. For operations, this needs to use compath.py
, since GFS is an external package.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, however, this is only used on Hera. On wcos2, I will use compath.py to get the path, but it can not used on Hera.
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: parm/config/config.base modified: parm/config/config.fcst modified: parm/config/config.resources modified: parm/config/config.ufs Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST modified: parm/config/config.base Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.fcst modified: parm/config/config.ufs modified: scripts/exgefs_forecast.sh Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.resources Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: parm/config/config.fcst Refs: NOAA-EMC#94
On branch feature/v13_atmos_fcst modified: jobs/JGEFS_FORECAST new file: parm/config/config.efcs modified: parm/config/config.fcst Refs: NOAA-EMC#94
@WalterKolczynski-NOAA @aerorahul I have decided the config.fcst into two files: config.fcst and config.efcs. Which will shorten the config.fcst. Please re-review my changes for this PR. I think I have resolved all the comments. If I missed something, please let me know. |
On branch feature/v13_atmos_fcst modified: parm/config/config.efcs Refs: NOAA-EMC#94
Since the GEFSv13 workflow will be in g-w, I will merge this PR to close this PR and for the future's reference. I will do the same PR for the other jobs. If you have any suggestions, please let me know. @WalterKolczynski-NOAA @aerorahul |
The job works on both WCOSS2 and HARA.