GEFS -CHEM PLOTS MPMD BUGFIX - Feature/gef splotmpmd #664
Conversation
sync repository 12/20/2024
sync repository 12/31/2024
sync repository 01/03/2024
sync repository 01/14/2025
sync with authritative repository 02/03/2025
sync local repository 02/10/2025
|
Initial look over:
|
|
test instructions revised to include stats email message |
|
Doing the missing data tests, I built a prep directory at /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/evs/v2.0/prep/global_ens and linked to all chem directories except for chem.20250206. COMIN is pointing to that directory for testing. COMOUT is /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/evs/v2.0. 1. jevs_global_ens_gefs_chem_grid2obs_aeronet_statsLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/EVS/dev/drivers/scripts/stats/global_ens/jevs_global_ens_gefs_chem_grid2obs_aeronet_stats.o6914362 2. jevs_global_ens_gefs_chem_grid2obs_airnow_statsLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/EVS/dev/drivers/scripts/stats/global_ens/jevs_global_ens_gefs_chem_grid2obs_airnow_stats.o6914417 |
|
The mail message generated as expected but I did not receive the email. I try the cmd line emailing submission, and I received the cmd line submitted email. Do you have any suggestion on the one sent in job queue? cat /lfs/h2/emc/stmp/mallory.row/evs_test/prod/tmp/jevs_global_ens_gefs_chem_grid2obs_aeronet_stats.6914362.dbqs01/mailmsg cat /lfs/h2/emc/stmp/mallory.row/evs_test/prod/tmp/jevs_global_ens_gefs_chem_grid2obs_airnow_stats.6914417.dbqs01/mailmsg |
|
I did not receive an email either, though I put myself on MAILTO. I don't know why we didn't get it. Is the purpose of these changes to put all missing data files in 1? |
|
Yes, I recall NCO said there are too many missing file email and they want it to be reduced. Later NCO cancel the MAIL option during operational. This is for EMC EVS parallel only for missing data. |
|
Yes, I too finding one email per file for stats step in one hours is too much. This changes is to put all missing file of an hour in one email. |
|
You are correct; however, we are going about resolving that in a different way. Here are the notes from the Verification Team Meeting on 1/7/2025 under the Missing Data Email section. If the changes are for thise, I would back out these changes and have them in another PR. There are other changes needed to meet what is needed for how we plan on delivering missing data emails to NCO. |
|
Hi, Mallory:
I retract my statements for exevs_global_ens_chem_grid2obs_stats.sh.
The changes is to fix the bug of current EMC parallel, that can not send
email for missing files
on develop version
(1) line 21, > should be >>
(2) already change directory to ${finalstat} in line 173 and no longer in
${DATA}. Thus it can not find mailmsg in line 189-190.
I will back it out to the current develop version and test the fix in your
new design later.
Ho-Chun
…--
Ho-Chun Huang, Ph.D.
Physical Scientist III, Contractor with Lynker in Support of
NOAA/NWS/NCEP/EMC, U.S. Department of Commerce
5830 University Research Ct., Rm. 2792
College Park, MD 20740
***@***.*** ***@***.***>
301-683-3958
On Mon, Feb 10, 2025 at 2:18 PM Mallory Row ***@***.***> wrote:
You are correct; however, we are going about resolving that in a different
way. Here are the notes from the Verification Team Meeting on 1/7/2025
<https://docs.google.com/document/d/1JwXz2SA99wbL7Io74Q7BWHbl_GPCOM-FROcmGi-02Uo/edit?tab=t.0#heading=h.hfko2wg14oea>
under the Missing Data Email section.
If the changes are for thise, I would back out these changes and have them
in another PR. There are other changes needed to meet what is needed for
how we plan on delivering missing data emails to NCO.
—
Reply to this email directly, view it on GitHub
<#664 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALPHE3FKUY7TRQSKCHACCBT2PD3PHAVCNFSM6AAAAABW2UYTZ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBZGAYDOMRVGU>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
|
Backout the changes in exevs_global_ens_chem_grid2obs_stats.sh and remove the instruction for testing the changes. |
|
I *retract* my statements for exevs_global_ens_chem_grid2obs_stats.sh.
|
sync local repository 02/12/2025
|
COMOUT is /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/evs/v2.0/plots/global_ens/chem.20250208. 1. jevs_global_ens_chem_gefs_grid2obs_aeronet_plots_last31daysLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/EVS/dev/drivers/scripts/plots/global_ens/jevs_global_ens_chem_gefs_grid2obs_aeronet_plots_last31days.o180371595 2. jevs_global_ens_chem_gefs_grid2obs_aeronet_plots_last90daysLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/EVS/dev/drivers/scripts/plots/global_ens/jevs_global_ens_chem_gefs_grid2obs_aeronet_plots_last90days.o180371597 3. jevs_global_ens_chem_gefs_grid2obs_airnow_plots_last31daysLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/EVS/dev/drivers/scripts/plots/global_ens/jevs_global_ens_chem_gefs_grid2obs_airnow_plots_last31days.o180371598 4. jevs_global_ens_chem_gefs_grid2obs_airnow_plots_last90daysLog File: /lfs/h2/emc/vpppg/noscrub/mallory.row/verification/EVS_PRs/pr664/EVS/dev/drivers/scripts/plots/global_ens/jevs_global_ens_chem_gefs_grid2obs_airnow_plots_last90days.o180371605 |
|
No error or warning message were found in the logs file (except for IP address error). The figures look okay to me https://www.emc.ncep.noaa.gov/users/verification/global/gefs/pr/chem/grid2obs/airnow_pm25/ |
|
Everything looks good to me too! One time though, I see the walltime for all the scripts are 1 hour, but they all ran in less than 15 minutes. Following the guidelines from NCO, could we get the walltimes reduced to 25 minutes (both dev drivers and ecf scripts)? |
|
Mallory, how do you check the cpu time in the GEFS-chem plots step? "cput=" ? since resources_used.walltime = 00:00:00. Because cput for nid* is over 12 min, will 15 min be cutting too close? Since the plots code has restart function, can we set it at higher walltime? |
|
Please ignore my previous comments, you did say 25 mins. I need to check my eysight. |
|
All good! Future reference, I usually grep for "resources_used.walltime". I don't think it works in some cases, particularly on Cactus, so if that has all zeros I grep for "stime =" and "mtime =" and do a rough estimate. |
There was a problem hiding this comment.
Code changes are good and testing successful.
Thanks for this work @Ho-ChunHuang-NOAA!
|
Thank you all for the help to get this PR done. |
…A/EVS into feature/href_MPMD * 'feature/href_MPMD' of https://github.com/BinbinZhou-NOAA/EVS: Update naming conventions per Executive Order 14172 (NOAA-EMC#666) GEFS -CHEM PLOTS MPMD BUGFIX - Feature/gef splotmpmd (NOAA-EMC#664)
…AA/EVS into feature/narre_MPMD * 'feature/narre_MPMD' of https://github.com/BinbinZhou-NOAA/EVS: Update naming conventions per Executive Order 14172 (NOAA-EMC#666) GEFS -CHEM PLOTS MPMD BUGFIX - Feature/gef splotmpmd (NOAA-EMC#664) bugfix/long_term_stats (NOAA-EMC#663) Global-det SST sea ice fix (NOAA-EMC#660) bugfix/dev_fix_ghrsst (NOAA-EMC#662)
Note to developers: You must use this PR template!
Description of Changes
Code revision to have mpmd processes running in its own child working directory
Developer Questions and Checklist
Is this a high priority PR? If so, why and is there a date it needs to be merged by?
NO
Do you have any planned upcoming annual leave/PTO?
YES, APRIL 2025
Are there any changes needed in the times when the jobs are supposed to run/kick-off?
NO
${USER}where necessary throughout the code.HOMEevsare removed from the code.dev/drivers/scriptsordev/modulefileshave been made in the correspondingecf/scriptsandecf/defs/evs-nco.def?Testing Instructions