-
Notifications
You must be signed in to change notification settings - Fork 253
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
update with fv3 & ccpp/physics updates in order to turn the NSST model on in the coupled model (Replace PR #453) #483
update with fv3 & ccpp/physics updates in order to turn the NSST model on in the coupled model (Replace PR #453) #483
Conversation
tests/rt.conf
Outdated
@@ -133,8 +133,9 @@ RUN | fv3_gfs_v16_ras_debug | |||
# CPLD tests # | |||
################################################################################################################################################################################### | |||
|
|||
COMPILE | APP=S2S SUITES=FV3_GFS_2017_coupled,FV3_GFS_2017_satmedmf_coupled,FV3_GFS_v15p2_coupled,FV3_GFS_v16_coupled | - wcoss_cray jet.intel | fv3 | | |||
COMPILE | APP=S2S SUITES=FV3_GFS_2017_coupled,FV3_GFS_2017_satmedmf_coupled,FV3_GFS_v15p2_coupled,FV3_GFS_v16_coupled,FV3_GFS_v15p2_couplednsst | - wcoss_cray jet.intel | fv3 | |
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.
Would you please add a GFSv16 coupled case as coupled P6 already moved to GFSv16_coupled? We may remove the old configures soon.
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.
You mean to add one ccpp suite, like FV3_GFS_v16_couplednsst, since FV3_GFS_v15p2_coupled will be removed soon.
If yes, I can create a ccpp suite, FV3_GFS_v16_couplednsst, based on FV3_GFS_v16_coupled, as I did for FV3_GFS_v15p2_couplednsst. But a test is required to deliver FV3_GFS_v16_couplednsst.
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.
@XuLi-NOAA Please close PR #453 since it has been replaced by PR #483.
PR #453 closed since it has been replaced by # 483
Yes, that is what I mean.
…On Thu, Mar 25, 2021 at 4:54 PM XuLi-NOAA ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In tests/rt.conf
<#483 (comment)>
:
> @@ -133,8 +133,9 @@ RUN | fv3_gfs_v16_ras_debug
# CPLD tests #
###################################################################################################################################################################################
-COMPILE | APP=S2S SUITES=FV3_GFS_2017_coupled,FV3_GFS_2017_satmedmf_coupled,FV3_GFS_v15p2_coupled,FV3_GFS_v16_coupled | - wcoss_cray jet.intel | fv3 |
+COMPILE | APP=S2S SUITES=FV3_GFS_2017_coupled,FV3_GFS_2017_satmedmf_coupled,FV3_GFS_v15p2_coupled,FV3_GFS_v16_coupled,FV3_GFS_v15p2_couplednsst | - wcoss_cray jet.intel | fv3 |
You mean to add one ccpp suite, like FV3_GFS_v16_couplednsst, since
FV3_GFS_v15p2_coupled will be removed soon.
If yes, I can create a ccpp suite, FV3_GFS_v16_couplednsst, based on
FV3_GFS_v16_coupled, as I did for FV3_GFS_v15p2_couplednsst. But a test is
required to deliver FV3_GFS_v16_couplednsst.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#483 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AI7D6TOQE6SVGMVVAHAMVP3TFOPGBANCNFSM4ZZ6XFZQ>
.
|
@XuLi-NOAA Please close PR #453 since it has been replaced by PR #483. |
OK, I have tested the created ccpp suite, FV3_GFS_v16_couplednsst, with P6. It works well. I will add it to fv3atm branch, feature/ufs-nst. |
@XuLi-NOAA This PR is behind the develop branch and needs to be updated. |
tests/tests/cpld_bmarkfrac_v16_nsst
Outdated
|
||
export TEST_DESCR="Fully coupled FV3-CCPP-MOM6-CICE-CMEPS system - C384L127 MX025 - Benchmark test - frac grid - v16" | ||
|
||
export CNTL_DIR="cpld_bmarkfrac_v16" |
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 needs to be cpld_bmarkfrac_v16_nsst. That is the test you are comparing against.
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 needs to be cpld_bmarkfrac_v16_nsst. That is the test you are comparing against.
@DeniseWorthen: Yes.
And another change is needed too:
export NSTF_NAME=2,1,0,0,0 to export nstf_name=2,1,0,0,0 in tests/cpld_bmarkfrac_v16_nsst.
Actually, I'd like to point out, it is good to always use lower (or upper) case for the parm files. Currently, both nstf_name and NSTF_NAME are used. This should be unified (for other variables as well, if apply).
…cpld_bmarkfrac_v16_nsst. 2. export NSTF_NAME=2,1,0,0,0 to export nstf_name=2,1,0,0,0.
tests/tests/cpld_bmarkfrac_v16_nsst
Outdated
export FIELD_TABLE="field_table_gfsv16" | ||
export DIAG_TABLE="diag_table_bmark_v16_template" | ||
export SUITE_NAME="FV3_GFS_v16_couplednsst" | ||
export nstf_name=2,1,0,0,0 |
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.
You need to make nstf_name in the file input.benchmark_v16.nml.IN to be a variable, which is defined here in the test. So, here define: NSTF_NAME='2,1,0,0,0'. Then in input.benchmark_v16.nml.IN set nstf_name = @[NSTF_NAME]. This will also need to be done in the same way for the control_nsst test that you added.
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.
You need to make nstf_name in the file input.benchmark_v16.nml.IN to be a variable, which is defined here in the test. So, here define: NSTF_NAME='2,1,0,0,0'. Then in input.benchmark_v16.nml.IN set nstf_name = @[NSTF_NAME]. This will also need to be done in the same way for the control_nsst test that you added.
@DeniseWorthen: Yes, it is better, since it is consistent with all the places of nstf_name in the parm/input*nml files. I change NSTF_NAME to be nstf_name since " nstf_name = 0,0,0,0,0" has been defined in input.benchmark_v16.nml.IN, I will remove this, and add nstf_name = @[NSTF_NAME] in input.benchmark_v16.nml.IN, then I can keep the upper case NSTF_NAME='2,1,0,0,0' in cpld_bmarkfrac_v16_nsst, as I did for cpld_control_nsst. This is the first time for me to create a test, a good experience.
…sistent definition of nstf_name namelist
…st field updates section
@XuLi-NOAA Please update your branch. Thanks! |
@binli2337 : The branches have been updated and the new baseline has been generated on hera, orion and wcoss_dell_p3, see: |
tests/tests/cpld_control_nsst
Outdated
RESTART/phy_data.tile4.nc \ | ||
RESTART/phy_data.tile5.nc \ | ||
RESTART/phy_data.tile6.nc \ | ||
RESTART/sfc_data.tile1.nc \ |
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.
I don't think this test cpld_control_nsst
is used? I don't see SUITE_NAME
being set, how is this test supposed to run/
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.
I think you should remove this file tests/tests/cpld_control_nsst
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.
@XuLi-NOAA You need to either remove this file from the PR or add it as a test in rt.conf. If you add it as a test, you need to set the SUITE_NAME in the test, otherwise it will use the default suite in input.mom6_ccpp.nml.IN
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.
I don't think this test
cpld_control_nsst
is used? I don't seeSUITE_NAME
being set, how is this test supposed to run/
@climbfuji : cpld_control_nsst is not used since it doesn't use v16 physics, as suggested by Bin & Jun, another one, has been created and tested, cpld_bmarkfrac_v16_nsst.
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.
Thanks, in this case, please remove tests/tests/cpld_control_nsst
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.
Thanks, in this case, please remove
tests/tests/cpld_control_nsst
@climbfuji : I will remove cpld_control_nsst and it doesn't affect anything we have done.
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.
@climbfuji : cpld_control_nsst removed
New baseline directory name is "develop-20210406". |
@XuLi-NOAA can you please update rt.sh for the newbaseline date (BL_DATE)? |
@XuLi-NOAA The new baseline directory name is "develop-20210406". Please update line 416 of the "rt.sh" file before you run regression tests to generate log files. |
…ld_bmarkfrac_v16_nsst
Looks like both didn't have baseline files? |
I saw time-outs in both err files. We need to save this log file and then either Bin or Xu can re-run just the failed tests and append it. |
MISSING file means that test didn't run (at all or to completion). If the baseline was missing, it would say "missing baseline" |
@XuLi-NOAA Are your rt tests on orion completed successfully? |
@binli2337 : still running, the same for hera and dell. |
@XuLi-NOAA So far the following log files are ready: jet: /mnt/lfs4/HFIP/hfv3gfs/Bin.Li/update Would you please copy those RT log files to your ufs-weather-model branch tests directory and commit them? |
@XuLi-NOAA I have transferred the following 7 log files to /scratch1/NCEPDEV/nems/Bin.Li/PR483 on Hera: What is the status of your runs on Hera and wcoss_p3? |
@binli2337 : tests on hera stopped at number 66 due to a disk problem (stemp2), I am running rt.sh again since No. 59 test, and then, can we just generate the log file by appending/editing? wcoss_p3 has reached test 65. |
@XuLi-NOAA Please save the log files for the successful runs. It is OK to combine the log files. |
@XuLi-NOAA can you please write run-ci in one of your future commit message? This will trigger the CI test. |
@XuLi-NOAA The wcoss_dell_p3 log file is in /scratch1/NCEPDEV/nems/Bin.Li/PR483 now. |
@binli2337 @BrianCurtis-NOAA : Thanks. Where is the log file for RegressionTests_hera.intel.log? |
CI tests passed except the debug test; this is an EC2 issue. I suggest we accept the results and move on with the commit. |
Please don't merge it. |
Ok, now that fv3atm has been corrected, please go to your codebase (from which you created the branch feature/nsst_ufs), go to FV3, make sure that |
* First test * Change branch name * github event name... * change to tests/ci dir * change repo name * print out stderr * try recursive checkout * checkout myself * in a hurry * fetch owner id * Make setup depend on prcheck * update with fv3 & ccpp/physics updates in order to turn the NSST model on in the coupled model (Replace PR ufs-community#453) (ufs-community#483) * point to fv3 branch * Add one more test cpld_control_nsst to test added ccpp suite (FV3_GFS_v15p2_couplednsst) * Modify rt.conf to add FV3_GFS_v15p2_couplednsst * Modify rt.conf to add a new test, cpld_bmarkfrac_v16_nsst, remove test cpld_control_nsst * Modify tests/tests/cpld_bmarkfrac_v16_nsst: 1. cpld_bmarkfrac_v16 to cpld_bmarkfrac_v16_nsst. 2. export NSTF_NAME=2,1,0,0,0 to export nstf_name=2,1,0,0,0. * Modify input.benchmark_v16.nml.IN & cpld_bmarkfrac_v16_nsst for a consistent definition of nstf_name namelist * Modify cpld_bmarkfrac_v16_nsst by moving the NSTF_NAME to the namelist field updates section * RegressionTests_orion.intel.log of the rt run and BL_DATE=20210406 in rt.sh * RT JOBS PASSED: hera.intel. Log file uploaded. * run-ci, commit 7 RegressionTest log files * Push RegressionTests_wcoss_dell_p3.log Co-authored-by: Brian Curtis <brian.curtis@noaa.gov> * remove develop * quiet git commands. run-ci * Update CMEPS for latest ESCOMP/master; Update PIO to 2.5.2; Refactor modules (ufs-community#492) * update ufs for CMEPS master with PIO changes * update pio to 2.5.2 across platforms * replace variable SUITE_NAME with CCPP_SUITE * Merge remote-tracking branch 'DusanJovic/module_common' into feature/updcmeps * switch to h-nems area on jet Co-authored-by: Bin Li <Bin.Li@gaea13.ncrc.gov> Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov> * debug this again... * debug again * debug continues * missed repo field * remove debug related texts * remove -x flag. run-ci * reduce sleep time * diag * diag2 * Fix pr_uid * diag again Co-authored-by: XuLi-NOAA <55100838+XuLi-NOAA@users.noreply.github.com> Co-authored-by: Brian Curtis <brian.curtis@noaa.gov> Co-authored-by: Denise Worthen <denise.worthen@noaa.gov> Co-authored-by: Bin Li <Bin.Li@gaea13.ncrc.gov> Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov>
* First test * Change branch name * github event name... * change to tests/ci dir * change repo name * print out stderr * try recursive checkout * checkout myself * in a hurry * fetch owner id * Make setup depend on prcheck * update with fv3 & ccpp/physics updates in order to turn the NSST model on in the coupled model (Replace PR ufs-community#453) (ufs-community#483) * point to fv3 branch * Add one more test cpld_control_nsst to test added ccpp suite (FV3_GFS_v15p2_couplednsst) * Modify rt.conf to add FV3_GFS_v15p2_couplednsst * Modify rt.conf to add a new test, cpld_bmarkfrac_v16_nsst, remove test cpld_control_nsst * Modify tests/tests/cpld_bmarkfrac_v16_nsst: 1. cpld_bmarkfrac_v16 to cpld_bmarkfrac_v16_nsst. 2. export NSTF_NAME=2,1,0,0,0 to export nstf_name=2,1,0,0,0. * Modify input.benchmark_v16.nml.IN & cpld_bmarkfrac_v16_nsst for a consistent definition of nstf_name namelist * Modify cpld_bmarkfrac_v16_nsst by moving the NSTF_NAME to the namelist field updates section * RegressionTests_orion.intel.log of the rt run and BL_DATE=20210406 in rt.sh * RT JOBS PASSED: hera.intel. Log file uploaded. * run-ci, commit 7 RegressionTest log files * Push RegressionTests_wcoss_dell_p3.log Co-authored-by: Brian Curtis <brian.curtis@noaa.gov> * remove develop * quiet git commands. run-ci * Update CMEPS for latest ESCOMP/master; Update PIO to 2.5.2; Refactor modules (ufs-community#492) * update ufs for CMEPS master with PIO changes * update pio to 2.5.2 across platforms * replace variable SUITE_NAME with CCPP_SUITE * Merge remote-tracking branch 'DusanJovic/module_common' into feature/updcmeps * switch to h-nems area on jet Co-authored-by: Bin Li <Bin.Li@gaea13.ncrc.gov> Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov> * debug this again... * debug again * debug continues * missed repo field * remove debug related texts * remove -x flag. run-ci * reduce sleep time * diag * diag2 * Fix pr_uid * diag again * still debugging * lets see if this works * use context as main yml key Co-authored-by: XuLi-NOAA <55100838+XuLi-NOAA@users.noreply.github.com> Co-authored-by: Brian Curtis <brian.curtis@noaa.gov> Co-authored-by: Denise Worthen <denise.worthen@noaa.gov> Co-authored-by: Bin Li <Bin.Li@gaea13.ncrc.gov> Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov>
* First test * Change branch name * github event name... * change to tests/ci dir * change repo name * print out stderr * try recursive checkout * checkout myself * in a hurry * fetch owner id * Make setup depend on prcheck * update with fv3 & ccpp/physics updates in order to turn the NSST model on in the coupled model (Replace PR ufs-community#453) (ufs-community#483) * point to fv3 branch * Add one more test cpld_control_nsst to test added ccpp suite (FV3_GFS_v15p2_couplednsst) * Modify rt.conf to add FV3_GFS_v15p2_couplednsst * Modify rt.conf to add a new test, cpld_bmarkfrac_v16_nsst, remove test cpld_control_nsst * Modify tests/tests/cpld_bmarkfrac_v16_nsst: 1. cpld_bmarkfrac_v16 to cpld_bmarkfrac_v16_nsst. 2. export NSTF_NAME=2,1,0,0,0 to export nstf_name=2,1,0,0,0. * Modify input.benchmark_v16.nml.IN & cpld_bmarkfrac_v16_nsst for a consistent definition of nstf_name namelist * Modify cpld_bmarkfrac_v16_nsst by moving the NSTF_NAME to the namelist field updates section * RegressionTests_orion.intel.log of the rt run and BL_DATE=20210406 in rt.sh * RT JOBS PASSED: hera.intel. Log file uploaded. * run-ci, commit 7 RegressionTest log files * Push RegressionTests_wcoss_dell_p3.log Co-authored-by: Brian Curtis <brian.curtis@noaa.gov> * remove develop * quiet git commands. run-ci * Update CMEPS for latest ESCOMP/master; Update PIO to 2.5.2; Refactor modules (ufs-community#492) * update ufs for CMEPS master with PIO changes * update pio to 2.5.2 across platforms * replace variable SUITE_NAME with CCPP_SUITE * Merge remote-tracking branch 'DusanJovic/module_common' into feature/updcmeps * switch to h-nems area on jet Co-authored-by: Bin Li <Bin.Li@gaea13.ncrc.gov> Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov> * debug this again... * debug again * debug continues * missed repo field * remove debug related texts * remove -x flag. run-ci * reduce sleep time * diag * diag2 * Fix pr_uid * diag again * still debugging * lets see if this works * use context as main yml key * Try pull request now. run-ci * minor change. run-ci * typo fix * debug 1 * debug2 * minor mods. run-ci * check another dir * sleep * sigh * sigh2 * increase time again * test again Co-authored-by: XuLi-NOAA <55100838+XuLi-NOAA@users.noreply.github.com> Co-authored-by: Brian Curtis <brian.curtis@noaa.gov> Co-authored-by: Denise Worthen <denise.worthen@noaa.gov> Co-authored-by: Bin Li <Bin.Li@gaea13.ncrc.gov> Co-authored-by: Dusan Jovic <dusan.jovic@noaa.gov>
* Add modulefiles for verification on WCOSS * Modify workflow template for WCOSS * Add a new parameter for the directory name of MET executables
Incorporate the changes in fv3atm and ccpp/physics for the NSST model in the coupled system .
See issue #350
fv3atm changes: see NOAA-EMC/fv3atm#219
Reduce the initialized warm layer thickness to be 20 m when the NSST model is cold start in FV3 io.
Add one ccpp suite: suite_FV3_GFS_v15p2_couplednsst.xml
Include the change in ccpp/physics submodule (see NCAR/ccpp-physics#537).
ccpp/physics changes: see NCAR/ccpp-physics#537
Three changes in this issue and a bug fix
(1) Scheme-A to Scheme-A.1 in ccpp/physics
The cooling effect is removed from the formula to determine Tf
(2) Apply a safeguard to tref in sfc_nst_pre_run
(3) Reduce z_w_max from 30 m to 20 in ccpp/physics and FV3/io
A bug fix in ccpp physics related to NSST model in the coupled mode
The new baseline results are at: /scratch1/NCEPDEV/stmp4/Xu.Li/FV3_RT/REGRESSION_TEST_INTEL, the new baseline has been generated since the changes alter the results related to the NSST model.
There is no library update.
The results are as expected.
Issue(s) addressed
#350
NOAA-EMC/fv3atm#219
NCAR/ccpp-physics#537
Testing
A new test, cpld_control_nsst has been generated and included in the regression tests, the added ccpp suite, suite_FV3_GFS_v15p2_couplednsst.xml, has been added in rt.conf
The regression tests have been done on Hera by:
./rt.sh -ce >baseline_new
./rt.sh -me
Dependencies:
FV3atm PR#255
CCPP physics PR #589