Skip to content

Update gsd/develop from NOAA-EMC develop#35

Merged
DomHeinzeller merged 11 commits into
NOAA-GSL:gsd/developfrom
climbfuji:update_gsd_develop_from_develop
Aug 7, 2020
Merged

Update gsd/develop from NOAA-EMC develop#35
DomHeinzeller merged 11 commits into
NOAA-GSL:gsd/developfrom
climbfuji:update_gsd_develop_from_develop

Conversation

@climbfuji
Copy link
Copy Markdown

@climbfuji climbfuji commented Aug 6, 2020

  • Tom's dycore DA fix
  • CCPP metadata corrections (incl. units)
  • EMC FLake model
  • New suites for SRW App release
  • Build system updates to use imported targets for NCEPLIBS
  • Make stochastic physics independent of fv3atm
  • Downgrade FMS to 2019.01.03.

Associated PRs:

#35
NOAA-GSL/fv3atm#43
NOAA-GSL/ccpp-physics#48
NOAA-GSL/ccpp-framework#7
NOAA-GSL/GFDL_atmos_cubed_sphere#7

For regression testing information, see below.

junwang-noaa and others added 10 commits July 14, 2020 15:11
* add Toms dycore regional DA fix
* update hera intel RT log files
* update fv3 with levp change
* upate baseline to develop-20200713
* copy regional control RESTART

Co-authored-by: Jun Wang <junwang-noaa@users.noreply.github.com>
…ommunity#162)

* Allow multiple regtest ecflow suites to run on the same server
* implement running a single rt test
* Update 'usage' function in rt.sh (added -n option)

Co-authored-by: MinsukJi-NOAA <Minsuk.Ji@noaa.gov>
…ty#167)

* add PR templates for feature and bugfix
Change-Id: Iad15dd1a584f4bd2fbaadf88359a55efd7b1925c
* add codeowners for automatic review requests
Change-Id: If6050e6411519223f514435801e09a63ed696a02
* Update CODEOWNERS
* This is EMC flake model
* Updated .gitmodules file under ufs-weather-model and tests/rt.conf
* Updated .gitmodules file under usf-weather-model
* Point to fv3 flake branch
* Updated ccpp_gfs_v16_flake_run.IN to use the new oro data for flake test
* Updated the following files and directories
* update to  the latest ccpp/physics branch
* Updated compile and regression test log files
* Updated ccpp-physics
* point to NOAA-EMC fv3 repo
…0/07/21) (ufs-community#173)

- new regression test configurations for `fv3_ccpp_rrfs_v1beta` and `fv3_ccpp_rrfs_v1beta_debug` (required for SRW App release)
- added `fv3_ccpp_rrfs_v1beta` test to various `rt*.conf` files; note: not adding `fv3_ccpp_rrfs_v1beta_debug` because this test crashes on hera with both Intel and GNU (it runs on Cheyenne with Intel but not with GNU) - see issue ufs-community#174
- moved the flake regression test in `rt_gnu.conf` from the DEBUG to the PROD section (this was a mistake not caught in the previous PR when flake was merged)
- remove all `tests/Compile*.log` log files
Top level CMakeLists.txt and CMakeLists.txt file in FV3 have been changed (and cleaned up) to use imported targets (mostly for NCEPLIBS).
Remove *.appBuilder compsets and log.
Remove outdated doc directory
…on Tier-2 platforms (ufs-community#177)

- updates the modulefiles and cmake configurations for Tier-2 platforms in order to work with the new NCEP libraries (imported targets)
- cleans up cmake configurations (AVX2 flags, DEBUG and REPRO mode) and declares SIMDMULTIARCH flags as a real alternative to the AVX2 flags
- removes old build targets cheyenne.intel-mpi and supermuc-phase2.intel
- clean up of build.sh
Downgrade FMS to 2019.01.03.
Small changes in cmake build system due to FMS version change.
Improve error handling in rt.sh when ecflow is used. Scripts cleanup and refactoring.
…-community#171)

* Update CMakeLists.txt to build stochastic_physics
* Update gnumake build to include stochastic_physics wrapper
* Update gnumake build config for macosx.gnu to work with new NCEPLIBS and downgraded FMS
* Update submodule pointer for fv3atm
* Regression test logs for orion.intel, hera.intel, hera.gnu
* Rename rt_stampede_cheyenne.conf to rt_stampede.conf, add rt_cheyenne.conf for cheyenne.intel from NCAR dtc/develop branch
* Regression test logs for wcoss_cray and wcoss_dell_p3
* Revert stochastic_physics change in .gitmodules and update submodule pointer for stochastic_physics
* Revert change to .gitmodules and update submodule pointer for fv3atm
* Revert unintential addition of tests/rt_stochy_ca.conf
@climbfuji
Copy link
Copy Markdown
Author

@jkhender for your information - will be merging in a couple of hours

@climbfuji
Copy link
Copy Markdown
Author

Regression testing on hera.intel using rt.conf and on hera.gnu using rt_gnu.conf against existing baselines: all tests pass.

rt_intel.log
rt_gnu.log

@climbfuji
Copy link
Copy Markdown
Author

Regression testing on hera.intel using rt_ccpp_gsd.conf: first create new baseline (to make sure all tests run to completion), then verify against it (to test run-to-run and restart reproducibility).

rt_ccpp_gsd_create.log
rt_ccpp_gsd_verify.log

@DomHeinzeller DomHeinzeller merged commit ca19b6e into NOAA-GSL:gsd/develop Aug 7, 2020
DomHeinzeller pushed a commit that referenced this pull request Nov 20, 2020
* bring GFS.v16 code changes back to develop branch
* update upp lib in hera module
* use test hpc stack lib
* update upp lib in fv3 for cmake
* update post control files
* update upp lib in hpcstack on hera, orion and dell and in nceplibs on cray
* add change in fv3 dycore #35
* remove duplicate read_data in fms_mod in external_ic.F90
Co-authored-by: Jun Wang <junwang-noaa@users.noreply.github.com>
Co-authored-by: Dom Heinzeller <climbfuji@ymail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants