Skip to content
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

Rework physics versions used in our System tests (least clm4_5, midlevel clm5_0, and most clm6_0) #2404

Open
ekluzek opened this issue Mar 6, 2024 · 9 comments
Labels
code health improving internal code structure to make easier to maintain (sustainability) next this should get some attention in the next week or two. Normally each Thursday SE meeting. priority: high High priority to fix/merge soon, e.g., because it is a problem in important configurations testing additions or changes to tests

Comments

@ekluzek
Copy link
Collaborator

ekluzek commented Mar 6, 2024

With #2378 coming in we need to look at our test lists and revamp them in view of the CESM3 release. Mainly I think most tests should be clm6_0, but some still for clm5_0 and clm4_5.

With #2379 we also should have minimal testing for clm5_1. But, we can only completely remove it when CAM and CESM have moved to remove clm5_1 in their compsets.

Maybe it's time to drop some of the clm4_5 testing with a new physics version available? I think the testing should scale up from clm4_5 with more for clm5_0 and then the most for clm6_0.

@ekluzek ekluzek added testing additions or changes to tests next this should get some attention in the next week or two. Normally each Thursday SE meeting. labels Mar 6, 2024
@ekluzek ekluzek added this to the CESM3 milestone Mar 6, 2024
@ekluzek
Copy link
Collaborator Author

ekluzek commented Mar 7, 2024

There are also some long standing testing issues that could be addressed with this:

#47
#275
#615
#1186
#1242
#1341
#1584
#1859
#2078
#2151

@ekluzek
Copy link
Collaborator Author

ekluzek commented Mar 7, 2024

This came up because I'm working on #2378 and I'll need to do something with clm5_1 and clm6_0 testing for it. From scoping this out, I think this is more important to do some careful thinking about and designing what we want it to look like. So I'll make this a post CTSM5.2 issue.

@ekluzek
Copy link
Collaborator Author

ekluzek commented Mar 7, 2024

Another complexity is that we need to add a LND_TUNING_MODE for clm6_0_cam7.0. And the same way that we should do most testing for clm6_0, we should do the most for clm6_0_cam7.0 LND_TUNING_MODE.

@ekluzek
Copy link
Collaborator Author

ekluzek commented Mar 7, 2024

It looks like the only testing that CAM and CESM have with clm5_1 are for the ne30np4.pg3 grid for SMS and ERP, on derecho with debug. So the clm5_1 testing can be very limited.

@ekluzek ekluzek removed the next this should get some attention in the next week or two. Normally each Thursday SE meeting. label Mar 14, 2024
@samsrabin
Copy link
Collaborator

samsrabin commented Mar 14, 2024

I will create a Project called "Revamp testing in preparation for the CESM3 release" with this and other issues in it.

@samsrabin
Copy link
Collaborator

Added Project 30.

@ekluzek Do you want to change access permissions so that anyone can view it? I don't have permissions to do that (even though I'm an admin of the project, which is annoying).

It might also be good to rename this issue to something more specific, like "Rework CTSM versions used in tests."

@ekluzek
Copy link
Collaborator Author

ekluzek commented Mar 14, 2024

@samsrabin thanks for pointing that out. Since this is ESCOMP level I've asked @billsacks to update the privacy level of the project board. Only ESCOMP admin's can do that.

@wwieder wwieder moved this to Back Burner (or lower priority) in CTSM-CLM6 development highlights Jun 17, 2024
@ekluzek ekluzek changed the title Revamp testing in preparation for the CESM3 release Rework physics versions used in our System tests (least clm4_5, midlevel clm4_5, and most clm6_0) Feb 6, 2025
@ekluzek
Copy link
Collaborator Author

ekluzek commented Feb 6, 2025

#2954 also relates to this, as the compsets that we keep should only be the ones that we test. And we should remove the compsets that we remove the tests for. I updated the title to what @samsrabin suggested.

@ekluzek
Copy link
Collaborator Author

ekluzek commented Feb 6, 2025

Here's the list of tests for the different physics versions in ctsm5.3.021:

(ctsm_pylib) main_dev/cime_config> grep Clm45 testdefs/testlist_clm.xml | grep '<test ' | wc
     23     115    2425
(ctsm_pylib) main_dev/cime_config> grep Clm50 testdefs/testlist_clm.xml | grep '<test ' | wc
    220    1096   23923
(ctsm_pylib) main_dev/cime_config> grep Clm60 testdefs/testlist_clm.xml | grep '<test ' | wc
    147     732   16638

Note, that the MOST are for clm5_0 rather than clm6_0. The scale should be something more like...

Clm60 -- 250
Clm50 -- 125
Clm45 -- 15

@ekluzek ekluzek changed the title Rework physics versions used in our System tests (least clm4_5, midlevel clm4_5, and most clm6_0) Rework physics versions used in our System tests (least clm4_5, midlevel clm5_0, and most clm6_0) Feb 6, 2025
@ekluzek ekluzek added priority: high High priority to fix/merge soon, e.g., because it is a problem in important configurations code health improving internal code structure to make easier to maintain (sustainability) next this should get some attention in the next week or two. Normally each Thursday SE meeting. labels Feb 6, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
code health improving internal code structure to make easier to maintain (sustainability) next this should get some attention in the next week or two. Normally each Thursday SE meeting. priority: high High priority to fix/merge soon, e.g., because it is a problem in important configurations testing additions or changes to tests
Projects
Status: Back Burner (or lower priority)
Status: Todo
Development

No branches or pull requests

2 participants