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

add target AMIP runs on GPU #632

Merged
merged 1 commit into from
Mar 2, 2024
Merged

add target AMIP runs on GPU #632

merged 1 commit into from
Mar 2, 2024

Conversation

juliasloan25
Copy link
Member

@juliasloan25 juliasloan25 commented Feb 17, 2024

Purpose

closes #630

Content

  • use ClimaAtmos v0.21.0 for gpu-compatible topography
  • add amip topography longrun
    • following config/longrun_configs/amip_target_topo.yml
    • set anim:false (not gpu-compatible)
  • amip topography + diagnostic edmf longrun
    • following config/longrun_configs/amip_target_topo_diagedmf.yml
    • set anim:false (not gpu-compatible)
  • add shortruns for both cases (decrease t_end to 1day)
    • I think we should keep these around with shorter t_end (200secs?) to make sure any future changes don't break these cases
  • rename config/model_configs/gpu_target_amip_albedo_function.yml -> config/model_configs/gpu_amip_albedo_function.yml to reduce confusion with target GPU AMIP runs
  • remove scalar indexing in atmos update_field! method for co2 (replace (...)[1] with mean(...))

@juliasloan25 juliasloan25 force-pushed the js/gpu-amip-target branch 2 times, most recently from 09828af to 756232a Compare February 17, 2024 01:14
@juliasloan25 juliasloan25 force-pushed the js/gpu-amip-target branch 3 times, most recently from 94336e4 to 2ea7a56 Compare March 1, 2024 00:44
@juliasloan25 juliasloan25 force-pushed the js/gpu-amip-target branch 2 times, most recently from 58fea36 to 09f04e3 Compare March 1, 2024 05:07
Copy link
Collaborator

@LenkaNovak LenkaNovak left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thank you, @juliasloan25. Just a couple of comments. It would be good to increase the t_ends in the long runs, but we can do that later. I've triggered a long run here to see where we get, but we can merge this PR for now.

.buildkite/longruns/pipeline.yml Outdated Show resolved Hide resolved
.buildkite/pipeline.yml Show resolved Hide resolved
.buildkite/pipeline.yml Outdated Show resolved Hide resolved
.buildkite/pipeline.yml Outdated Show resolved Hide resolved
@juliasloan25
Copy link
Member Author

LGTM, thank you, @juliasloan25. Just a couple of comments. It would be good to increase the t_ends in the long runs, but we can do that later. I've triggered a long run here to see where we get, but we can merge this PR for now.

that longrun is actually for a different branch where I'm trying DYAMOND on GPU, but I started a longrun for this branch here

@juliasloan25 juliasloan25 merged commit 385ee23 into main Mar 2, 2024
6 of 7 checks passed
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.

run target AMIP runs on GPU
2 participants