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.

Sign up for GitHub

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

Fix spatially varying boundary conditions on GPU #2215

Merged
merged 3 commits into from
Oct 16, 2023
Merged

Fix spatially varying boundary conditions on GPU #2215

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2c4a70c
simplify benchmark
simonbyrne Oct 16, 2023
3f05a5c
remove soft_fail, clean up pipelines
simonbyrne Oct 16, 2023
3cd5080
reduce length of run, increase diagnostic freq
simonbyrne Oct 16, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 0 additions & 2 deletions .buildkite/gpu_pipeline/pipeline.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,7 +61,6 @@ steps:
artifact_paths: "gpu_aquaplanet_dyamond/*"
agents:
slurm_gpus: 1
slurm_time: 23:00:00

- label: "moist Held-Suarez"
key: "gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km"
Expand All @@ -74,7 +73,6 @@ steps:
artifact_paths: "gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km/*"
agents:
slurm_gpus: 1
slurm_time: 23:00:00

- label: "dry baroclinic wave - 4 gpus"
key: "target_gpu_implicit_baroclinic_wave_4process"
Expand Down
12 changes: 5 additions & 7 deletions .buildkite/pipeline.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -942,21 +942,19 @@ steps:
slurm_mem: 32G

- label: "GPU: GPU moist Held-Suarez"
command: >
julia --project -e 'using CUDA; CUDA.versioninfo()'

julia --color=yes --project=examples examples/hybrid/driver.jl
--config_file $GPU_CONFIG_PATH/central_gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km.yml
command:
- julia --project -e 'using CUDA; CUDA.versioninfo()'
- >
julia --color=yes --project=examples examples/hybrid/driver.jl
--config_file $GPU_CONFIG_PATH/central_gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km.yml
artifact_paths: "central_gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km/*"
agents:
slurm_gpus: 1
soft_fail: true

- label: "GPU: gpu_aquaplanet_dyamond"
command:
- mkdir -p gpu_aquaplanet_dyamond
- >
nsys profile --trace=nvtx,cuda --output=gpu_aquaplanet_dyamond/report
julia --color=yes --project=examples examples/hybrid/driver.jl
--config_file ${GPU_CONFIG_PATH}gpu_aquaplanet_dyamond.yml
artifact_paths: "gpu_aquaplanet_dyamond/*"
Expand Down
4 changes: 2 additions & 2 deletions config/gpu_configs/central_gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
dt_save_to_disk: "10days"
dt_save_to_disk: "1days"
dt: "150secs"
t_end: "300days"
t_end: "2days"
h_elem: 16
z_elem: 63
dz_bottom: 30.0
Expand Down
8 changes: 4 additions & 4 deletions config/gpu_configs/gpu_hs_rhoe_equilmoist_nz63_0M_55km_rs35km.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,12 +1,12 @@
dt_save_to_disk: "10days"
dt: "150secs"
t_end: "300days"
h_elem: 16
dt: "100secs"
t_end: "1days"
Copy link
Member

Choose a reason for hiding this comment

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

I suggest we go at least 10 days, so that the callbacks are exercised.

Copy link
Member Author

Choose a reason for hiding this comment

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

what callbacks need to be exercised?

Copy link
Member

Choose a reason for hiding this comment

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

The default are probably fine, which will probably include a bunch of model-specific pieces (e.g., diagnostics for moisture etc.).

Copy link
Member Author

Choose a reason for hiding this comment

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

can we just reduce the dt_save_to_disk?

Copy link
Member Author

Choose a reason for hiding this comment

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

okay, but that doesn't need to be in the perf target. We should make sure it is in the CI though.

Copy link
Member

Choose a reason for hiding this comment

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

can we just reduce the dt_save_to_disk?

Yeah, that works too

h_elem: 30
z_elem: 63
dz_bottom: 30.0
dz_top: 3000.0
z_max: 55000.0
kappa_4: 2.0e16
kappa_4: 1.0e15
vert_diff: "true"
moist: "equil"
precip_model: "0M"
Expand Down
Loading