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

Scaling pipeline for moist baroclinic wave #3550

Merged
merged 1 commit into from
Feb 12, 2025
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
69 changes: 69 additions & 0 deletions .buildkite/cpu_scaling_pipeline/generate_pipeline.jl
Original file line number Diff line number Diff line change
@@ -0,0 +1,69 @@
# Generate a YAML pipeline to run CPU scaling tests on buildkite
# To run: `julia --project=.buildkite .buildkite/cpu_scaling_pipeline/generate_pipeline.jl`
# nodes = (1, 2, 4, 8, 16, 32)
# helems = (30, 42, 60, 84, 120, 170)
nodes = (1, 2, 4)
helems = (30, 42, 60)

import YAML

agents = Dict("modules" => "climacommon/2024_12_16", "queue" => "new-central")
env = Dict(
"SLURM_KILL_BAD_EXIT" => 1,
"OPENBLAS_NUM_THREADS" => 1,
"JULIA_NVTX_CALLBACKS" => "gc",
"JULIA_LOAD_PATH" => "\${JULIA_LOAD_PATH}:\${BUILDKITE_BUILD_CHECKOUT_PATH}/.buildkite",
)

CPU_TYPE = "icelake"
MBW_SCALING_CONFIG_PATH = "config/mbw_scaling_configs"

init_step = Dict(
"label" => "init :computer:",
"key" => "init_cpu_env",
"agents" => Dict("slurm_cpus_per_task" => 8),
"command" => [
"""echo --- Instantiate project""",
"julia --project -e 'using Pkg; Pkg.instantiate(;verbose=true)'",
"julia --project -e 'using Pkg; Pkg.precompile(); Pkg.status()'",
"""echo --- Instantiate .buildkite""",
"julia --project=.buildkite -e 'using Pkg; Pkg.instantiate(;verbose=true)'",
"julia --project=.buildkite -e 'using Pkg; Pkg.precompile(); Pkg.status()'",
],
"env" => Dict(
"JULIA_NUM_PRECOMPILE_TASKS" => 8,
"JULIA_MAX_NUM_PRECOMPILE_FILES" => 50,
),
)

function generate_step(nodes::Int, helems::Int)
return Dict(
"label" => ":computer: $nodes node, 16 processes per node, helem = $helems",
"command" => "srun julia --color=yes --project=.buildkite .buildkite/ci_driver.jl --config_file $MBW_SCALING_CONFIG_PATH/moist_baroclinic_wave_helem_$(helems)_0M_ws.yml --job_id moist_baroclinic_wave_helem_$(helems)_0M_ws",
"artifact_paths" => "moist_baroclinic_wave_helem_$(helems)_0M_ws/output_active/*",
"agents" => Dict(
"slurm_constraint" => CPU_TYPE,
"queue" => "new-central",
"slurm_nodes" => nodes,
"slurm_tasks_per_node" => 16,
"slurm_cpus_per_task" => 1,
"slurm_mem" => 0,
"slurm_time" => "1:00:00",
"slurm_exclusive" => true,
),
)
end

pipeline = Dict(
"agents" => agents,
"env" => env,
"steps" => [
init_step,
"wait",
Dict(
"group" => "Moist Baroclinic Wave, weak scaling",
"steps" => [generate_step.(nodes, helems)...],
),
],
)
YAML.write_file(".buildkite/cpu_scaling_pipeline/pipeline.yml", pipeline)
62 changes: 62 additions & 0 deletions .buildkite/cpu_scaling_pipeline/pipeline.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,62 @@
agents:
modules: "climacommon/2024_12_16"
queue: "new-central"
steps:
- label: "init :computer:"
key: "init_cpu_env"
agents:
slurm_cpus_per_task: 8
command:
- "echo --- Instantiate project"
- "julia --project -e 'using Pkg; Pkg.instantiate(;verbose=true)'"
- "julia --project -e 'using Pkg; Pkg.precompile(); Pkg.status()'"
- "echo --- Instantiate .buildkite"
- "julia --project=.buildkite -e 'using Pkg; Pkg.instantiate(;verbose=true)'"
- "julia --project=.buildkite -e 'using Pkg; Pkg.precompile(); Pkg.status()'"
env:
JULIA_NUM_PRECOMPILE_TASKS: 8
JULIA_MAX_NUM_PRECOMPILE_FILES: 50
- "wait"
- steps:
- label: ":computer: 1 node, 16 processes per node, helem = 30"
agents:
slurm_time: "1:00:00"
slurm_cpus_per_task: 1
slurm_tasks_per_node: 16
queue: "new-central"
slurm_nodes: 1
slurm_exclusive: true
slurm_mem: 0
slurm_constraint: "icelake"
command: "srun julia --color=yes --project=.buildkite .buildkite/ci_driver.jl --config_file config/mbw_scaling_configs/moist_baroclinic_wave_helem_30_0M_ws.yml --job_id moist_baroclinic_wave_helem_30_0M_ws"
artifact_paths: "moist_baroclinic_wave_helem_30_0M_ws/output_active/*"
- label: ":computer: 2 node, 16 processes per node, helem = 42"
agents:
slurm_time: "1:00:00"
slurm_cpus_per_task: 1
slurm_tasks_per_node: 16
queue: "new-central"
slurm_nodes: 2
slurm_exclusive: true
slurm_mem: 0
slurm_constraint: "icelake"
command: "srun julia --color=yes --project=.buildkite .buildkite/ci_driver.jl --config_file config/mbw_scaling_configs/moist_baroclinic_wave_helem_42_0M_ws.yml --job_id moist_baroclinic_wave_helem_42_0M_ws"
artifact_paths: "moist_baroclinic_wave_helem_42_0M_ws/output_active/*"
- label: ":computer: 4 node, 16 processes per node, helem = 60"
agents:
slurm_time: "1:00:00"
slurm_cpus_per_task: 1
slurm_tasks_per_node: 16
queue: "new-central"
slurm_nodes: 4
slurm_exclusive: true
slurm_mem: 0
slurm_constraint: "icelake"
command: "srun julia --color=yes --project=.buildkite .buildkite/ci_driver.jl --config_file config/mbw_scaling_configs/moist_baroclinic_wave_helem_60_0M_ws.yml --job_id moist_baroclinic_wave_helem_60_0M_ws"
artifact_paths: "moist_baroclinic_wave_helem_60_0M_ws/output_active/*"
group: "Moist Baroclinic Wave, weak scaling"
env:
SLURM_KILL_BAD_EXIT: 1
OPENBLAS_NUM_THREADS: 1
JULIA_NVTX_CALLBACKS: "gc"
JULIA_LOAD_PATH: "${JULIA_LOAD_PATH}:${BUILDKITE_BUILD_CHECKOUT_PATH}/.buildkite"
12 changes: 0 additions & 12 deletions .buildkite/scaling/README.md

This file was deleted.

210 changes: 0 additions & 210 deletions .buildkite/scaling/pipeline.sh

This file was deleted.

Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
h_elem: 120
z_elem: 43
dz_bottom: 30.0
dt: "40secs"
t_end: "1days"
ode_algo: ARS343
initial_condition: "MoistBaroclinicWave"
moist: "equil"
precip_model: "0M"
dt_save_state_to_disk: "Inf"
dt_save_to_sol: "Inf"
output_default_diagnostics: false
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
h_elem: 120
z_elem: 43
dz_bottom: 30.0
dt: "6secs"
t_end: "1days"
ode_algo: ARS343
initial_condition: "MoistBaroclinicWave"
moist: "equil"
precip_model: "0M"
dt_save_state_to_disk: "Inf"
dt_save_to_sol: "Inf"
output_default_diagnostics: false
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
h_elem: 170
z_elem: 43
dz_bottom: 30.0
dt: "6secs"
t_end: "1days"
ode_algo: ARS343
initial_condition: "MoistBaroclinicWave"
moist: "equil"
precip_model: "0M"
dt_save_state_to_disk: "Inf"
dt_save_to_sol: "Inf"
output_default_diagnostics: false
Loading
Loading