Regrid truncated data

.buildkite/longruns/pipeline.yml

@@ -1,7 +1,7 @@
 agents:
   queue: new-central
   slurm_time: 24:00:00
-  modules: climacommon/2024_03_18
+  modules: climacommon/2024_04_05
 
 env:
   JULIA_LOAD_PATH: "${JULIA_LOAD_PATH}:${BUILDKITE_BUILD_CHECKOUT_PATH}/.buildkite"
@@ -13,7 +13,6 @@ env:
   SLURM_KILL_BAD_EXIT: 1
 
   CONFIG_PATH: "config/longrun_configs"
-  PERF_CONFIG_PATH: "config/perf_configs"
 
 timeout_in_minutes: 1440
 
@@ -34,11 +33,6 @@ steps:
       - "julia --project=artifacts -e 'using Pkg; Pkg.status()'"
       - "julia --project=artifacts artifacts/download_artifacts.jl"
 
-      - echo "--- Instantiate perf env"
-      - "julia --project=perf/ -e 'using Pkg; Pkg.instantiate(;verbose=true)'"
-      - "julia --project=perf/ -e 'using Pkg; Pkg.precompile()'"
-      - "julia --project=perf/ -e 'using Pkg; Pkg.status()'"
-
     agents:
       slurm_cpus_per_task: 8
     env:
@@ -75,6 +69,7 @@ steps:
       slurm_mem: 20GB
       slurm_gpus: 1
       modules: common
+    soft_fail: true
 
   - group: "Coupler integration and conservation tests"
 
@@ -91,6 +86,7 @@ steps:
           slurm_ntasks_per_node: 1
           slurm_nodes: 1
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "Slabplanet_aqua: couple"
         key: "slabplanet_aqua_atmos_sf_couple" # SF at each Atmos stage, coupling, prescribed SST from coupler - identical results to the above confirm 1) initial conditions in Atmos are unchanged compared to the slab, 2) coupling not introducing variability when constant surface
@@ -102,6 +98,7 @@ steps:
           slurm_ntasks_per_node: 1
           slurm_nodes: 1
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "Slabplanet_aqua: coupler fluxes"
         key: "slabplanet_aqua_coupler_sf" # SF at each coupler timestep, constant ocean - comparing to the above runs, this tests the sensitivity of less frequent flux calculation
@@ -113,6 +110,7 @@ steps:
           slurm_ntasks_per_node: 1
           slurm_nodes: 1
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "Slabplanet_aqua: coupler fluxes, evolving ocean"
         key: "slabplanet_aqua_coupler_sf_evolve_ocn" # SF at each coupler timestep, evolving ocean - comparing to the above run, tests the sensitivity of evolving ocean
@@ -124,6 +122,7 @@ steps:
           slurm_ntasks_per_node: 1
           slurm_nodes: 1
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "Slabplanet_terra: coupler fluxes, evolving bucket"
         key: "slabplanet_terra" # SF at each coupler timestep, evolving ocean - comparing to the above run, tests the sensitivity of evolving bucket
@@ -135,6 +134,7 @@ steps:
           slurm_ntasks_per_node: 1
           slurm_nodes: 1
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "Slabplanet: coupler fluxes, evolving ocean and land"
         key: "slabplanet_coupler_sf_evolve_ocn"
@@ -146,8 +146,9 @@ steps:
           slurm_ntasks_per_node: 1
           slurm_nodes: 1
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
-  - group: "Current target tests: idealized surface"
+  - group: "Current target tests: idealized surfaces"
 
     steps:
 
@@ -162,6 +163,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "TARGET IDEALIZED: new target aqua - fixed ocean T, nocouple, atmos flux calc"
         key: "slabplanet_aqua_target_nocouple"
@@ -174,6 +176,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "TARGET IDEALIZED: new target aqua - fixed ocean T, coupler flux calc"
         key: "slabplanet_aqua_target"
@@ -186,6 +189,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "TARGET IDEALIZED: new target aqua - evolving slab ocean T"
         key: "slabplanet_aqua_target_evolve_ocn"
@@ -198,6 +202,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "TARGET IDEALIZED: new target slab - fixed ocean T, bucket"
         key: "slabplanet_target"
@@ -210,6 +215,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "TARGET IDEALIZED: new target slab - evolving slab ocean T, bucket"
         key: "slabplanet_target_evolve_ocn"
@@ -222,9 +228,10 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
 
-  - group: "Current target tests: AMIP surface"
+  - group: "Current target tests: AMIP surfaces"
 
     steps:
 
@@ -239,6 +246,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "MPI AMIP FINE: new target amip: topo"
         key: "amip_target_topo"
@@ -251,6 +259,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 16G
+        soft_fail: true
 
       - label: "MPI AMIP FINE: new target amip: topo + diagedmf"
         key: "amip_target_topo_diagedmf"
@@ -263,6 +272,7 @@ steps:
           slurm_ntasks_per_node: 16
           slurm_nodes: 4
           slurm_mem_per_cpu: 20G
+        soft_fail: true
 
   - group: "Current target tests on GPU: AMIP surface"
 
@@ -275,6 +285,7 @@ steps:
         agents:
           slurm_gpus: 1
           slurm_mem: 16GB
+        soft_fail: true
 
       - label: "GPU AMIP FINE: new target amip: topo + diagedmf"
         key: "gpu_amip_target_topo_diagedmf"
@@ -283,105 +294,13 @@ steps:
         agents:
           slurm_gpus: 1
           slurm_mem: 16GB
+        soft_fail: true
 
-  - group: "Other AMIP targets"
-
-    steps:
-
-      # DYAMOND AMIP: 1 day (convection resolving)
-      - label: "MPI AMIP SUPERFINE: dyamond_target"
-        key: "dyamond_target"
-        command: "srun julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/dyamond_target.yml"
-        artifact_paths: "experiments/AMIP/output/amip/dyamond_target_artifacts/*"
-        env:
-          CLIMACORE_DISTRIBUTED: "MPI"
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 16
-          slurm_nodes: 4
-          slurm_mem_per_cpu: 16G
-
-      # mid-resolution AMIP: MPI performance scaling (10 days)
-      - label: "MPI AMIP FINE: n64"
-        key: "mpi_amip_fine_n64"
-        command: "srun julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/amip_n64_shortrun.yml"
-        artifact_paths: "experiments/AMIP/output/amip/amip_n64_shortrun_artifacts/*"
-        env:
-          CLIMACORE_DISTRIBUTED: "MPI"
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 16
-          slurm_nodes: 4
-          slurm_mem_per_cpu: 16G
-
-      - label: "MPI AMIP FINE: n32"
-        key: "mpi_amip_fine_n32"
-        command: "srun julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/amip_n32_shortrun.yml"
-        artifact_paths: "experiments/AMIP/output/amip/amip_n32_shortrun_artifacts/*"
-        env:
-          CLIMACORE_DISTRIBUTED: "MPI"
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 8
-          slurm_nodes: 4
-          slurm_mem_per_cpu: 16G
-
-      - label: "MPI AMIP FINE: n8"
-        key: "mpi_amip_fine_n8"
-        command: "srun julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/amip_n8_shortrun.yml"
-        artifact_paths: "experiments/AMIP/output/amip/amip_n8_shortrun_artifacts/*"
-        env:
-          CLIMACORE_DISTRIBUTED: "MPI"
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 8
-          slurm_nodes: 1
-          slurm_mem_per_cpu: 16G
-
-      - label: "MPI AMIP FINE: n2" # 10d take 21h, so reducing to 1d
-        key: "mpi_amip_fine_n2"
-        command: "srun julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/amip_n2_shortrun.yml"
-        artifact_paths: "experiments/AMIP/output/amip/amip_n2_shortrun_artifacts/*"
-        env:
-          CLIMACORE_DISTRIBUTED: "MPI"
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 2
-          slurm_nodes: 1
-          slurm_mem_per_cpu: 16G
-
-      - label: "MPI AMIP FINE: n1" # also reported by longruns with a flame graph; 10d take 21h, so reducing to 1d
-        key: "mpi_amip_fine_n1"
-        command: "julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/amip_n1_shortrun.yml"
-        artifact_paths: "experiments/AMIP/output/amip/amip_n1_shortrun_artifacts/*"
-        env:
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 1
-          slurm_nodes: 1
-          slurm_mem_per_cpu: 16G
-
-      - label: "MPI AMIP FINE: n1 no couple" # sim time = Δt_cpl (~ benchmarking with standalone models)
-        key: "mpi_amip_fine_n1_nocouple"
-        command: "julia --color=yes --project=experiments/AMIP/ experiments/AMIP/coupler_driver.jl --config_file $CONFIG_PATH/amip_n1_shortrun_nocouple.yml"
-        artifact_paths: "experiments/AMIP/output/amip/amip_n1_shortrun_nocouple_artifacts/*"
-        env:
-          BUILD_HISTORY_HANDLE: ""
-        agents:
-          slurm_ntasks_per_node: 1
-          slurm_nodes: 1
-          slurm_mem_per_cpu: 16G
+  - wait
 
-      # mpi_amip_fine_n1 flame graph report (NB: arguments passed from the ci pipeline.yml)
-      - label: ":rocket: performance: flame graph diff: perf_target_amip_n1_shortrun"
-        command: "julia --color=yes --project=perf perf/flame_diff.jl --config_file $PERF_CONFIG_PATH/perf_diff_target_amip_n1_shortrun.yml"
-        artifact_paths: "perf/output/perf_diff_target_amip_n1_shortrun/*"
-        agents:
-          slurm_ntasks_per_node: 1
-          slurm_nodes: 1
-          slurm_mem_per_cpu: 16G
+  - group: "Job analysis and reporting"
 
-      - wait
+    steps:
 
       # plot job performance history
       - label: ":chart_with_downwards_trend: build history"
@@ -394,21 +313,20 @@ steps:
 
       - label: ":envelope: Slack report: build_history"
         command:
-          - slack-upload -c "#coupler-report" -f build_history.html -m html -n build_history -x "Overall job performance"
+          - |
+            slack-upload -c "#coupler-report" -f build_history.html -m html -n build_history -x ":rocket: Interactive overall job performance history (download the attached file and view in browser) :rocket:"
 
       - label: ":envelope: Slack report: Slabplanet"
         command:
-          - slack-upload -c "#coupler-report" -f experiments/AMIP/output/slabplanet/slabplanet_coupler_sf_evolve_ocn_artifacts/total_energy_log_bucket.png -m png -n slab_coarse_log -x "Slabplanet energy conservation (log error)"
           - slack-upload -c "#coupler-report" -f experiments/AMIP/output/slabplanet/slabplanet_coupler_sf_evolve_ocn_artifacts/total_energy_bucket.png -m png -n slab_coarse -x "Slabplanet energy conservation"
-          - slack-upload -c "#coupler-report" -f experiments/AMIP/output/slabplanet/slabplanet_coupler_sf_evolve_ocn_artifacts/total_water_log_bucket.png -m png -n slab_coarse_w_log -x "Slabplanet water conservation (log error)"
           - slack-upload -c "#coupler-report" -f experiments/AMIP/output/slabplanet/slabplanet_coupler_sf_evolve_ocn_artifacts/total_water_bucket.png -m png -n slab_coarse_w -x "Slabplanet water conservation"
 
       - label: ":envelope: Slack report: target AMIP"
         command:
-          - slack-upload -c "#coupler-report" -f experiments/AMIP/output/amip/amip_target_artifacts/amip_paperplots.png -m png -n amip_fine -x "AMIP Target Longrun"
-          - slack-upload -c "#coupler-report" -f experiments/AMIP/output/amip/amip_target_artifacts/biases.png -m png -n amip_fine -x "AMIP Target Longrun"
+          - slack-upload -c "#coupler-report" -f experiments/AMIP/output/amip/amip_target_topo_diagedmf_artifacts/amip_ncep.png -m png -n amip_fine -x "Target AMIP v NCEP Last Month Mean"
+          - |
+            find experiments/AMIP/output/amip/amip_target_topo_diagedmf_artifacts/ -type f -name 'bias*.png' -print0 | while IFS= read -r -d '' file; do
+              slack-upload -c "#coupler-report" -f "$$file" -m png -n "$$(basename "$$file" .png)" -x "$$(basename "$$file" .png)"
+            done
+
 
-      - label: ":envelope: Slack report: Flame Diff"
-        command:
-          - slack-upload -c "#coupler-report" -f perf/output/perf_diff_target_amip_n1_shortrun/flame_diff.html -m png -n amip_fine_flamegraphdiff -x "AMIP Longrun FlameGraphDiff"
-          - slack-upload -c "#coupler-report" -f perf/output/perf_diff_target_amip_n1_shortrun/flame_diff_self_count.html -m png -n amip_fine_flamegraphdiffself -x "AMIP Longrun FlameGraphDiffSelf"

.buildkite/pipeline.yml

@@ -1,7 +1,7 @@
 agents:
   queue: new-central
   slurm_time: 4:00:00
-  modules: climacommon/2024_03_18
+  modules: climacommon/2024_04_05
 
 env:
   JULIA_LOAD_PATH: "${JULIA_LOAD_PATH}:${BUILDKITE_BUILD_CHECKOUT_PATH}/.buildkite"

Project.toml

@@ -1,7 +1,7 @@
 name = "ClimaCoupler"
 uuid = "4ade58fe-a8da-486c-bd89-46df092ec0c7"
 authors = ["CliMA Contributors <clima-software@caltech.edu>"]
-version = "0.1.0"
+version = "0.0.1"
 
 [deps]
 ClimaAtmos = "b2c96348-7fb7-4fe0-8da9-78d88439e717"
@@ -34,7 +34,7 @@ Dates = "1"
 DocStringExtensions = "0.8, 0.9"
 Insolation = "0.9.2"
 JLD2 = "0.4"
-NCDatasets = "0.11, 0.12, 0.13, =0.14.1"
+NCDatasets = "0.11, 0.12, 0.13, 0.14.2"
 Plots = "1.39.0"
 SciMLBase = "1, 2"
 StaticArrays = "1"

artifacts/artifact_funcs.jl

@@ -1,4 +1,5 @@
 import ArtifactWrappers as AW
+#using NCDatasets
 
 function sst_dataset_path()
     sst_dataset = AW.ArtifactWrapper(
@@ -59,3 +60,20 @@ function pr_obs_data_path()
     )
     return AW.get_data_folder(pr_obs_data)
 end
+
+"""
+    artifact_data(datapath_full, name; datapath_trunc, date0, time_start, time_end, comms_ctx)
+
+Truncates given data set, and constructs a new dataset containing only the dates that are used in the simulation
+"""
+function artifact_data(datapath_full, name)
+    datafile_truncated = joinpath(datapath_full, string(name, ".nc"))
+    return datafile_truncated
+end
+
+function artifact_data(datapath_full, name, datapath_trunc, date0, time_start, time_end, comms_ctx)
+    datafile = joinpath(datapath_full, string(name, ".nc"))
+    datafile_truncated =
+        Regridder.truncate_dataset(datafile, name, datapath_trunc, date0, time_start, time_end, comms_ctx)
+    return datafile_truncated
+end

config/model_configs/slabplanet_atmos_diags.yml

@@ -1,5 +1,6 @@
 anim: true
 apply_limiter: false
+ci_plots: true
 dt: "200secs"
 dt_cpl: 200
 dt_save_to_sol: "9days"
@@ -9,11 +10,13 @@ job_id: "slabplanet_atmos_diags"
 mode_name: "slabplanet"
 moist: "equil"
 mono_surface: true
+output_default_diagnostics: false
 precip_model: "0M"
 rad: "gray"
 run_name: "slabplanet_atmos_diags"
 t_end: "10days"
 vert_diff: "true"
 diagnostics:
-  - short_name: [mse, lr, ediff]
+  - short_name: [mse, lr, ediff, ts]
+    reduction_time: average
     period: 1days

config/model_configs/slabplanet_default.yml

@@ -14,3 +14,8 @@ rad: "gray"
 run_name: "slabplanet_default"
 t_end: "10days"
 vert_diff: "true"
+output_default_diagnostics: false
+diagnostics:
+ - short_name: [mse, lr, ediff, hfes, evspsbl, ts]
+   period: 1days
+