Aplus3

example-configs/density_skj_fat/Snakefile

@@ -3,17 +3,14 @@ import csv
 import pandas as pd
 import matplotlib.pyplot as plt
 
-# snakemake --jobs 10 --cluster "sbatch -p genoa --mem=20g --nodes=1 --ntasks-per-node={threads} --cpus-per-task=1 --time=01:30:00 --extra-node-info=1:*:1 --distribution=*:block:* --mem-bind=local"
+# snakemake --jobs 10 --cluster "sbatch -p genoa --mem=60g --nodes=1 --ntasks-per-node={threads} --cpus-per-task=1 --time=01:30:00 --extra-node-info=1:*:1 --distribution=*:block:* --mem-bind=local"
 # 
 
 # MPI benchmarking configurations
-#N_VALUES = [2,3,4,5,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,28,29,30,31,32]
-#N_VALUES = [2,3,4,5,6,7,8,10,12,14,16,20,24,28,32,36,40,50]
-N_VALUES = [2,3,4,5,6,7,8,9,10,11,12,14,16,17,18,20,22,24,26,28,32,36,40,44,51]
-#N_VALUES = [10, 20, ]
-REPEATS = range(0,6)
-#HOSTS = ["milan", "genoa"]
+PROCS = [3, 11, 21, 31, 41, 51, 61, 71, 81, 91, 101]
+REPEATS = range(0,3)
 HOSTS = ["genoa",]
+CASES = ["aplus", "aplus3"]
 
 # -----------------------------
 # Top-level workflow
@@ -27,20 +24,17 @@ rule all:
 # -----------------------------
 rule run_mpi:
     output:
-        main="results/run_{n}_{r}_{h}-main.out",
-        best="results/run_{n}_{r}_{h}-best.out"
+        exp="results/run_{n}_{r}_{h}_{c}.out",
     params:
-        exe_main="../../bin/seapodym_cohort-main",
-        exe_best="../../bin/seapodym_cohort-best",
-        parfile="skj_fat.xml"
+        parfile="../density_skj_fat/skj_Fat.xml"
     threads:
         lambda wc: int(wc.n)
     shell:
         r"""
         set -x
-        cd ~/nesi99999/seapodym-codebase-test_shm/
+        cd ~/nesi99999/seapodym-codebase-exp/
 
-        source sourceme-main.sh
+        source sourceme-exp.sh
 
         cd example-configs/density_skj_fat
         mkdir -p results
@@ -51,19 +45,14 @@ rule run_mpi:
         rm -rf $rundir
         mkdir $rundir
         cd $rundir
-        # copy the input file
-        cp ../density_skj_fat/{params.parfile} .
-
-
-        # run main
-        echo "jobid=$SLURM_JOB_ID" > ../density_skj_fat/{output.main}
-        srun --distribution=block:block --cpu-bind=cores --cpus-per-task=1 \
-             {params.exe_main} -s {params.parfile} >> ../density_skj_fat/{output.main} 2>&1
+
+        # create directories
+        mkdir -p output/output_F0/
 
-        # run best
-        echo "jobid=$SLURM_JOB_ID" > ../density_skj_fat/{output.best}
-        srun --distribution=block:block --cpu-bind=cores --cpus-per-task=1 \
-             {params.exe_best} -s {params.parfile} >> ../density_skj_fat/{output.best} 2>&1
+        # run 
+        echo "jobid=$SLURM_JOB_ID" > ../density_skj_fat/{output.exp}
+        srun --ntasks={threads} --distribution=block:block --cpu-bind=cores --cpus-per-task=1 \
+             ../../bin/seapodym_cohort_{wildcards.c} -s {params.parfile} >> ../density_skj_fat/{output.exp} 2>&1
 
         cd ../
         # clean up
@@ -77,13 +66,12 @@ rule run_mpi:
 # -----------------------------
 rule collect_results:
     input:
-        expand("results/run_{n}_{r}_{h}-main.out", n=N_VALUES, r=REPEATS, h=HOSTS),
-        expand("results/run_{n}_{r}_{h}-best.out", n=N_VALUES, r=REPEATS, h=HOSTS)
+        expand("results/run_{n}_{r}_{h}_{c}.out", n=PROCS, r=REPEATS, h=HOSTS, c=CASES),
     output:
         "results/timings.csv"
     shell:
         r"""
         module load Python
-        python assemble_timings.py -d results -o results/timings.csv
+        python assemble_timings_aplus.py -d results -o results/timings.csv
         """
 

example-configs/density_skj_fat/assemble_timings_aplus.py

@@ -0,0 +1,48 @@
+import re
+import glob
+import defopt
+import csv
+
+
+def main(*, dir : str='results', output : str='results/timings.csv'):
+    """
+    Assemble timings into a csv file
+    :param dir directory
+    """
+    pattern = re.compile(r"time manager\s*=\s*([0-9.]+)")
+    rows = []
+
+    for f in glob.glob(dir + '/run*.out'):
+
+         print(f'working on {f}...')
+
+         parts = f.split("/")[-1].replace(".out","").split("_")
+         n = int(parts[1])
+         r = int(parts[2])
+         host = parts[3]
+         case = parts[4]
+
+         jobid = None
+         manager_time = None
+
+         with open(f) as fh:
+            for line in fh:
+                if line.startswith("jobid="):
+                    jobid = line.split("=")[1].strip()
+                m = pattern.search(line)
+                if m:
+                    manager_time = float(m.group(1))
+                    print(f'manager time = {manager_time}')
+
+            rows.append((n, jobid, host, case, manager_time))
+
+         rows.sort()
+
+         with open(output, "w", newline="") as csvfile:
+            writer = csv.writer(csvfile)
+            writer.writerow(["nprocs","jobid","host","case", "manager_time"])
+            writer.writerows(rows)
+
+if __name__ == '__main__':
+	defopt.run(main)
+

example-configs/density_skj_fat/plot_timings_aplus.py

@@ -0,0 +1,59 @@
+import os
+import pandas as pd
+import matplotlib.pyplot as plt
+import numpy as np
+import defopt
+
+
+def main(*, input_file: str='results/timings.csv', output_file: str='results/timings.png', show: bool=False):
+        """
+        input_file: CSV input file (e.g. results/timings.csv)
+        output_file: PNG output file
+        show: set to True to display
+        """
+
+        # Load CSV
+        df = pd.read_csv(input_file)
+
+        # Ensure manager_time column exists
+        if "manager_time" not in df.columns:
+            print(f"ERROR: 'manager_time' column not found in CSV. Cannot plot.")
+
+        # Drop rows with missing manager_time
+        df = df.dropna(subset=["manager_time"])
+        if df.empty:
+            print(f"ERROR: No valid 'manager_time' values found. Cannot plot.")
+
+        # Plot curves for Milan & Genoa
+        cols = {
+ 		'aplus': 'b', 
+ 		'aplus3': 'm',
+ 		}
+        plt.figure(figsize=(8,6))
+        for case in df["case"].unique():
+            df2 = df[df.case == case]
+            df_min = df2.groupby(["nprocs"])["manager_time"].min().reset_index()
+            df_avg = df2.groupby(["nprocs"])["manager_time"].mean().reset_index()
+            df_max = df2.groupby(["nprocs"])["manager_time"].max().reset_index()
+
+            col = cols[case]
+            plt.loglog(df_min["nprocs"], df_min["manager_time"], col + '-.', label=case + ' min')
+            plt.loglog(df_avg["nprocs"], df_avg["manager_time"], col + '-', label=case + ' avg')
+            plt.loglog(df_max["nprocs"], df_max["manager_time"], col + '--', label=case + ' max')
+
+        plt.loglog([1,100], 1000./np.array([1,100]), 'k-', label='ideal')
+
+        plt.xlabel("Number of processes (nprocs)")
+        plt.ylabel("Manager time s")
+        plt.title("Seapodym cohort timings 1983-2022 skj_Fat.xml A+")
+        plt.xticks(sorted(df["nprocs"].unique()))
+        plt.grid(True)
+        plt.legend()
+        plt.tight_layout()
+        plt.savefig(output_file)
+        if show:
+            plt.show()
+        print(f"plot saved to {output_file}")
+
+if __name__ == '__main__':
+	defopt.run(main)

example-configs/density_skj_fat/rungenoa.sl

@@ -3,13 +3,9 @@
 #SBATCH --time=00:10:00           # walltime (adjust as needed)
 #SBATCH --partition=genoa
 #SBATCH --nodes=1
-#SBATCH --ntasks=24
-#SBATCH --mem-per-cpu=4g
+#SBATCH --ntasks=51
+#SBATCH --mem=25g
 #SBATCH --cpus-per-task=1
-#SBATCH --output=seapodym_%j.out
-#SBATCH --error=seapodym_%j.err
-##SBATCH --profile task
-##SBATCH --acctg-freq=1
 #SBATCH --extra-node-info=1:*:1 
 #SBATCH --distribution=*:block:* 
 #SBATCH --mem-bind=local
@@ -20,4 +16,4 @@ module load gimkl/2020a
 rm log_*.txt
 
 # Run the MPI job
-srun ../../bin/seapodym_cohort -s skj_fat.xml >& results/n${SLURM_NTASKS}.txt
+srun ../../bin/seapodym_cohort -s skj_fat.xml

example-configs/density_skj_fat/skj_fat.xml

@@ -244,7 +244,7 @@
 </life_stage>
 
 <nb_cohort_life_stage>
-        <skj>1 2 47</skj>
+        <skj>1 2 97</skj>
 </nb_cohort_life_stage>
 
 <age_recruit skj="3"/>
@@ -255,23 +255,39 @@
 		  30 30 30 30 30 30 30 30 30 30 30 30 
 		  30 30 30 30 30 30 30 30 30 30 30 30 
 		  30 30 30 30 30 30 30 30 30 30 30 30 
-		  30 330</skj>
+		  30 30 30 30 30 30 30 30 30 30 30 30 
+		  30 30 30 30 30 30 30 30 30 30 30 30 		  
+		  30 30 30 30 30 30 30 30 30 30 30 30 
+		  30 30 30 30 30 30 30 30 30 30 30 30 
+		  30 30 30 2970
+	</skj>
 </sp_unit_cohort>
 
 <length>
 	<skj>3     4.5   7.52  11.65 16.91 21.83 26.43 30.72 34.73 38.49 41.99 45.27 
 		 48.33 51.19 53.86 56.36 58.70 60.88 62.92 64.83 66.61 68.27 69.83 71.28 
 		 72.64 73.91 75.10 76.21 77.25 78.22 79.12 79.97 80.76 81.50 82.19 82.83 
 		 83.50 83.50 83.50 83.50 83.50 83.50 83.50 83.50 83.50 83.50 83.50 83.50
-		 85.00 87.96 </skj>
+		 85.00 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96
+		 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96
+		 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96
+		 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96
+		 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96 87.96		 
+		 </skj>
 </length>
 
 <weight>
 	<skj>0.0003 0.00109 0.00571 0.023 0.08  0.18  0.32  0.53  0.78  1.09  1.44  1.84
 		 2.27   2.73    3.21    3.72  4.23  4.76  5.30  5.83  6.36  6.89  7.40  7.91 
 		 8.41   8.89    9.36    9.81  10.25 10.66 11.07 11.45 11.82 12.17 12.51 12.83 
 		 13.5   13.5    13.5    13.5  13.50 13.50 13.50 13.50 13.50 13.50 13.50 13.50
-		 14.0	15.56  </skj>
+		 14.0	15.56   15.56   15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56
+		 15.56  15.56   15.56   15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56
+		 15.56  15.56   15.56   15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56
+		 15.56  15.56   15.56   15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56
+		 15.56  15.56   15.56   15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56
+		 15.56  15.56   15.56   15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56 15.56		 		 
+		 </skj>
 </weight>