IMB-Benchmarking-tools/postprocess_data.py
2025-05-28 19:46:24 +02:00

73 lines
2.5 KiB
Python
Executable File

import pandas as pd
import os
data_markers = {
"block_separator": "#----------------------------------------------------------------",
"benchmark_type": "# Benchmarking",
"processes_num": "# #processes = ",
"min_bytelen": "# Minimum message length in bytes",
"max_bytelen": "# Maximum message length in bytes",
"mpi_datatype": "# MPI_Datatype :",
"mpi_red_datatype": "# MPI_Datatype for reductions :",
"mpi_red_op": "# MPI_Op",
"end_of_table": "# All processes entering MPI_Finalize",
}
column_names = [
"benchmark_type",
"proc_num",
"msg_size_bytes",
"repetitions",
"t_min_usec",
"t_max_usec",
"t_avg_usec",
"mpi_datatype",
"mpi_red_datatype",
"mpi_red_op",
]
data = list()
for file in os.listdir("data/"):
with open("data/"+file, 'r') as f:
lines = f.readlines()
past_preheader = False
in_header = False
in_body = False
btype = None
proc_num = None
mpi_datatype = None
mpi_red_datatype = None
mpi_red_op = None
for line in lines:
if data_markers["block_separator"] in line:
if in_header and not past_preheader:
past_preheader = True
elif in_header and past_preheader:
in_body = True
in_header = not in_header
continue
if not in_header and not in_body and past_preheader:
if data_markers["mpi_datatype"] in line:
mpi_datatype = line.split()[-1]
elif data_markers["mpi_red_datatype"] in line:
mpi_red_datatype = line.split()[-1]
elif data_markers["mpi_red_op"] in line:
mpi_red_op = line.split()[-1]
if past_preheader and in_header:
if data_markers["benchmark_type"] in line:
btype = line.split()[2]
if data_markers["processes_num"] in line:
proc_num = int(line.split()[3])
if in_body:
if "#" in line or "".join(line.split()) == "":
continue
if data_markers["end_of_table"] in line:
break
data.append([btype, proc_num]+[int(s) if s.isdigit()
else float(s) for s in line.split()] + [mpi_datatype, mpi_red_datatype, mpi_red_op])
df = pd.DataFrame(data, columns=column_names)
df.to_csv("data.csv", index=False)