-
Notifications
You must be signed in to change notification settings - Fork 0
/
plot_benchmark.py
executable file
·109 lines (91 loc) · 2.97 KB
/
plot_benchmark.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
#!/usr/bin/env python3
"""
Make a plot of CP2K benchmark data.
"""
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.ticker
import argparse
# Base directory
BASE = "/home/jeremy/cp2k-uk/results/"
YSHIFT = 1.1
def makePlot(benchmarkName, machineName, shift, showPlot):
"""
Make a plot of benchmark results.
Arguments:
benchmarkName -- the name of the benchmark to plot results for
machineName -- the name of the machine to get results for
shift -- the shift of the data point label on the y-axis
showPlot -- whether to show the plot in a window when done
"""
# Turn interactive mode off (required in ipython)
plt.ioff()
# Construct file name based on machine name and benchmark name
fileName = (
BASE
+ machineName.lower()
+ "_benchmarks/"
+ benchmarkName
+ "/"
+ benchmarkName
+ "_besttimes.txt"
)
print("Processing benchmark results for %s from file:" % machineName)
print(fileName)
# Load data from file
data = np.loadtxt(
fileName,
dtype={"names": ("nodes", "time", "label"), "formats": ("i4", "f4", "a5")},
)
# Create plot
fig = plt.figure()
ax = fig.add_subplot(111)
# Plot data
ax.loglog(data["nodes"], data["time"], marker="o")
# Change axes ticks to integers
ax.get_xaxis().set_major_formatter(matplotlib.ticker.FormatStrFormatter("%d"))
ax.get_yaxis().set_major_formatter(matplotlib.ticker.FormatStrFormatter("%d"))
# Add title and axes labels
title = "Performance of the " + benchmarkName + " benchmark on " + machineName
plt.title(title)
plt.xlabel("Number of nodes used")
plt.ylabel("Time (seconds)")
# Label the points with the best performance
for i, labelText in enumerate(data["label"]):
ax.annotate(
labelText,
xy=(data["nodes"][i], data["time"][i]),
xytext=(data["nodes"][i], shift * data["time"][i]),
)
# Save plot to file
outFileName = benchmarkName + "-" + machineName.lower() + ".png"
plt.savefig(outFileName)
if showPlot:
plt.show()
plt.clf()
if __name__ == "__main__":
# Command line options
parser = argparse.ArgumentParser(description="Plot benchmark results.")
parser.add_argument(
"--name", default="fayalite-FIST", help="name of the benchmark to process"
)
parser.add_argument(
"--machine", default="Magnus", help="name of the machine used to obtain results"
)
parser.add_argument(
"--shift",
default=YSHIFT,
help="Fractional shift of label in y-direction",
type=float,
)
parser.add_argument(
"--show",
action="store_true",
help="show the plot window as well as saving to file",
)
args = parser.parse_args()
benchmarkName = args.name
machineName = args.machine
shift = args.shift
showPlot = args.show
makePlot(benchmarkName, machineName, shift, showPlot)