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
|
# This file is part of xrayutilities.
#
# xrayutilities is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, see <http://www.gnu.org/licenses/>.
#
# Copyright (C) 2025 Dominik Kriegner <dominik.kriegner@gmail.com>
import time
import unittest
import numpy as np
import xrayutilities as xu
from scipy.stats import uniform
# get maximum number of threads
# if this is -1 no openmp is available and we skip the test class
max_threads = xu.cxrayutilities.get_max_openmp_threads()
@unittest.skipIf(max_threads == -1, "extension was not compiled with OpenMP")
class TestParallelPerformance(unittest.TestCase):
def setUp(self):
"""Set up the HXRD object and generate random angles."""
mat = xu.materials.Si # Example material
self.hxrd = xu.HXRD(mat.Q(1, 1, 0), mat.Q(0, 0, 1))
self.num_angles = 10000000 # Number of random angles to test
self.tt_angles = uniform.rvs(size=self.num_angles, loc=0, scale=80)
self.pp_angles = uniform.rvs(size=self.num_angles, loc=0, scale=140)
def test_parallel_performance(self):
"""Test the performance scaling of parallel processing.
Tests angle to Q-space conversion performance with varying numbers of
threads. Calculates speedup and efficiency metrics for different thread
counts. Checks that parallel processing achieves reasonable efficiency
thresholds.
"""
if max_threads < 2:
self.skipTest("OpenMP available, but not enough allowed threads")
return
times = {}
n_threads_tested = list(range(1, min(13, max_threads + 1)))
for nthreads in n_threads_tested:
xu.config.NTHREADS = nthreads
start_time = time.time()
qx, qy, qz = self.hxrd.Ang2Q(self.pp_angles, self.tt_angles)
end_time = time.time()
times[nthreads] = end_time - start_time
print(f"Nthreads: {nthreads}, Time: {times[nthreads]:.4f} seconds")
# Calculate Efficiency
efficiencies = {}
time_1 = times[1] # Time for 1 thread
for nthreads in n_threads_tested:
speedup = time_1 / times[nthreads]
efficiency = speedup / nthreads
efficiencies[nthreads] = efficiency
print(
f"Nthreads: {nthreads}, Speedup: {speedup:.2f}, "
f"Efficiency: {efficiency:.2f}"
)
# Average Efficiency (excluding 1 thread, as efficiency=1.0)
avg_efficiency = np.mean(list(efficiencies.values())[1:])
print(f"Average Efficiency (excluding 1 thread): {avg_efficiency:.2f}")
# Assertions based on the average efficiency
self.assertGreater(
avg_efficiency, 0.25, "Average efficiency should be reasonable"
)
self.assertGreater(
efficiencies[2], 0.5, "Efficiency at 2 threads should be high."
)
if __name__ == "__main__":
unittest.main()
|
