-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathplot_hegemony_similarity.py
274 lines (238 loc) · 10.5 KB
/
plot_hegemony_similarity.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
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
import argparse
import configparser
import logging
import os
import sys
from collections import namedtuple
from datetime import datetime
import matplotlib.pyplot as plt
import numpy as np
from network_dependency.utils.helper_functions import parse_timestamp_argument
from network_dependency.utils.scope import read_legacy_scopes
Stats = namedtuple('Stats', 'min max avg median ixp')
def plot_scopes(data: dict, x_metric: str, y_metric: str, output: str, title: str) -> None:
x = list()
y = list()
x_ixp = list()
y_ixp = list()
for as_ in data:
if data[as_].ixp:
x_ixp.append(getattr(data[as_], x_metric))
y_ixp.append(getattr(data[as_], y_metric))
else:
x.append(getattr(data[as_], x_metric))
y.append(getattr(data[as_], y_metric))
# definitions for the axes
left, width = 0.1, 0.65
bottom, height = 0.1, 0.65
spacing = 0.035
rect_scatter = [left, bottom, width, height]
rect_histx = [left, bottom + height + spacing, width, 0.2]
rect_histy = [left + width + spacing, bottom, 0.2, height]
# start with a square Figure
fig = plt.figure(figsize=(8, 8))
ax = fig.add_axes(rect_scatter)
ax_histx = fig.add_axes(rect_histx, sharex=ax)
ax_histy = fig.add_axes(rect_histy, sharey=ax)
ax_histx.set_title(title)
# no labels
ax_histx.tick_params(axis="x", labelbottom=False)
ax_histy.tick_params(axis="y", labelleft=False)
# the scatter plot:
ax.set_xlim(0, 1)
ax.set_xticks(np.arange(0, 1.1, 0.1))
ax.set_xlabel(x_metric)
ax.set_ylim(0, 1)
ax.set_yticks(np.arange(0, 1.1, 0.1))
ax.set_ylabel(y_metric)
ax.grid(which='both')
ax.plot([0, 1], [0, 1], c='gray', ls='--')
if x:
ax.scatter(x, y)
if x_ixp:
ax.scatter(x_ixp, y_ixp, c='red')
# now determine nice limits by hand:
binwidth = 0.05
xymax = max(np.max(np.abs(x + x_ixp)), np.max(np.abs(y + y_ixp)))
lim = (int(xymax / binwidth) + 1) * binwidth
bins = np.arange(-lim, lim + binwidth, binwidth)
ax_histx.hist(x + x_ixp, bins=bins)
ax_histy.hist(y + y_ixp, bins=bins, orientation='horizontal')
plt.savefig(output, bbox_inches='tight')
def plot_raw_scores(data: dict, output: str, title: str) -> None:
# definitions for the axes
left, width = 0.1, 0.65
bottom, height = 0.1, 0.65
spacing = 0.035
rect_scatter = [left, bottom, width, height]
rect_histx = [left, bottom + height + spacing, width, 0.2]
rect_histy = [left + width + spacing, bottom, 0.2, height]
# start with a square Figure
fig = plt.figure(figsize=(8, 8))
ax = fig.add_axes(rect_scatter)
ax_histx = fig.add_axes(rect_histx, sharex=ax)
ax_histy = fig.add_axes(rect_histy, sharey=ax)
ax_histx.set_title(title)
# no labels
ax_histx.tick_params(axis="x", labelbottom=False)
ax_histy.tick_params(axis="y", labelleft=False)
# the scatter plot:
ax.set_xlim(0, 1)
ax.set_xticks(np.arange(0, 1.1, 0.1))
ax.set_xlabel('traceroute')
ax.set_ylim(0, 1)
ax.set_yticks(np.arange(0, 1.1, 0.1))
ax.set_ylabel('BGP')
ax.grid(which='both')
ax.plot([0, 1], [0, 1], c='gray', ls='--')
agg_x = list()
agg_y = list()
for as_ in data:
x, y = zip(*data[as_])
agg_x += x
agg_y += y
ax.scatter(x, y)
# now determine nice limits by hand:
binwidth = 0.05
xymax = max(np.max(np.abs(agg_x)), np.max(np.abs(agg_y)))
lim = (int(xymax / binwidth) + 1) * binwidth
bins = np.arange(0, lim + binwidth, binwidth)
ax_histx.hist(agg_x, bins=bins)
ax_histy.hist(agg_y, bins=bins, orientation='horizontal')
plt.savefig(output, bbox_inches='tight')
def get_stats(data: list, ixp_dependent: bool):
min_ = np.min(data)
max_ = np.max(data)
avg = np.mean(data)
median = np.median(data)
return Stats(min_, max_, avg, median, ixp_dependent)
def write_data(data: dict, output_file: str):
out_lines = ['as min max avg median\n']
for as_ in data:
stats = data[as_]
out_lines.append(' '.join(map(str,
[as_, stats.min, stats.max,
stats.avg, stats.median]))
+ '\n')
with open(output_file, 'w') as f:
f.writelines(out_lines)
def write_raw_data(data: dict, output_file: str):
out_lines = list()
for as_ in data:
out_lines.append(' '.join(map(str, [as_] + data[as_]))
+ '\n')
with open(output_file, 'w') as f:
f.writelines(out_lines)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('config')
parser.add_argument('-tt', '--traceroute_timestamp',
help='Timestamp (as UNIX epoch in seconds or '
'milliseconds, or in YYYY-MM-DDThh:mm format)')
parser.add_argument('-bt', '--bgp_timestamp',
help='Timestamp (as UNIX epoch in seconds or '
'milliseconds, or in YYYY-MM-DDThh:mm format)')
# Logging
FORMAT = '%(asctime)s %(processName)s %(message)s'
logging.basicConfig(
format=FORMAT, # filename='../compare_results.log',
level=logging.DEBUG, datefmt='%Y-%m-%d %H:%M:%S',
)
logging.info("Started: %s" % sys.argv)
args = parser.parse_args()
# Read config
config = configparser.ConfigParser()
config.read(args.config)
traceroute_timestamp_argument = config.get('input', 'traceroute_timestamp', fallback=None)
if args.traceroute_timestamp is not None:
logging.info('Overriding config traceroute timestamp.')
traceroute_timestamp_argument = args.traceroute_timestamp
traceroute_timestamp = parse_timestamp_argument(traceroute_timestamp_argument)
if traceroute_timestamp == 0:
logging.error('Invalid traceroute timestamp specified: {}'
.format(traceroute_timestamp_argument))
exit(1)
bgp_timestamp_argument = config.get('input', 'bgp_timestamp', fallback=None)
if args.bgp_timestamp is not None:
logging.info('Overriding config BGP timestamp.')
bgp_timestamp_argument = args.bgp_timestamp
bgp_timestamp = parse_timestamp_argument(bgp_timestamp_argument)
if bgp_timestamp == 0:
logging.error('Invalid BGP timestamp specified: {}'
.format(bgp_timestamp_argument))
exit(1)
logging.info('Timestamp: {} {}'
.format(datetime.utcfromtimestamp(traceroute_timestamp)
.strftime('%Y-%m-%dT%H:%M'), traceroute_timestamp))
bgp_kafka_topic = config.get('input', 'bgp_kafka_topic',
fallback='ihr_hegemony')
traceroute_kafka_topic = config.get('input', 'traceroute_kafka_topic',
fallback='ihr_hegemony_traceroutev4')
bootstrap_servers = config.get('kafka', 'bootstrap_servers',
fallback='kafka2:9092')
scope_as_filter = config.get('input', 'asns', fallback=None)
if not scope_as_filter:
scope_as_filter = None
if scope_as_filter is not None:
scope_as_filter = set(scope_as_filter.split(','))
logging.info('Filtering for AS scopes: {}'.format(scope_as_filter))
bgp_scopes = read_legacy_scopes(bgp_kafka_topic,
bgp_timestamp * 1000,
bootstrap_servers,
scope_as_filter)
traceroute_scopes = read_legacy_scopes(traceroute_kafka_topic,
traceroute_timestamp * 1000,
bootstrap_servers,
scope_as_filter)
if not bgp_scopes or not traceroute_scopes:
logging.error('No results for BGP ({} entries) or traceroute ({} '
'entries).'.format(len(bgp_scopes),
len(traceroute_scopes)))
exit(1)
if traceroute_scopes.keys() - bgp_scopes.keys():
logging.info('Traceroute scopes without matching BGP scope: {}'
.format(traceroute_scopes.keys() - bgp_scopes.keys()))
# if bgp_scopes.keys() - traceroute_scopes.keys():
# logging.info('BGP scopes without matching traceroute scope: {}'
# .format(bgp_scopes.keys() - traceroute_scopes.keys()))
plot_data_overlap = dict()
raw_data_overlap = dict()
plot_data_union = dict()
raw_data_union = dict()
raw_scores = dict()
for as_ in traceroute_scopes.keys() & bgp_scopes.keys():
if as_ == '-1':
continue
traceroute_scope = traceroute_scopes[as_]
bgp_scope = bgp_scopes[as_]
raw_scores[as_] = list()
for dep_as in traceroute_scope.union(bgp_scope):
raw_scores[as_].append((traceroute_scope.get_score(dep_as, True),
bgp_scope.get_score(dep_as, True)))
ixp_dependent = traceroute_scope.is_ixp_dependent
deltas_overlap = traceroute_scope \
.get_score_deltas_for_overlap(bgp_scope)
deltas_union = traceroute_scope.get_score_deltas_for_union(bgp_scope)
scores_overlap = [abs(s) for _, s in deltas_overlap]
scores_union = [abs(s) for _, s in deltas_union]
if scores_overlap:
plot_data_overlap[as_] = get_stats(scores_overlap, ixp_dependent)
raw_data_overlap[as_] = deltas_overlap
plot_data_union[as_] = get_stats(scores_union, ixp_dependent)
raw_data_union[as_] = deltas_union
output_dir = config.get('output', 'directory', fallback='./')
if not output_dir.endswith('/'):
output_dir += '/'
out_date = datetime.utcfromtimestamp(traceroute_timestamp).strftime('%Y-%m-%dT%H:%M')
output_dir += out_date + '/'
os.makedirs(output_dir, exist_ok=True)
write_data(plot_data_overlap, output_dir + 'overlap.dat')
write_raw_data(raw_data_overlap, output_dir + 'overlap-raw.dat')
write_data(plot_data_union, output_dir + 'union.dat')
write_raw_data(raw_data_union, output_dir + 'union-raw.dat')
write_raw_data(raw_scores, output_dir + 'raw.dat')
plot_scopes(plot_data_overlap, 'max', 'avg', output_dir + 'overlap_max_avg.pdf', out_date)
plot_scopes(plot_data_overlap, 'max', 'median', output_dir + 'overlap_max_median.pdf', out_date)
plot_scopes(plot_data_union, 'max', 'avg', output_dir + 'union_max_avg.pdf', out_date)
plot_scopes(plot_data_union, 'max', 'median', output_dir + 'union_max_median.pdf', out_date)
plot_raw_scores(raw_scores, output_dir + 'raw.pdf', out_date)