-
Notifications
You must be signed in to change notification settings - Fork 2
/
bursts_duration_bytes_wcdf.py
executable file
·270 lines (231 loc) · 10.7 KB
/
bursts_duration_bytes_wcdf.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
#! /usr/bin/python
# -*- coding: utf-8 -*-
#
# Copyright 2015 Quentin De Coninck
#
# This program 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 3 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, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
# MA 02110-1301, USA.
#
# To install on this machine: matplotlib, numpy
from __future__ import print_function
import argparse
import common as co
import common_graph as cog
import matplotlib
# Do not use any X11 backend
matplotlib.use('Agg')
matplotlib.rcParams['pdf.fonttype'] = 42
matplotlib.rcParams['ps.fonttype'] = 42
import matplotlib.pyplot as plt
import mptcp
import numpy as np
import os
import tcp
##################################################
## ARGUMENTS ##
##################################################
parser = argparse.ArgumentParser(
description="Summarize stat files generated by analyze")
parser.add_argument("-s",
"--stat", help="directory where the stat files are stored", default=co.DEF_STAT_DIR + '_' + co.DEF_IFACE)
parser.add_argument('-S',
"--sums", help="directory where the summary graphs will be stored", default=co.DEF_SUMS_DIR + '_' + co.DEF_IFACE)
parser.add_argument("-d",
"--dirs", help="list of directories to aggregate", nargs="+")
args = parser.parse_args()
stat_dir_exp = os.path.abspath(os.path.expanduser(args.stat))
sums_dir_exp = os.path.abspath(os.path.expanduser(args.sums))
co.check_directory_exists(sums_dir_exp)
##################################################
## GET THE DATA ##
##################################################
connections = cog.fetch_valid_data(stat_dir_exp, args)
multiflow_connections, singleflow_connections = cog.get_multiflow_connections(connections)
##################################################
## PLOTTING RESULTS ##
##################################################
TINY = '0B-10KB'
SMALL = '10KB-100KB'
MEDIUM = '100KB-1MB'
LARGE = '>=1MB'
results_duration_bytes = {co.C2S: {TINY: [], SMALL: [], MEDIUM: [], LARGE: []}, co.S2C: {TINY: [], SMALL: [], MEDIUM: [], LARGE: []}}
results_pkts = {co.C2S: {TINY: [], SMALL: [], MEDIUM: [], LARGE: []}, co.S2C: {TINY: [], SMALL: [], MEDIUM: [], LARGE: []}}
min_duration = 0.001
for fname, conns in multiflow_connections.iteritems():
for conn_id, conn in conns.iteritems():
# Restrict to only 2SFs, but we can also see with more than 2
if co.START in conn.attr and len(conn.flows) >= 2:
# Rely here on MPTCP duration, maybe should be duration at TCP level?
# Also rely on the start time of MPTCP; again, should it be the TCP one?
conn_start_time = conn.attr[co.START]
conn_start_time_int = long(conn_start_time)
conn_start_time_dec = float('0.' + str(conn_start_time - conn_start_time_int).split('.')[1])
conn_duration = conn.attr[co.DURATION]
if conn_duration < min_duration:
continue
for direction in co.DIRECTIONS:
tot_packs = 0
to_add_pkts = []
# First count all bytes sent (including retransmissions)
tcp_conn_bytes = 0
for flow_id, flow in conn.flows.iteritems():
tcp_conn_bytes += flow.attr[direction].get(co.BYTES_DATA, 0)
# To cope with unseen TCP connections
conn_bytes = max(conn.attr[direction][co.BYTES_MPTCPTRACE], tcp_conn_bytes)
for flow_id, bytes, pkts, burst_duration, burst_start_time in conn.attr[direction][co.BURSTS]:
frac_bytes = (bytes + 0.0) / conn_bytes
if frac_bytes > 1.1:
print(frac_bytes, bytes, pkts, conn_bytes, direction, conn_id, flow_id)
continue
if frac_bytes < 0:
print(frac_bytes, bytes, pkts, conn_bytes, direction, conn_id, flow_id)
continue
burst_start_time_int = long(burst_start_time)
burst_start_time_dec = float('0.' + str(burst_start_time - burst_start_time_int).split('.')[1])
relative_time_int = burst_start_time_int - conn_start_time_int
relative_time_dec = burst_start_time_dec - conn_start_time_dec
relative_time = relative_time_int + relative_time_dec
frac_duration = relative_time / conn_duration
if frac_duration >= 0.0 and frac_duration <= 2.0:
if conn_bytes < 10000:
label = TINY
elif conn_bytes < 100000:
label = SMALL
elif conn_bytes < 1000000:
label = MEDIUM
else:
label = LARGE
results_duration_bytes[direction][label].append((frac_duration, frac_bytes))
to_add_pkts.append(pkts)
tot_packs += pkts
if conn_bytes < 10000:
label = TINY
elif conn_bytes < 100000:
label = SMALL
elif conn_bytes < 1000000:
label = MEDIUM
else:
label = LARGE
for p in to_add_pkts:
results_pkts[direction][label].append(p * 1.0 / tot_packs)
base_graph_name = 'bursts_'
color = {TINY: 'red', SMALL: 'blue', MEDIUM: 'green', LARGE: 'orange'}
ls = {TINY: ':', SMALL: '-.', MEDIUM: '--', LARGE: '-'}
for direction in co.DIRECTIONS:
plt.figure()
plt.clf()
fig, ax = plt.subplots()
graph_fname = os.path.splitext(base_graph_name)[0] + "duration_wcdf_" + direction + ".pdf"
graph_full_path = os.path.join(sums_dir_exp, graph_fname)
for label in [TINY, SMALL, MEDIUM, LARGE]:
x_val = [x[0] for x in results_duration_bytes[direction][label]]
sample = np.array(sorted(x_val))
sorted_array = np.sort(sample)
tot = 0.0
yvals = []
for elem in sorted_array:
tot += elem
yvals.append(tot)
yvals = [x / tot for x in yvals]
if len(sorted_array) > 0:
# Add a last point
sorted_array = np.append(sorted_array, sorted_array[-1])
yvals = np.append(yvals, 1.0)
ax.plot(sorted_array, yvals, color=color[label], linestyle=ls[label], linewidth=2, label=label)
# Shrink current axis's height by 10% on the top
# box = ax.get_position()
# ax.set_position([box.x0, box.y0,
# box.width, box.height * 0.9])
# ax.set_xscale('log')
# Put a legend above current axis
# ax.legend(loc='lower center', bbox_to_anchor=(0.5, 1.05), fancybox=True, shadow=True, ncol=ncol)
ax.legend(loc='best')
plt.xlabel('Fraction of connection duration', fontsize=24)
plt.ylabel("Weighted CDF", fontsize=24)
plt.savefig(graph_full_path)
plt.close('all')
plt.figure()
plt.clf()
fig, ax = plt.subplots()
graph_fname = os.path.splitext(base_graph_name)[0] + "bytes_wcdf_" + direction + ".pdf"
graph_full_path = os.path.join(sums_dir_exp, graph_fname)
for label in [TINY, SMALL, MEDIUM, LARGE]:
y_val = [x[1] for x in results_duration_bytes[direction][label]]
sample = np.array(sorted(y_val))
sorted_array = np.sort(sample)
tot = 0.0
yvals = []
for elem in sorted_array:
tot += elem
yvals.append(tot)
yvals = [x / tot for x in yvals]
if len(sorted_array) > 0:
# Add a last point
sorted_array = np.append(sorted_array, sorted_array[-1])
yvals = np.append(yvals, 1.0)
ax.plot(sorted_array, yvals, color=color[label], linestyle=ls[label], linewidth=2, label=label)
# Shrink current axis's height by 10% on the top
# box = ax.get_position()
# ax.set_position([box.x0, box.y0,
# box.width, box.height * 0.9])
# ax.set_xscale('log')
# Put a legend above current axis
# ax.legend(loc='lower center', bbox_to_anchor=(0.5, 1.05), fancybox=True, shadow=True, ncol=ncol)
ax.legend(loc='best')
plt.xlim(0.0, 1.0)
plt.xlabel('Fraction of connection bytes', fontsize=24)
plt.ylabel("Weighted CDF", fontsize=24)
plt.savefig(graph_full_path)
plt.close('all')
plt.figure()
plt.clf()
fig, ax = plt.subplots()
graph_fname = os.path.splitext(base_graph_name)[0] + "pkts_wcdf_" + direction + ".pdf"
graph_full_path = os.path.join(sums_dir_exp, graph_fname)
for label in [TINY, SMALL, MEDIUM, LARGE]:
sample = np.array(sorted(results_pkts[direction][label]))
sorted_array = np.sort(sample)
tot = 0.0
yvals = []
for elem in sorted_array:
tot += elem
yvals.append(tot)
yvals = [x / tot for x in yvals]
print("PERCENTAGE 1 BLOCK", direction, label, len([x for x in sorted_array if x >= 0.99]) * 100. / tot)
i = 0
for elem in sorted_array:
if elem >= 0.2:
break
else:
i += 1
print("PERCENTAGE 0.2 block conn", direction, label, yvals[i])
if len(sorted_array) > 0:
# Add a last point
sorted_array = np.append(sorted_array, sorted_array[-1])
yvals = np.append(yvals, 1.0)
ax.plot(sorted_array, yvals, color=color[label], linestyle=ls[label], linewidth=2, label=label)
# Shrink current axis's height by 10% on the top
# box = ax.get_position()
# ax.set_position([box.x0, box.y0,
# box.width, box.height * 0.9])
# ax.set_xscale('log')
# Put a legend above current axis
# ax.legend(loc='lower center', bbox_to_anchor=(0.5, 1.05), fancybox=True, shadow=True, ncol=ncol)
ax.legend(loc='best')
plt.xlim(0.0, 1.0)
plt.xlabel('Fraction of connection packets', fontsize=24, labelpad=-1)
plt.ylabel("Weighted CDF", fontsize=24)
plt.savefig(graph_full_path)
plt.close('all')