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
|
namebench has an opt-in feature that enables users to upload and share results.
This is a technical description to show exactly what data is communicated to
the server via JSON.
Many performance numbers are duplicated within the submitted data. For
instance, we upload the individual durations for each query, along with an
average. This is done as a performance enhancement, since the client has
already calculated all of the averages, saving AppEngine from duplicating
effort.
-------------------------------------------------------------------------------
# Lets start with the most interesting number from a privacy perspective:
#
# The client_id is a simple signed integer, designed to detect subsequent runs
# from the same user, so that the new runs can be hidden. To arrive at a unique
# integer, we create a long string containing the following text, and create an
# integer checksum of that data (see site_connector.py):
#
# platform.platform() Operating system name and version
# sys.version Python interpreter version
# platform.node() Hostname of machine
# os.getenv('HOME') Filesystem path to home directory
#
# The checksum is calculated using the crc32() function from the zlib library.
# We purposefully chose such a simple algorithm with a higher rate of collision
# than MD5 or SHA1. For our purpose, a simple 32-bit integer is good enough to
# detect subsequent submissions from the same user within 6 hours.
# This item is deleted from the datastore within 12h.
'client_id': 1078802240,
# This is a simple random integer to prevent a duplicate submission in the case of a
# client or server error. This item is deleted from the datastore within 12h.
'submit_id': 34923485,
# A user may request that this submission is unlisted directly.
'hidden': False,
'data':
{
"nameservers":[
{
"timeout_count":0,
# How long the health checks took for this host.
"check_average":433.1902027130127,
# Any node_ids reported by a nameserver. This is removed for internal hosts.
"node_ids":[
"3.nyc"
],
# This is obfuscated for internal hosts.
"ip":"208.67.220.220",
# How long each benchmark query took in ms. Normally there would be 250 of these.
"durations":[
[
671.98014259338379,
611.4649772644043,
625.26202201843262,
630.34987449645996,
628.66091728210449,
697.69501686096191,
716.68291091918945,
715.0270938873291,
713.67907524108887,
715.93189239501953,
716.09187126159668,
714.63704109191895,
715.09003639221191,
721.9398021697998,
710.00409126281738,
717.65494346618652,
711.21597290039062,
716.58420562744141,
715.98601341247559,
715.15107154846191,
182.373046875,
64.732074737548828,
37.251949310302734,
62.382221221923828,
54.202795028686523
]
],
"is_disabled":false,
"is_global":false,
"error_count":0,
"nx_count":0,
# The performance ranking of this server in your tests.
"position":0,
"is_failure_prone":false,
# pre-computed average for all runs.
"overall_average":571.28124237060547,
# Index hosts are a pre-defined list of queries that are always
# queried for submissions. This allows providers to compare results
# on an individual query basis. The hosts do not differ between submissions.
"index":[
[
"www.google.com.",
"A",
716.68291091918945,
2,
30,
"google.navigation.opendns.com. -> 208.67.217.231, 208.67.217.230"
],
[
"www.amazon.com.",
"A",
95.599889755249023,
1,
15,
"207.171.166.252"
],
[
"www.wikipedia.org.",
"A",
142.0438289642334,
3,
3308,
"rr.wikimedia.org. -> rr.pmtpa.wikimedia.org. -> 208.80.152.2"
],
[
"www.twitter.com.",
"A",
151.8549919128418,
2,
509,
"twitter.com. -> 128.242.240.148"
],
[
"yahoo.com.",
"A",
166.63408279418945,
1,
19068,
"209.191.122.70, 67.195.160.76, 69.147.125.65, 72.30.2.43, 98.137.149.56"
],
[
"a.root-servers.net.",
"A",
101.4409065246582,
1,
604799,
"198.41.0.4"
],
[
"www.google-analytics.com.",
"A",
60.657978057861328,
2,
82164,
"www-google-analytics.l.google.com. -> 72.14.204.102, 72.14.204.113, 72.14.204.100, 72.14.204.101"
],
[
"www.youtube.com.",
"A",
65.762996673583984,
2,
86118,
"youtube-ui.l.google.com. -> 74.125.93.113, 74.125.93.100, 74.125.93.139, 74.125.93.102, 74.125.93.138, 74.125.93.101"
],
[
"www.baidu.com.",
"A",
61.342954635620117,
2,
959,
"www.a.shifen.com. -> 220.181.6.175"
],
[
"www.facebook.com.",
"A",
100.84199905395508,
1,
40,
"66.220.146.11"
]
],
"is_reference":true,
"is_custom":false,
"name":"",
# Any errors the nameserver has.
"notes":[
{
"url":"http://code.google.com/p/namebench/wiki/FAQ#What_does_\"Incorrect_result_for...\"_mean?",
"text":"www.google.com is hijacked: google.navigation.opendns.com"
}
],
# This is masked for internal hostnames.
"hostname":"resolver2.opendns.com",
# If this host is currently in use by the system, this is the positional element. 0 would signify
# a primary nameserver.
"sys_position":null,
# If the nameserver returns a version number, it would be returned here. This is removed for
# private IP's.
"version":"",
"is_regional":false,
# OARC port diversity data - only available for public IP's.
"port_behavior":"208.67.217.8 is GREAT: 26 queries in 2.0 seconds from 26 ports with std dev 18425",
# Precomputed slowest and fastest times for any query.
"duration_max":721.9398021697998,
"duration_min":37.251949310302734,
# An average for each run.
"averages":[
571.28124237060547
]
},
{
# ... additional nameservers would have data stored here ...
}
],
# This is GeoIP data returned by the Google Gears JSON API, and is
# used to make results available on a geographic basis.
"geodata":{
"city":"Manchester",
"region_name":"New Hampshire",
# Whether or not we used Google (default) or Maxmind for GeoIP data.
"source":"gloc",
# The lat/long accuracy is reduced to only provide city-level
# accuracy, which is enough for us to keep up with city name changes.
"longitude":-71.454,
"latitude":42.995,
"country_code":"US",
"country_name":"United States"
},
# These are the options selected by the user.
"config":{
"query_count":25,
"benchmark_thread_count":2,
"template":"html",
"health_thread_count":40,
"version":"1.3-BETA1",
"only":true,
"select_mode":"automatic",
"platform":"Mac OS X",
"num_servers":11,
"hide_results":0,
"ping_timeout":0.5,
"timeout":3.5,
"run_count":1,
"input_source":"chrome",
"health_timeout":3.75,
"upload_results":true
}
}
|
