-
Notifications
You must be signed in to change notification settings - Fork 0
/
osm_utils.py
289 lines (245 loc) · 10.3 KB
/
osm_utils.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
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
#!/usr/bin/python
"""
date >> df.txt
df | sort | grep /dev/sd >> df.txt
Fri Apr 29 21:47:35 CDT 2016
/dev/sda1 1922728752 263520744 1561515952 15% /media/Borg_LS
/dev/sdb6 351784100 238139416 95752032 72% /
/dev/sdb7 40185208 49036 38071788 1% /media/OSM040
/dev/sdb8 71897424 24760156 43461980 37% /media/OSM070
/dev/sdc1 302248384 64352 286807648 1% /media/OSM300
/dev/sdc2 151058636 60872 143301380 1% /media/OSM150
/dev/sdc3 201454560 60692 191137484 1% /media/OSM200
/dev/sdc4 329221796 68304 312406948 1% /media/OSM325
"""
import csv
import os
from datetime import datetime
from datetime import timedelta
from matplotlib.dates import date2num
import matplotlib.pyplot as plt
def dstat_format_func(keys, x, st=None):
if "time" in keys[1]:
# 30-04 12:47:38
assert st is not None
dt = datetime.strptime(x, "%d-%m %H:%M:%S")
dt = dt.replace(year=2016)
dt = dt - st
return dt.total_seconds()
elif "read" in keys[1]:
return float(x)/1024.0/1024.0
elif "writ" in keys[1]:
return float(x)/1024.0/1024.0
elif "swap" in keys[0]:
return float(x)/1024.0/1024.0/1024.0
elif "paging" in keys[0]:
return float(x)/1024.0/1024.0
elif "memory usage" in keys[0]:
if "buff" in keys[1]:
return float(x)/1024.0/1024.0
else:
return float(x)/1024.0/1024.0/1024.0
else:
return float(x)
def read_osm2pgsql(filename):
osm2pgsqldict = {}
with open(filename) as csvfile:
reader = csv.reader(csvfile, delimiter=',')
row = reader.next()
assert row[0].lower() == 'i_flag'
has_parallel_indexing = row[1].lower() == 'true'
row = reader.next()
dt0 = datetime.strptime(row[1], "%a %b %d %H:%M:%S %Z %Y")
osm2pgsqldict[row[0]] = dt0
dt = dt0
for _ in range(4):
row = reader.next()
dt += timedelta(seconds=float(row[1]))
osm2pgsqldict[row[0]] = dt
if has_parallel_indexing:
for row in reader:
dt += timedelta(seconds=float(row[1]))
osm2pgsqldict[row[0]] = dt
osm2pgsqldict.pop('result', None)
else:
temp_dict = {}
for row in reader:
if row[0].startswith('#'):
continue
temp_dict[row[0]] = float(row[1])
osm2pgsqldict['result'] = dt0 + timedelta(seconds=temp_dict['result'])
osm2pgsqldict['planet_osm_rels'] = osm2pgsqldict['result'] - timedelta(seconds=temp_dict['planet_osm_rels'])
osm2pgsqldict['planet_osm_ways'] = osm2pgsqldict['planet_osm_rels'] - timedelta(seconds=temp_dict['planet_osm_ways'])
return osm2pgsqldict
def read_df(filename, start_time):
dfdict = {}
startsize = -1
with open(filename) as ifs:
for line in ifs:
line = line.strip()
if line.startswith('/'):
cols = line.split()
if cols[0] not in dfdict:
dfdict[cols[0]] = []
if cols[5] == '/':
if startsize < 0:
startsize = float(cols[2])
col = float(cols[2]) - startsize
else:
col = float(cols[2])
dfdict[cols[0]].append(float(col)/1024.0/1024.0)
else:
dt = datetime.strptime(line, "%a %b %d %H:%M:%S %Z %Y")
dt = dt - start_time
if 'datetime' not in dfdict:
dfdict['datetime'] = [dt.total_seconds()]
else:
dfdict['datetime'].append(dt.total_seconds())
return dfdict
def read_du(filename, start_time):
dudict = {}
headers = []
with open(filename) as csvfile:
reader = csv.reader(csvfile, delimiter=',')
header = reader.next()
for h in header:
if h != '':
headers.append(h)
dudict[h] = []
for row in reader:
for i, col in enumerate(row):
if col != '':
if headers[i] == 'time':
dt = datetime.strptime(col, "%a %b %d %H:%M:%S %Z %Y")
dt = dt - start_time
dudict[headers[i]].append(dt.total_seconds())
else:
dudict[headers[i]].append(float(col)/1024.0/1024.0)
return dudict
def read_dstat(filename, start_time):
dstatdict = {}
with open(filename) as csvfile:
reader = csv.reader(csvfile, delimiter=',')
# Skip the first five lines
reader.next()
reader.next()
reader.next()
reader.next()
reader.next()
# lines 6 and 7 contain the header information
header0 = reader.next()
header1 = reader.next()
headers = []
for i, (hh, h1) in enumerate(zip(header0, header1)):
if hh != '':
h0 = hh
if h0 not in dstatdict:
dstatdict[h0] = {}
if h1 not in dstatdict[h0]:
dstatdict[h0][h1] = []
else:
dstatdict[h0][h1] = []
if h1 == "time":
time_col = i
headers.append([h0, h1])
for row in reader:
for i, col in enumerate(row):
if i == time_col:
dstatdict[headers[i][0]][headers[i][1]].append(dstat_format_func(headers[i], col, st=start_time))
else:
dstatdict[headers[i][0]][headers[i][1]].append(dstat_format_func(headers[i], col))
return dstatdict
run_idx = 'benchmarks/f10'
osm2pgsql_file = run_idx+'/osm2pgsql.txt'
osm2pgsql_dict = read_osm2pgsql(osm2pgsql_file)
start_time = osm2pgsql_dict['Start time']
df_file = run_idx+'/df.txt'
if os.path.exists(df_file):
df_dict = read_df(df_file, start_time)
# df_dates = date2num(df_dict['datetime'])
else:
df_dict = None
du_file = run_idx+'/du.txt'
if os.path.exists(du_file):
du_dict = read_du(du_file, start_time)
# du_dates = date2num(du_dict['time'])
else:
du_dict = None
dstat_file = run_idx+'/dstat.txt'
if os.path.exists(dstat_file):
dstat_dict = read_dstat(dstat_file, start_time)
# dstat_dates = date2num(dstat_dict['system']['time'])
else:
dstat_dict = None
ymaxs = [250, 500, 250, 250, 250, 128, 100, 25, 20]
fig, ax = plt.subplots(len(ymaxs), sharex=True, figsize=(16, 10))
if df_dict is not None:
if 'dsk/sdb8' in dstat_dict:
ax[0].plot(df_dict['datetime'], df_dict['/dev/sdb8'], 'r-', lw=1)
ax[0].plot(df_dict['datetime'], df_dict['/dev/sdb6'], 'g-', lw=2)
if dstat_dict is not None:
if 'dsk/sdc1' in dstat_dict:
ax[1].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc1']['writ'], 'y-')
ax[1].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc1']['read'], 'b-')
if df_dict is not None:
if 'dsk/sdc1' in dstat_dict:
ax[1].plot(df_dict['datetime'], df_dict['/dev/sdc1'], 'g-', lw=2)
if du_dict is not None:
ax[1].plot(du_dict['time'], du_dict['main_data'], 'k-', lw=2)
if dstat_dict is not None:
if 'dsk/sdc2' in dstat_dict:
ax[2].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc2']['writ'], 'y-')
ax[2].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc2']['read'], 'b-')
if df_dict is not None:
if 'dsk/sdc2' in dstat_dict:
ax[2].plot(df_dict['datetime'], df_dict['/dev/sdc2'], 'g-', lw=2)
if du_dict is not None:
ax[2].plot(du_dict['time'], du_dict['main_index'], 'k-', lw=2)
if dstat_dict is not None:
if 'dsk/sdc3' in dstat_dict:
ax[3].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc3']['writ'], 'y-')
ax[3].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc3']['read'], 'b-')
if df_dict is not None:
if 'dsk/sdc3' in dstat_dict:
ax[3].plot(df_dict['datetime'], df_dict['/dev/sdc3'], 'g-', lw=2)
if du_dict is not None:
ax[3].plot(du_dict['time'], du_dict['slim_data'], 'k-', lw=2)
if dstat_dict is not None:
if 'dsk/sdc4' in dstat_dict:
ax[4].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc4']['writ'], 'y-')
ax[4].plot(dstat_dict['system']['time'], dstat_dict['dsk/sdc4']['read'], 'b-')
if df_dict is not None:
if 'dsk/sdc4' in dstat_dict:
ax[4].plot(df_dict['datetime'], df_dict['/dev/sdc4'], 'g-', lw=2)
if du_dict is not None:
ax[4].plot(du_dict['time'], du_dict['slim_index'], 'k-', lw=2)
if dstat_dict is not None:
ax[5].plot(dstat_dict['system']['time'], dstat_dict['memory usage']['used'], 'g-', lw=2)
ax[5].plot(dstat_dict['system']['time'], dstat_dict['memory usage']['cach'], 'b-', lw=2)
ax[5].plot(dstat_dict['system']['time'], dstat_dict['memory usage']['free'], 'k-', lw=2)
if dstat_dict is not None:
ax[6].plot(dstat_dict['system']['time'], dstat_dict['total cpu usage']['usr'], '-')
ax[6].plot(dstat_dict['system']['time'], dstat_dict['total cpu usage']['sys'], '-')
if dstat_dict is not None:
if 'swap' in dstat_dict:
ax[7].plot(dstat_dict['system']['time'], dstat_dict['swap']['used'], '-')
if dstat_dict is not None:
if 'paging' in dstat_dict:
ax[8].plot(dstat_dict['system']['time'], dstat_dict['paging']['out'], 'r-')
ax[8].plot(dstat_dict['system']['time'], dstat_dict['paging']['in'], 'b-')
# ax[8].plot(dstat_dict['system']['time'], dstat_dict['memory usage']['buff'], '-')
# ax[9].plot(dstat_dict['system']['time'], dstat_dict['total cpu usage']['wai'], '-', color='r')
# ax[9].plot(dstat_dict['system']['time'], dstat_dict['total cpu usage']['hiq'], '-')
# ax[9].plot(dstat_dict['system']['time'], dstat_dict['total cpu usage']['siq'], '-')
# ax[10].plot(dstat_dict['system']['time'], dstat_dict['load avg']['1m'], '-')
# ax[10].plot(dstat_dict['system']['time'], dstat_dict['load avg']['5m'], '-')
# ax[10].plot(dstat_dict['system']['time'], dstat_dict['load avg']['15m'], '-')
for i, ymax in enumerate(ymaxs):
for key, value in osm2pgsql_dict.iteritems():
value = value - start_time
ax[i].vlines(value.total_seconds(), 0, ymax, label=key)
ax[i].set_ylim([0, ymax])
ax[i].set_xlim([-300, 72000])
fig.subplots_adjust(hspace=0)
plt.setp([a.get_xticklabels() for a in fig.axes[:-1]], visible=False)
plt.show()