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
290
291
292
|
from __future__ import print_function
import functools
import hashlib
import logging
import os
import shutil
import sys
import time
from azure.datalake.store import core, multithread
from azure.datalake.store.transfer import ADLTransferClient
from azure.datalake.store.utils import WIN
from tests.testing import md5sum
def benchmark(f):
@functools.wraps(f)
def wrapped(*args, **kwargs):
print('[%s] starting...' % (f.__name__))
start = time.time()
result = f(*args, **kwargs)
stop = time.time()
elapsed = stop - start
print('[%s] finished in %2.4fs' % (f.__name__, elapsed))
return result, elapsed
return wrapped
def mock_client(adl, nthreads):
def transfer(adlfs, src, dst, offset, size, buffersize, blocksize, shutdown_event=None):
pass
def merge(adlfs, outfile, files, shutdown_event=None):
pass
return ADLTransferClient(
adl,
'foo',
transfer=transfer,
merge=merge,
nthreads=nthreads)
def checksum(path):
""" Generate checksum for file/directory content """
if not os.path.exists(path):
return None
if os.path.isfile(path):
return md5sum(path)
partial_sums = []
for root, dirs, files in os.walk(path):
for f in files:
filename = os.path.join(root, f)
if os.path.exists(filename):
partial_sums.append(str.encode(md5sum(filename)))
return hashlib.md5(b''.join(sorted(partial_sums))).hexdigest()
def du(path):
""" Find total size of content used by path """
if os.path.isfile(path):
return os.path.getsize(path)
size = 0
for root, dirs, files in os.walk(path):
for f in files:
size += os.path.getsize(os.path.join(root, f))
return size
def verify(instance):
""" Confirm whether target file matches source file """
adl = instance.client._adlfs
lfile = instance.lpath
rfile = instance.rpath
print("finish w/o error:", instance.successful())
print("local file :", lfile)
if os.path.exists(lfile):
print("local file size :", du(lfile))
else:
print("local file size :", None)
print("remote file :", rfile)
if adl.exists(rfile, invalidate_cache=False):
print("remote file size:", adl.du(rfile, total=True, deep=True))
else:
print("remote file size:", None)
@benchmark
def bench_upload_1_50gb(adl, lpath, rpath, config):
return multithread.ADLUploader(
adl,
lpath=lpath,
rpath=rpath,
**config[bench_upload_1_50gb.__name__])
@benchmark
def bench_upload_50_1gb(adl, lpath, rpath, config):
return multithread.ADLUploader(
adl,
lpath=lpath,
rpath=rpath,
**config[bench_upload_50_1gb.__name__])
@benchmark
def bench_download_1_50gb(adl, lpath, rpath, config):
return multithread.ADLDownloader(
adl,
lpath=lpath,
rpath=rpath,
**config[bench_download_1_50gb.__name__])
@benchmark
def bench_download_50_1gb(adl, lpath, rpath, config):
return multithread.ADLDownloader(
adl,
lpath=lpath,
rpath=rpath,
**config[bench_download_50_1gb.__name__])
def setup_logging(level='INFO'):
""" Log only Azure messages, ignoring 3rd-party libraries """
levels = dict(
CRITICAL=logging.CRITICAL,
ERROR=logging.ERROR,
WARNING=logging.WARNING,
INFO=logging.INFO,
DEBUG=logging.DEBUG)
if level in levels:
level = levels[level]
else:
raise ValueError('invalid log level: {}'.format(level))
logging.basicConfig(
format='%(asctime)s %(name)-17s %(levelname)-8s %(message)s')
logger = logging.getLogger('azure.datalake.store')
logger.setLevel(level)
def print_summary_statistics(stats):
from statistics import mean, median, pstdev
print("benchmark min mean sd median max")
for benchmark, samples in stats.items():
if samples:
metrics = [int(round(fn(samples), 0)) for fn in [min, mean, pstdev, median, max]]
else:
metrics = [0, 0, 0, 0, 0]
print(benchmark, *metrics)
if __name__ == '__main__':
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('local_path', type=str)
parser.add_argument('remote_path', type=str)
parser.add_argument('-l', '--log-level', default='INFO')
parser.add_argument('-n', '--iterations', default=1, type=int)
parser.add_argument('-q', '--quiet', dest='verbose', action='store_false')
parser.add_argument('-s', '--statistics', action='store_true')
parser.add_argument('--no-verify', dest='verify', action='store_false')
parser.add_argument('--no-checksum', dest='validate', action='store_false')
args = parser.parse_args(sys.argv[1:])
setup_logging(level=args.log_level)
adl = core.AzureDLFileSystem()
# Required setup until outstanding issues are resolved
adl.mkdir(args.remote_path)
# OS-specific settings
if WIN:
config = {
'bench_upload_1_50gb': {
'nthreads': 64,
'buffersize': 32 * 2**20,
'blocksize': 4 * 2**20
},
'bench_upload_50_1gb': {
'nthreads': 64,
'buffersize': 32 * 2**20,
'blocksize': 4 * 2**20
},
'bench_download_1_50gb': {
'nthreads': 64,
'buffersize': 32 * 2**20,
'blocksize': 4 * 2**20
},
'bench_download_50_1gb': {
'nthreads': 64,
'buffersize': 32 * 2**20,
'blocksize': 4 * 2**20
}
}
else:
config = {
'bench_upload_1_50gb': {
'nthreads': 64,
'buffersize': 4 * 2**20,
'blocksize': 4 * 2**20
},
'bench_upload_50_1gb': {
'nthreads': 64,
'buffersize': 4 * 2**20,
'blocksize': 4 * 2**20
},
'bench_download_1_50gb': {
'nthreads': 16,
'buffersize': 4 * 2**20,
'blocksize': 4 * 2**20
},
'bench_download_50_1gb': {
'nthreads': 16,
'buffersize': 4 * 2**20,
'blocksize': 4 * 2**20
}
}
for benchmark in config:
config[benchmark]['verbose'] = args.verbose
stats = {}
for _ in range(args.iterations):
# Upload/download 1 50GB files
lpath_up = os.path.join(args.local_path, '50gbfile.txt')
lpath_down = os.path.join(args.local_path, '50gbfile.txt.out')
rpath = args.remote_path + '/50gbfile.txt'
if adl.exists(rpath, invalidate_cache=False):
adl.rm(rpath)
if os.path.exists(lpath_down):
os.remove(lpath_down)
result, elapsed = bench_upload_1_50gb(adl, lpath_up, rpath, config)
if args.verify:
verify(result)
if result.successful:
stats.setdefault('up-1-50gb', []).append(elapsed)
result, elapsed = bench_download_1_50gb(adl, lpath_down, rpath, config)
if args.verify:
verify(result)
if result.successful:
stats.setdefault('down-1-50gb', []).append(elapsed)
if args.validate:
print(checksum(lpath_up), lpath_up)
print(checksum(lpath_down), lpath_down)
# Upload/download 50 1GB files
lpath_up = os.path.join(args.local_path, '50_1GB_Files')
lpath_down = os.path.join(args.local_path, '50_1GB_Files.out')
rpath = args.remote_path + '/50_1GB_Files'
if adl.exists(rpath):
adl.rm(rpath, recursive=True)
if os.path.exists(lpath_down):
shutil.rmtree(lpath_down)
result, elapsed = bench_upload_50_1gb(adl, lpath_up, rpath, config)
if args.verify:
verify(result)
if result.successful:
stats.setdefault('up-50-1gb', []).append(elapsed)
result, elapsed = bench_download_50_1gb(adl, lpath_down, rpath, config)
if args.verify:
verify(result)
if result.successful:
stats.setdefault('down-50-1gb', []).append(elapsed)
if args.validate:
print(checksum(lpath_up), lpath_up)
print(checksum(lpath_down), lpath_down)
if args.statistics:
print_summary_statistics(stats)
|
