/
launch.py
777 lines (668 loc) · 41.7 KB
/
launch.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
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
# Copyright 2019 Uber Technologies, Inc. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the 'License');
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an 'AS IS' BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
import argparse
import hashlib
import logging
import io
import os
import sys
import textwrap
import warnings
import yaml
import horovod
from horovod.common.util import (extension_available,
gloo_built, mpi_built,
nccl_built, ddl_built, ccl_built)
from horovod.runner.common.util import config_parser, hosts, safe_shell_exec, secret, timeout
from horovod.runner.common.util import settings as hvd_settings
from horovod.runner.driver import driver_service
from horovod.runner.elastic import settings as elastic_settings
from horovod.runner.elastic import discovery
from horovod.runner.util import cache, threads, network, lsf
from horovod.runner.gloo_run import gloo_run, gloo_run_elastic
from horovod.runner.mpi_run import mpi_run
from horovod.runner.js_run import js_run, is_jsrun_installed
from horovod.runner.http.http_client import read_data_from_kvstore, put_data_into_kvstore
from horovod.runner.http.http_server import KVStoreServer
from horovod.runner.util.remote import get_remote_command
# Cached information of horovod functions be stored in this directory
CACHE_FOLDER = os.path.join(os.path.expanduser('~'), '.horovod')
# Cache entries will be stale if they are older than this number of minutes
CACHE_STALENESS_THRESHOLD_MINUTES = 60
# Number of attempts for sshing into the hosts
SSH_ATTEMPTS = 5
SSH_CONNECT_TIMEOUT_S = 10
@cache.use_cache()
def _check_all_hosts_ssh_successful(host_addresses, ssh_port=None, ssh_identity_file=None):
"""
checks if ssh can successfully be performed to all the hosts.
:param host_addresses: list of addresses to ssh into. for example,
['worker-0','worker-1']
['10.11.11.11', '10.11.11.12']
:type host_addresses: list(strings)
:return: Returns True if all ssh was successful into all the addresses.
"""
def exec_command(command):
exit_code = 1
output_msg = ''
# Try ssh 5 times
for i in range(SSH_ATTEMPTS):
output = io.StringIO()
try:
exit_code = safe_shell_exec.execute(command,
stdout=output,
stderr=output)
if exit_code == 0:
break
output_msg = output.getvalue()
finally:
output.close()
return exit_code, output_msg
args_list = [[get_remote_command(local_command='true',
host=host_address,
port=ssh_port,
identity_file=ssh_identity_file,
timeout_s=SSH_CONNECT_TIMEOUT_S)]
for host_address in host_addresses]
ssh_exit_codes = \
threads.execute_function_multithreaded(exec_command,
args_list)
ssh_successful_to_all_hosts = True
for index, ssh_status in ssh_exit_codes.items():
exit_code, output_msg = ssh_status[0], ssh_status[1]
if exit_code != 0:
print('ssh not successful for host {host}:\n{msg_output}'
.format(host=host_addresses[index],
msg_output=output_msg))
ssh_successful_to_all_hosts = False
if not ssh_successful_to_all_hosts:
return None # we could return False here but do not want it to be cached
return True
def check_build(verbose):
def get_check(value):
return 'X' if value else ' '
output = '''{verbose_newline}\
Horovod v{version}:
Available Frameworks:
[{tensorflow}] TensorFlow
[{torch}] PyTorch
[{mxnet}] MXNet
Available Controllers:
[{mpi}] MPI
[{gloo}] Gloo
Available Tensor Operations:
[{nccl_ops}] NCCL
[{ddl_ops}] DDL
[{ccl_ops}] CCL
[{mpi_ops}] MPI
[{gloo_ops}] Gloo\
'''.format(verbose_newline='\n' if verbose else '',
version=horovod.__version__,
tensorflow=get_check(extension_available('tensorflow', verbose=verbose)),
torch=get_check(extension_available('torch', verbose=verbose)),
mxnet=get_check(extension_available('mxnet', verbose=verbose)),
mpi=get_check(mpi_built(verbose=verbose)),
gloo=get_check(gloo_built(verbose=verbose)),
nccl_ops=get_check(nccl_built(verbose=verbose)),
ddl_ops=get_check(ddl_built(verbose=verbose)),
mpi_ops=get_check(mpi_built(verbose=verbose)),
ccl_ops=get_check(ccl_built(verbose=verbose)),
gloo_ops=get_check(gloo_built(verbose=verbose)))
print(textwrap.dedent(output))
os._exit(0)
def make_check_build_action(np_arg):
class CheckBuildAction(argparse.Action):
def __call__(self, parser, args, values, option_string=None):
# If -cb is specified, make -np optional
np_arg.required = False
args.check_build = True
return CheckBuildAction
def make_override_action(override_args):
class StoreOverrideAction(argparse.Action):
def __init__(self,
option_strings,
dest,
default=None,
type=None,
choices=None,
required=False,
help=None):
super(StoreOverrideAction, self).__init__(
option_strings=option_strings,
dest=dest,
nargs=1,
default=default,
type=type,
choices=choices,
required=required,
help=help)
def __call__(self, parser, args, values, option_string=None):
override_args.add(self.dest)
setattr(args, self.dest, values[0])
return StoreOverrideAction
def make_override_bool_action(override_args, bool_value):
class StoreOverrideBoolAction(argparse.Action):
def __init__(self,
option_strings,
dest,
required=False,
help=None):
super(StoreOverrideBoolAction, self).__init__(
option_strings=option_strings,
dest=dest,
const=bool_value,
nargs=0,
default=None,
required=required,
help=help)
def __call__(self, parser, args, values, option_string=None):
override_args.add(self.dest)
setattr(args, self.dest, self.const)
return StoreOverrideBoolAction
def make_override_true_action(override_args):
return make_override_bool_action(override_args, True)
def make_override_false_action(override_args):
return make_override_bool_action(override_args, False)
def make_deprecated_bool_action(override_args, bool_value, replacement_option):
class StoreOverrideBoolAction(argparse.Action):
def __init__(self,
option_strings,
dest,
required=False,
help=None):
super(StoreOverrideBoolAction, self).__init__(
option_strings=option_strings,
dest=dest,
const=bool_value,
nargs=0,
default=None,
required=required,
help=help)
def __call__(self, parser, args, values, option_string=None):
deprecated_option = '|'.join(self.option_strings)
warnings.warn(f'Argument {deprecated_option} has been replaced by {replacement_option} and will be removed in v0.21.0',
DeprecationWarning)
override_args.add(self.dest)
setattr(args, self.dest, self.const)
return StoreOverrideBoolAction
def parse_args():
override_args = set()
parser = argparse.ArgumentParser(description='Horovod Runner')
parser.add_argument('-v', '--version', action='version', version=horovod.__version__,
help='Shows Horovod version.')
np_arg = parser.add_argument('-np', '--num-proc', action='store', dest='np',
type=int, required=not lsf.LSFUtils.using_lsf(),
help='Total number of training processes. In elastic mode, '
'number of processes required before training can start.')
parser.add_argument('-cb', '--check-build', action=make_check_build_action(np_arg), nargs=0,
help='Shows which frameworks and libraries have been built into Horovod.')
parser.add_argument('--disable-cache', action='store_true',
dest='disable_cache',
help='If the flag is not set, horovodrun will perform '
'the initialization checks only once every 60 '
'minutes -- if the checks successfully pass. '
'Otherwise, all the checks will run every time '
'horovodrun is called.')
parser.add_argument('--start-timeout', action='store',
dest='start_timeout', type=int,
help='Horovodrun has to perform all the checks and '
'start the processes before the specified '
'timeout. The default value is 30 seconds. '
'Alternatively, The environment variable '
'HOROVOD_START_TIMEOUT can also be used to '
'specify the initialization timeout.')
parser.add_argument('--network-interface', action='store', dest='nics',
help='Network interfaces that can be used for communication separated by '
'comma. If not specified, Horovod will find the common NICs among all '
'the workers and use it; example, --network-interface "eth0,eth1".')
parser.add_argument('--output-filename', action='store',
help='For Gloo, writes stdout / stderr of all processes to a filename of the form '
'<output_filename>/rank.<rank>/<stdout | stderr>. The <rank> will be padded with 0 '
'characters to ensure lexicographical order. For MPI, delegates its behavior to mpirun.')
parser.add_argument('--verbose', action='store_true',
dest='verbose',
help='If this flag is set, extra messages will '
'be printed.')
parser.add_argument('command', nargs=argparse.REMAINDER,
help='Command to be executed.')
parser.add_argument('--config-file', action='store', dest='config_file',
help='Path to YAML file containing runtime parameter configuration for Horovod. '
'Note that this will override any command line arguments provided before '
'this argument, and will be overridden by any arguments that come after it.')
group_ssh = parser.add_argument_group('SSH arguments')
group_ssh.add_argument('-p', '--ssh-port', action='store', dest='ssh_port',
type=int, help='SSH port on all the hosts.')
group_ssh.add_argument('-i', '--ssh-identity-file', action='store', dest='ssh_identity_file',
help='File on the driver from which the identity (private key) is read.')
group_params = parser.add_argument_group('tuneable parameter arguments')
group_params.add_argument('--fusion-threshold-mb', action=make_override_action(override_args), type=int,
help='Fusion buffer threshold in MB. This is the maximum amount of '
'tensor data that can be fused together into a single batch '
'during allreduce / allgather. Setting 0 disables tensor fusion. '
'(default: 128)')
group_params.add_argument('--cycle-time-ms', action=make_override_action(override_args), type=float,
help='Cycle time in ms. This is the delay between each tensor fusion '
'cycle. The larger the cycle time, the more batching, but the '
'greater latency between each allreduce / allgather operations. '
'(default: 1')
group_params.add_argument('--cache-capacity', action=make_override_action(override_args), type=int,
help='Maximum number of tensor names that will be cached to reduce amount '
'of coordination required between workers before performing allreduce / '
'allgather. (default: 1024')
group_hierarchical_allreduce = group_params.add_mutually_exclusive_group()
group_hierarchical_allreduce.add_argument('--hierarchical-allreduce',
action=make_override_true_action(override_args),
help='Perform hierarchical allreduce between workers instead of '
'ring allreduce. Hierarchical allreduce performs a local '
'allreduce / gather within a host, then a parallel cross allreduce '
'between equal local ranks across workers, and finally a '
'local gather.')
group_hierarchical_allreduce.add_argument('--no-hierarchical-allreduce', dest='hierarchical_allreduce',
action=make_override_false_action(override_args),
help='Explicitly disable hierarchical allreduce to prevent autotuning '
'from adjusting it.')
group_hierarchical_allgather = group_params.add_mutually_exclusive_group()
group_hierarchical_allgather.add_argument('--hierarchical-allgather',
action=make_override_true_action(override_args),
help='Perform hierarchical allgather between workers instead of '
'ring allgather. See hierarchical allreduce for algorithm details.')
group_hierarchical_allgather.add_argument('--no-hierarchical-allgather', dest='hierarchical_allgather',
action=make_override_false_action(override_args),
help='Explicitly disable hierarchical allgather to prevent autotuning '
'from adjusting it.')
group_autotune = parser.add_argument_group('autotune arguments')
group_autotune_enabled = group_autotune.add_mutually_exclusive_group()
group_autotune_enabled.add_argument('--autotune', action=make_override_true_action(override_args),
help='Perform autotuning to select parameter argument values that maximimize '
'throughput for allreduce / allgather. Any parameter explicitly set will '
'be held constant during tuning.')
group_autotune_enabled.add_argument('--no-autotune', dest='autotune',
action=make_override_false_action(override_args), help=argparse.SUPPRESS)
group_autotune.add_argument('--autotune-log-file', action=make_override_action(override_args),
help='Comma-separated log of trials containing each hyperparameter and the '
'score of the trial. The last row will always contain the best value '
'found.')
group_autotune.add_argument('--autotune-warmup-samples', action=make_override_action(override_args),
type=int,
help='Number of samples to discard before beginning the optimization process '
'during autotuning. Performance during the first few batches can be '
'affected by initialization and cache warmups. (default: 3')
group_autotune.add_argument('--autotune-steps-per-sample', action=make_override_action(override_args),
type=int,
help='Number of steps (approximate) to record before observing a sample. The sample '
'score is defined to be the median score over all batches within the sample. The '
'more batches per sample, the less variance in sample scores, but the longer '
'autotuning will take. (default: 10')
group_autotune.add_argument('--autotune-bayes-opt-max-samples', action=make_override_action(override_args),
type=int,
help='Maximum number of samples to collect for each Bayesian optimization process. '
'(default: 20')
group_autotune.add_argument('--autotune-gaussian-process-noise', action=make_override_action(override_args),
type=float,
help='Regularization value [0, 1] applied to account for noise in samples. '
'(default: 0.8')
group_elastic = parser.add_argument_group('elastic arguments')
group_elastic.add_argument('--min-np', action='store', dest='min_np', type=int,
help='Minimum number of processes running for training to continue. If number of '
'available processes dips below this threshold, then training will wait for '
'more instances to become available. Defaults to --num-proc.')
group_elastic.add_argument('--max-np', action='store', dest='max_np', type=int,
help='Maximum number of training processes, beyond which no additional '
'processes will be created. If not specified, then will be unbounded.')
group_elastic.add_argument('--slots-per-host', action='store', dest='slots', type=int,
help='Number of slots for processes per host. Normally 1 slot per GPU per host. '
'If slots are provided by the output of the host discovery script, then '
'that value will override this parameter.')
group_elastic.add_argument('--elastic-timeout', action='store', dest='elastic_timeout', type=int,
help='Timeout for elastic initialisation after re-scaling the cluster. '
'The default value is 600 seconds. Alternatively, '
'The environment variable HOROVOD_ELASTIC_TIMEOUT '
'can also be used to.')
group_elastic.add_argument('--reset-limit', action='store', dest='reset_limit', type=int,
help='Maximum number of times that the training job can scale up or down '
'the number of workers after which the job is terminated. (default: None)')
group_elastic.add_argument('--blacklist-cooldown-range', action='store', dest='cooldown_range', type=int, nargs=2,
help='Range of seconds(min, max) a failing host will remain in blacklist. (default: None)')
group_timeline = parser.add_argument_group('timeline arguments')
group_timeline.add_argument('--timeline-filename', action=make_override_action(override_args),
help='JSON file containing timeline of Horovod events used for debugging '
'performance. If this is provided, timeline events will be recorded, '
'which can have a negative impact on training performance.')
group_timeline_cycles = group_timeline.add_mutually_exclusive_group()
group_timeline_cycles.add_argument('--timeline-mark-cycles', action=make_override_true_action(override_args),
help='Mark cycles on the timeline. Only enabled if the timeline filename '
'is provided.')
group_timeline_cycles.add_argument('--no-timeline-mark-cycles', dest='timeline_mark_cycles',
action=make_override_false_action(override_args), help=argparse.SUPPRESS)
group_stall_check = parser.add_argument_group('stall check arguments')
group_stall_check_enabled = group_stall_check.add_mutually_exclusive_group()
group_stall_check_enabled.add_argument('--no-stall-check', action=make_override_true_action(override_args),
help='Disable the stall check. The stall check will log a warning when '
'workers have stalled waiting for other ranks to submit tensors.')
group_stall_check_enabled.add_argument('--stall-check', dest='no_stall_check',
action=make_override_false_action(override_args), help=argparse.SUPPRESS)
group_stall_check.add_argument('--stall-check-warning-time-seconds', action=make_override_action(override_args),
type=int,
help='Seconds until the stall warning is logged to stderr. (default: 60')
group_stall_check.add_argument('--stall-check-shutdown-time-seconds', action=make_override_action(override_args),
type=int,
help='Seconds until Horovod is shutdown due to stall. Shutdown will only take '
'place if this value is greater than the warning time. (default: 0')
group_library_options = parser.add_argument_group('library arguments')
group_mpi_threads_disable = group_library_options.add_mutually_exclusive_group()
group_mpi_threads_disable.add_argument('--mpi-threads-disable', action=make_override_true_action(override_args),
help='Disable MPI threading support. Only applies when running in MPI '
'mode. In some cases, multi-threaded MPI can slow down other '
'components, but is necessary if you wish to run mpi4py on top '
'of Horovod.')
group_mpi_threads_disable.add_argument('--no-mpi-threads-disable', dest='mpi_threads_disable',
action=make_override_false_action(override_args), help=argparse.SUPPRESS)
group_library_options.add_argument('--mpi-args', action='store', dest='mpi_args',
help='Extra MPI arguments to pass to mpirun. '
'They need to be passed with the equal sign to avoid parsing issues. '
'e.g. --mpi-args="--map-by ppr:6:node"')
group_library_options.add_argument('--tcp', action='store_true', dest='tcp_flag',
help='If this flag is set, only TCP is used for communication.')
group_library_options.add_argument('--binding-args', action='store', dest='binding_args',
help='Process binding arguments. Default is socket for Spectrum MPI '
'and no binding for other cases. e.g. --binding-args="--rankfile myrankfile"')
group_library_options.add_argument('--num-nccl-streams', action=make_override_action(override_args),
type=int,
help='Number of NCCL streams. Only applies when running with NCCL support. '
'(default: %(default)s)')
group_library_options.add_argument('--thread-affinity', action=make_override_action(override_args),
type=int,
help='Horovod background thread affinity. '
'(default: 0')
group_library_options.add_argument('--gloo-timeout-seconds', action=make_override_action(override_args),
type=int,
help='Timeout in seconds for Gloo operations to complete. '
'(default: 30')
group_logging = parser.add_argument_group('logging arguments')
group_logging.add_argument('--log-level', action=make_override_action(override_args),
choices=config_parser.LOG_LEVELS,
help='Minimum level to log to stderr from the Horovod backend. (default: WARNING).')
group_logging_timestamp = group_logging.add_mutually_exclusive_group()
group_logging_timestamp.add_argument('--log-with-timestamp',
action=make_override_true_action(override_args),
help=argparse.SUPPRESS)
group_logging_timestamp.add_argument('--log-without-timestamp', dest='log_with_timestamp',
action=make_override_false_action(override_args),
help='Hide the timestamp from Horovod internal log messages.')
group_logging_timestamp.add_argument('-prefix-timestamp', '--prefix-output-with-timestamp', action='store_true',
dest='prefix_output_with_timestamp',
help='Timestamp each line of output to stdout, stderr, and stddiag.')
group_logging_timestamp.add_argument('--log-hide-timestamp',
dest='log_with_timestamp',
action=make_deprecated_bool_action(override_args, False, '--log-without-timestamp'),
help=argparse.SUPPRESS)
group_logging_timestamp.add_argument('--no-log-hide-timestamp',
dest='log_with_timestamp',
action=make_deprecated_bool_action(override_args, True, '--log-with-timestamp'),
help=argparse.SUPPRESS)
group_hosts_parent = parser.add_argument_group('host arguments')
group_hosts = group_hosts_parent.add_mutually_exclusive_group()
group_hosts.add_argument('-H', '--hosts', action='store', dest='hosts',
help='List of host names and the number of available slots '
'for running processes on each, of the form: <hostname>:<slots> '
'(e.g.: host1:2,host2:4,host3:1 indicating 2 processes can run on host1, '
'4 on host2, and 1 on host3). If not specified, defaults to using '
'localhost:<np>')
group_hosts.add_argument('-hostfile', '--hostfile', action='store', dest='hostfile',
help='Path to a host file containing the list of host names and the number of '
'available slots. Each line of the file must be of the form: '
'<hostname> slots=<slots>')
group_hosts.add_argument('--host-discovery-script', action=make_override_action(override_args),
help='Used for elastic training (autoscaling and fault tolerance). '
'An executable script that will print to stdout every available host (one per '
'newline character) that can be used to run worker processes. Optionally '
'specifies number of slots on the same line as the hostname as: "hostname:slots".'
'Providing a discovery script enables elastic training (see elastic arguments).'
'The job will fail immediately if execution of the script returns a non-zero exit '
'code on the first call. Subsequent calls will be retried until timeout.')
group_controller_parent = parser.add_argument_group('controller arguments')
group_controller = group_controller_parent.add_mutually_exclusive_group()
group_controller.add_argument('--gloo', action='store_true', dest='use_gloo',
help='Run Horovod using the Gloo controller. This will '
'be the default if Horovod was not built with MPI support.')
group_controller.add_argument('--mpi', action='store_true', dest='use_mpi',
help='Run Horovod using the MPI controller. This will '
'be the default if Horovod was built with MPI support.')
group_controller.add_argument('--jsrun', action='store_true', dest='use_jsrun',
help='Launch Horovod processes with jsrun and use the MPI controller. '
'This will be the default if jsrun is installed and Horovod '
'was built with MPI support.')
args = parser.parse_args()
if args.config_file:
with open(args.config_file, 'r') as f:
config = yaml.load(f, Loader=yaml.FullLoader)
config_parser.set_args_from_config(args, config, override_args)
config_parser.validate_config_args(args)
args.run_func = None
args.executable = None
if args.check_build:
check_build(args.verbose)
return args
def _run_static(args):
# horovodrun has to finish all the checks before this timeout runs out.
if args.start_timeout:
start_timeout = args.start_timeout
else:
# Lookup default timeout from the environment variable.
start_timeout = int(os.getenv('HOROVOD_START_TIMEOUT', '30'))
tmout = timeout.Timeout(start_timeout,
message='Timed out waiting for {activity}. Please '
'check connectivity between servers. You '
'may need to increase the --start-timeout '
'parameter if you have too many servers.')
settings = hvd_settings.Settings(verbose=2 if args.verbose else 0,
ssh_port=args.ssh_port,
ssh_identity_file=args.ssh_identity_file,
extra_mpi_args=args.mpi_args,
tcp_flag=args.tcp_flag,
binding_args=args.binding_args,
key=secret.make_secret_key(),
start_timeout=tmout,
num_proc=args.np,
hosts=args.hosts,
output_filename=args.output_filename,
run_func_mode=args.run_func is not None,
nics=args.nics,
prefix_output_with_timestamp=args.prefix_output_with_timestamp)
# This cache stores the results of checks performed by horovod
# during the initialization step. It can be disabled by setting
# --disable-cache flag.
fn_cache = None
if not args.disable_cache:
params = ''
if args.np:
params += str(args.np) + ' '
if args.hosts:
params += str(args.hosts) + ' '
if args.ssh_port:
params += str(args.ssh_port)
if args.ssh_identity_file:
params += args.ssh_identity_file
parameters_hash = hashlib.md5(params.encode('utf-8')).hexdigest()
fn_cache = cache.Cache(CACHE_FOLDER, CACHE_STALENESS_THRESHOLD_MINUTES,
parameters_hash)
all_host_names, _ = hosts.parse_hosts_and_slots(args.hosts)
if settings.verbose >= 2:
print('Filtering local host names.')
remote_host_names = network.filter_local_addresses(all_host_names)
if settings.verbose >= 2:
print('Remote host found: ' + ' '.join(remote_host_names))
if len(remote_host_names) > 0:
if settings.verbose >= 2:
print('Checking ssh on all remote hosts.')
# Check if we can ssh into all remote hosts successfully.
if not _check_all_hosts_ssh_successful(remote_host_names, args.ssh_port, args.ssh_identity_file,
fn_cache=fn_cache):
raise RuntimeError('could not connect to some hosts via ssh')
if settings.verbose >= 2:
print('SSH was successful into all the remote hosts.')
nics = driver_service.get_common_interfaces(settings, all_host_names,
remote_host_names, fn_cache)
if args.run_func:
# get the driver IPv4 address
driver_ip = network.get_driver_ip(nics)
run_func_server = KVStoreServer(verbose=settings.verbose)
run_func_server_port = run_func_server.start_server()
put_data_into_kvstore(driver_ip, run_func_server_port,
'runfunc', 'func', args.run_func)
executable = args.executable or sys.executable
command = [executable, '-m', 'horovod.runner.run_task', str(driver_ip), str(run_func_server_port)]
try:
_launch_job(args, settings, nics, command)
results = [None] * args.np
# TODO: make it parallel to improve performance
for i in range(args.np):
results[i] = read_data_from_kvstore(driver_ip, run_func_server_port,
'runfunc_result', str(i))
return results
finally:
run_func_server.shutdown_server()
else:
command = args.command
_launch_job(args, settings, nics, command)
return None
def _run_elastic(args):
# construct host discovery component
if args.host_discovery_script:
discover_hosts = discovery.HostDiscoveryScript(args.host_discovery_script, args.slots)
elif args.hosts:
_, available_host_slots = hosts.parse_hosts_and_slots(args.hosts)
if len(available_host_slots) < 2:
raise ValueError('Cannot run in fault tolerance mode with fewer than 2 hosts.')
discover_hosts = discovery.FixedHosts(available_host_slots)
else:
raise ValueError('One of --host-discovery-script, --hosts, or --hostnames must be provided')
# horovodrun has to finish all the checks before this timeout runs out.
if args.start_timeout:
start_timeout = args.start_timeout
else:
# Lookup default timeout from the environment variable.
start_timeout = int(os.getenv('HOROVOD_START_TIMEOUT', '30'))
tmout = timeout.Timeout(start_timeout,
message='Timed out waiting for {activity}. Please '
'check connectivity between servers. You '
'may need to increase the --start-timeout '
'parameter if you have too many servers.')
settings = elastic_settings.ElasticSettings(discovery=discover_hosts,
min_np=args.min_np or args.np,
max_np=args.max_np,
elastic_timeout=args.elastic_timeout,
reset_limit=args.reset_limit,
cooldown_range=args.cooldown_range,
num_proc=args.np,
verbose=2 if args.verbose else 0,
ssh_port=args.ssh_port,
ssh_identity_file=args.ssh_identity_file,
extra_mpi_args=args.mpi_args,
key=secret.make_secret_key(),
start_timeout=tmout,
output_filename=args.output_filename,
run_func_mode=args.run_func is not None,
nics=args.nics,
prefix_output_with_timestamp=args.prefix_output_with_timestamp)
if not gloo_built(verbose=(settings.verbose >= 2)):
raise ValueError('Gloo support is required to use elastic training, but has not been built. Ensure CMake is '
'installed and reinstall Horovod with HOROVOD_WITH_GLOO=1 to debug the build error.')
env = os.environ.copy()
config_parser.set_env_from_args(env, args)
gloo_run_elastic(settings, env, args.command)
def is_gloo_used(use_gloo=None, use_mpi=None, use_jsrun=None):
# determines whether run_controller will run gloo
# for the given (use_gloo, _, use_mpi, _, use_jsrun, _, _)
return use_gloo or (not use_mpi and not use_jsrun and not mpi_built())
def run_controller(use_gloo, gloo_run, use_mpi, mpi_run, use_jsrun, js_run, verbosity):
# keep logic in sync with is_gloo_used(...)
verbose = verbosity is not None and verbosity >= 2
if use_gloo:
if not gloo_built(verbose=verbose):
raise ValueError('Gloo support has not been built. If this is not expected, ensure CMake is installed '
'and reinstall Horovod with HOROVOD_WITH_GLOO=1 to debug the build error.')
gloo_run()
elif use_mpi:
if not mpi_built(verbose=verbose):
raise ValueError('MPI support has not been built. If this is not expected, ensure MPI is installed '
'and reinstall Horovod with HOROVOD_WITH_MPI=1 to debug the build error.')
mpi_run()
elif use_jsrun:
if not mpi_built(verbose=verbose):
raise ValueError('MPI support has not been built. If this is not expected, ensure MPI is installed '
'and reinstall Horovod with HOROVOD_WITH_MPI=1 to debug the build error.')
if not lsf.LSFUtils.using_lsf():
raise ValueError(
'Horovod did not detect an LSF job. The jsrun launcher can only be used in that environment. '
'Please, pick a different launcher for other environments.')
js_run()
else:
if mpi_built(verbose=verbose):
if lsf.LSFUtils.using_lsf() and is_jsrun_installed():
js_run()
else:
mpi_run()
elif gloo_built(verbose=verbose):
gloo_run()
else:
raise ValueError('Neither MPI nor Gloo support has been built. Try reinstalling Horovod ensuring that '
'either MPI is installed (MPI) or CMake is installed (Gloo).')
def _is_elastic(args):
return args.host_discovery_script is not None or args.min_np is not None
def _launch_job(args, settings, nics, command):
env = os.environ.copy()
config_parser.set_env_from_args(env, args)
def gloo_run_fn():
driver_ip = network.get_driver_ip(nics)
gloo_run(settings, nics, env, driver_ip, command)
def mpi_run_fn():
mpi_run(settings, nics, env, command)
def js_run_fn():
js_run(settings, nics, env, command)
run_controller(args.use_gloo, gloo_run_fn,
args.use_mpi, mpi_run_fn,
args.use_jsrun, js_run_fn,
args.verbose)
def _run(args):
# If LSF is used, use default values from job config
if lsf.LSFUtils.using_lsf():
if not args.np:
args.np = lsf.LSFUtils.get_num_processes()
if not args.hosts and not args.hostfile and not args.host_discovery_script:
args.hosts = ','.join('{host}:{np}'.format(host=host, np=lsf.LSFUtils.get_num_gpus())
for host in lsf.LSFUtils.get_compute_hosts())
# if hosts are not specified, either parse from hostfile, or default as
# localhost
if not args.hosts and not args.host_discovery_script:
if args.hostfile:
args.hosts = hosts.parse_host_files(args.hostfile)
else:
# Set hosts to localhost if not specified
args.hosts = 'localhost:{np}'.format(np=args.np)
# Convert nics into set
args.nics = set(args.nics.split(',')) if args.nics else None
if _is_elastic(args):
return _run_elastic(args)
else:
return _run_static(args)
def run_commandline():
args = parse_args()
if args.log_level:
logging.addLevelName(logging.NOTSET, 'TRACE')
logging.basicConfig(level=logging.getLevelName(args.log_level))
_run(args)
if __name__ == '__main__':
run_commandline()