Files
35f4d29ed6562b0f4d5db253deeb218db62078a4
swift /bin /swift-bench

176 lines
6.9 KiB
Plaintext
Raw Normal View History

2011年03月24日 15:51:44 -07:00
#!/usr/bin/env python
# Copyright (c) 2010-2012 OpenStack, LLC.
2010年10月01日 19:50:34 +00:00
#
# 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.
2010年09月27日 23:10:09 +00:00
import logging
import os
import sys
import signal
import uuid
from optparse import OptionParser
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
from swift.common.bench import (BenchController, DistributedBenchController,
create_containers, delete_containers)
from swift.common.utils import readconf, LogAdapter, config_true_value
2010年09月27日 23:10:09 +00:00
2010年09月28日 15:05:08 +00:00
# The defaults should be sufficient to run swift-bench on a SAIO
2010年09月27日 23:10:09 +00:00
CONF_DEFAULTS = {
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
'auth': os.environ.get('ST_AUTH', ''),
'user': os.environ.get('ST_USER', ''),
'key': os.environ.get('ST_KEY', ''),
'auth_version': '1.0',
'use_proxy': 'yes',
2010年09月27日 23:10:09 +00:00
'put_concurrency': '10',
'get_concurrency': '10',
'del_concurrency': '10',
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
'concurrency': '', # set all 3 in one shot
'object_sources': '', # set of file contents to read and use for PUTs
'lower_object_size': '10', # bounded random size used if these differ
'upper_object_size': '10',
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
'object_size': '1', # only if not object_sources and lower == upper
2010年09月27日 23:10:09 +00:00
'num_objects': '1000',
'num_gets': '10000',
'delete': 'yes',
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
'container_name': uuid.uuid4().hex, # really "container name base"
'num_containers': '20',
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
'url': '', # used when use_proxy = no or overrides auth X-Storage-Url
'account': '', # used when use_proxy = no
'devices': 'sdb1', # space-sep list
'log_level': 'INFO',
2010年09月27日 23:10:09 +00:00
'timeout': '10',
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
'auth_version': '1.0',
'bench_clients': [],
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
}
2010年09月27日 23:10:09 +00:00
SAIO_DEFAULTS = {
'auth': 'http://localhost:8080/auth/v1.0',
'user': 'test:tester',
'key': 'testing',
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
}
2010年09月27日 23:10:09 +00:00
if __name__ == '__main__':
usage = "usage: %prog [OPTIONS] [CONF_FILE]"
usage += """\n\nConf file with SAIO defaults:
2010年09月27日 23:10:09 +00:00
[bench]
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
auth = http://localhost:8080/auth/v1.0
2010年09月27日 23:10:09 +00:00
user = test:tester
key = testing
concurrency = 10
object_size = 1
num_objects = 1000
num_gets = 10000
delete = yes
auth_version = 1.0
2010年09月27日 23:10:09 +00:00
"""
parser = OptionParser(usage=usage)
parser.add_option('', '--saio', dest='saio', action='store_true',
default=False, help='Run benchmark with SAIO defaults')
2010年09月27日 23:10:09 +00:00
parser.add_option('-A', '--auth', dest='auth',
help='URL for obtaining an auth token')
parser.add_option('-U', '--user', dest='user',
help='User name for obtaining an auth token')
parser.add_option('-K', '--key', dest='key',
help='Key for obtaining an auth token')
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
parser.add_option('-b', '--bench-clients', action='append',
metavar='<ip>:<port>',
help=('A string of the form "<ip>:<port>" which matches '
'the arguments supplied to a swift-bench-client '
'process. This argument must be specified '
'once per swift-bench-client you want to '
'utilize.'))
2010年09月27日 23:10:09 +00:00
parser.add_option('-u', '--url', dest='url',
help='Storage URL')
parser.add_option('-c', '--concurrency', dest='concurrency',
help='Number of concurrent connections to use')
parser.add_option('-s', '--object-size', dest='object_size',
help='Size of objects to PUT (in bytes)')
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
parser.add_option('-l', '--lower-object-size', dest='lower_object_size',
help=('Lower size of objects (in bytes); '
'--object-size will be upper-object-size'))
2010年09月27日 23:10:09 +00:00
parser.add_option('-n', '--num-objects', dest='num_objects',
help='Number of objects to PUT')
parser.add_option('-g', '--num-gets', dest='num_gets',
help='Number of GET operations to perform')
parser.add_option('-x', '--no-delete', dest='delete', action='store_false',
help='If set, will not delete the objects created')
parser.add_option('-V', '--auth_version', dest='auth_version',
help='Authentication version')
2010年09月27日 23:10:09 +00:00
if len(sys.argv) == 1:
parser.print_help()
sys.exit(1)
options, args = parser.parse_args()
if options.saio:
CONF_DEFAULTS.update(SAIO_DEFAULTS)
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
if getattr(options, 'lower_object_size', None):
if options.object_size <= options.lower_object_size:
raise ValueError('--lower-object-size (%s) must be '
'< --object-size (%s)' %
(options.lower_object_size, options.object_size))
CONF_DEFAULTS['upper_object_size'] = options.object_size
2010年09月27日 23:10:09 +00:00
if args:
conf = args[0]
if not os.path.exists(conf):
sys.exit("No such conf file: %s" % conf)
conf = readconf(conf, 'bench', log_name='swift-bench',
Misc. swift-bench improvements. swift-bench now honors the environment variables, ST_AUTH, ST_USER, and ST_KEY like python-swiftclient does. Added --lower-object-size (or -l) command-line option which, if specified, will turn a specified --object-size into --upper-object-size. It will raise a ValueError if --object-size is not specified or is <= --lower-object-size. BenchController how handles SIGINT (KeyboardInterrupt) similarly to the swift command-line client: the first Ctrl-C will make it fast-track to completion (no new PUT or GET operations are started, but everything PUT is DELETE'ed). A second Ctrl-C will immediately exit. The behavior for SIGTERM is unchanged (a single SIGTERM will immediately terminate the process). Added a sample configuration file for swift-bench, with documenting comments. Change-Id: I6f394ad995300fc8af3d565d95c3b45559ee510a
2012年08月25日 16:02:45 -07:00
defaults=CONF_DEFAULTS)
conf['bench_clients'] = []
2010年09月27日 23:10:09 +00:00
else:
conf = CONF_DEFAULTS
parser.set_defaults(**conf)
options, _junk = parser.parse_args()
2010年09月27日 23:10:09 +00:00
if options.concurrency is not '':
options.put_concurrency = options.concurrency
options.get_concurrency = options.concurrency
options.del_concurrency = options.concurrency
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
options.containers = ['%s_%d' % (options.container_name, i)
for i in xrange(int(options.num_containers))]
# check boolean options vs config parameter values
if config_true_value(str(options.delete).lower()):
options.delete = 'yes'
else:
options.delete = 'no'
2010年09月27日 23:10:09 +00:00
def sigterm(signum, frame):
sys.exit('Termination signal received.')
signal.signal(signal.SIGTERM, sigterm)
logger = logging.getLogger()
logger.setLevel({
'debug': logging.DEBUG,
'info': logging.INFO,
'warning': logging.WARNING,
'error': logging.ERROR,
'critical': logging.CRITICAL}.get(
options.log_level.lower(), logging.INFO))
2010年09月27日 23:10:09 +00:00
loghandler = logging.StreamHandler()
logger.addHandler(loghandler)
logger = LogAdapter(logger, 'swift-bench')
logformat = logging.Formatter('%(server)s %(asctime)s %(levelname)s '
'%(message)s')
loghandler.setFormatter(logformat)
2010年09月27日 23:10:09 +00:00
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
if options.use_proxy:
create_containers(logger, options)
controller_class = DistributedBenchController if options.bench_clients \
else BenchController
controller = controller_class(logger, options)
2010年09月27日 23:10:09 +00:00
controller.run()
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
if config_true_value(options.delete.lower()):
Can run swift-bench across multiple cores/servers. You run one or more swift-bench-client processes like this: $ swift-bench-client 127.0.0.1 20001 $ swift-bench-client 127.0.0.1 20002 Then you run swift-bench with a new option, --bench-clients (-b), which is specified once for each swift-bench-client: $ swift-bench -b 127.0.0.1:20001 -b 127.0.0.1:20002 You get log lines from each client (interleaved) along with a final report for all clients: 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:06,148 INFO Auth version: 1.0 127.0.0.1:20001 swift-bench-server 2012年08月25日 22:44:12,249 INFO 83 PUTS [0 failures], 41.5/s 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:44:14,430 INFO 74 PUTS [0 failures], 34.3/s ... 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:18,942 INFO Auth version: 1.0 127.0.0.1:20002 swift-bench-server 2012年08月25日 22:45:20,946 INFO 238 DEL [2 failures], 118.9/s swift-bench 2012年08月25日 22:45:27,549 INFO 2000 PUTS **FINAL** [0 failures], 56.8/s swift-bench 2012年08月25日 22:45:27,550 INFO 30000 GETS **FINAL** [50 failures], 974.6/s swift-bench 2012年08月25日 22:45:27,550 INFO 2000 DEL **FINAL** [20 failures], 237.1/s The concurrency, PUT count, and GET count config settings are divided by the number of bench_clients. In other words, the same volume of work is attempted (vs. not specifying --bench-clients), but it can now span servers and CPU cores. Benchmark containers are created (if use_proxy = yes) and deleted (if delete = yes), with appropriate concurrency, in the initiating swift-bench process, not any of the swift-bench-client processes. Change-Id: Idbf31a23093244ab357a9bf77e6031257774f24a
2012年08月25日 23:02:53 -07:00
delete_containers(logger, options)