1 .TH "MEMASLAP" "1" "April 08, 2011" "0.47" "libmemcached"
3 memaslap \- libmemcached Documentation
5 .nr rst2man-indent-level 0
9 level \\n[rst2man-indent-level]
10 level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
17 .\" .rstReportMargin pre:
19 . nr rst2man-indent\\n[rst2man-indent-level] \\n[an-margin]
20 . nr rst2man-indent-level +1
21 .\" .rstReportMargin post:
25 .\" indent \\n[an-margin]
26 .\" old: \\n[rst2man-indent\\n[rst2man-indent-level]]
27 .nr rst2man-indent-level -1
28 .\" new: \\n[rst2man-indent\\n[rst2man-indent-level]]
29 .in \\n[rst2man-indent\\n[rst2man-indent-level]]u
31 .\" Man page generated from reStructeredText.
34 memaslap \- Load testing and benchmarking tool for memcached
44 \fBmemaslap\fP is a load generation and benchmark tool for memcached(1)
45 servers. It generates configurable workload such as threads, concurrencies, connections,
46 run time, overwrite, miss rate, key size, value size, get/set proportion,
47 expected throughput, and so on. Furthermore, it also testss data
48 verification, expire\-time verification, UDP, binary protocol, facebook test,
49 replication test, multi\-get and reconnection, etc.
51 Memslap manages network connections like memcached with
52 libevent. Each thread of memaslap is bound with a CPU core, all
53 the threads don\(aqt communicate with each other, and there are several socket
54 connections in each thread. Each connection keeps key size distribution,
55 value size distribution, and command distribution by itself.
57 You can specify servers via the \fB\-\-servers\fP option or via the
58 environment variable \fBMEMCACHED_SERVERS\fP.
61 Memslap is developed to for the following purposes:
63 Manages network connections with libevent asynchronously.
65 Set both TCP and UDP up to use non\-blocking IO.
67 Improves parallelism: higher performance in multi\-threads environments.
69 Improves time efficiency: faster processing speed.
71 Generates key and value more efficiently; key size distribution and value size distribution are configurable.
73 Supports get, multi\-get, and set commands; command distribution is configurable.
75 Supports controllable miss rate and overwrite rate.
77 Supports data and expire\-time verification.
79 Supports dumping statistic information periodically.
81 Supports thousands of TCP connections.
83 Supports binary protocol.
85 Supports facebook test (set with TCP and multi\-get with UDP) and replication test.
87 .SS Effective implementation of network.
89 For memaslap, both TCP and UDP use non\-blocking network IO. All
90 the network events are managed by libevent as memcached. The network module
91 of memaslap is similar to memcached. Libevent can ensure
92 memaslap can handle network very efficiently.
93 .SS Effective implementation of multi\-threads and concurrency
95 Memslap has the similar implementation of multi\-threads to
96 memcached. Memslap creates one or more self\-governed threads;
97 each thread is bound with one CPU core if the system testss setting CPU
100 In addition, each thread has a libevent to manage the events of the network;
101 each thread has one or more self\-governed concurrencies; and each
102 concurrency has one or more socket connections. All the concurrencies don’t
103 communicate with each other even though they are in the same thread.
105 Memslap can create thousands of socket connections, and each
106 concurrency has tens of socket connections. Each concurrency randomly or
107 sequentially selects one socket connection from its socket connection pool
108 to run, so memaslap can ensure each concurrency handles one
109 socket connection at any given time. Users can specify the number of
110 concurrency and socket connections of each concurrency according to their
112 .SS Effective implementation of generating key and value
114 In order to improve time efficiency and space efficiency,
115 memaslap creates a random characters table with 10M characters. All the
116 suffixes of keys and values are generated from this random characters table.
118 Memslap uses the offset in the character table and the length
119 of the string to identify a string. It can save much memory.
120 Each key contains two parts, a prefix and a suffix. The prefix is an
121 uint64_t, 8 bytes. In order to verify the data set before,
122 memaslap need to ensure each key is unique, so it uses the prefix to identify
123 a key. The prefix cannot include illegal characters, such as ‘r’, ‘n’,
124 ‘0’ and ‘ ‘. And memaslap has an algorithm to ensure that.
126 Memslap doesn’t generate all the objects (key\-value pairs) at
127 the beginning. It only generates enough objects to fill the task window
128 (default 10K objects) of each concurrency. Each object has the following
129 basic information, key prefix, key suffix offset in the character table, key
130 length, value offset in the character table, and value length.
132 In the work process, each concurrency sequentially or randomly selects an
133 object from the window to do set operation or get operation. At the same
134 time, each concurrency kicks objects out of its window and adds new object
136 .SS Simple but useful task scheduling
138 Memslap uses libevent to schedule all the concurrencies of
139 threads, and each concurrency schedules tasks based on the local task
140 window. Memslap assumes that if each concurrency keeps the same
141 key distribution, value distribution and commands distribution, from
142 outside, memaslap keeps all the distribution as a whole.
143 Each task window includes a lot of objects, each object stores its basic
144 information, such as key, value, expire time, and so on. At any time, all
145 the objects in the window keep the same and fixed key and value
146 distribution. If an object is overwritten, the value of the object will be
147 updated. Memslap verifies the data or expire\-time according to
148 the object information stored in the task window.
150 Libevent selects which concurrency to handle based on a specific network
151 event. Then the concurrency selects which command (get or set) to operate
152 based on the command distribution. If it needs to kick out an old object and
153 add a new object, in order to keep the same key and value distribution, the
154 new object must have the same key length and value length.
156 If memcached server has two cache layers (memory and SSD), running
157 memaslap with different window sizes can get different cache
158 miss rates. If memaslap adds enough objects into the windows at
159 the beginning, and the cache of memcached cannot store all the objects
160 initialized, then memaslap will get some objects from the second
161 cache layer. It causes the first cache layer to miss. So the user can
162 specify the window size to get the expected miss rate of the first cache
164 .SS Useful implementation of multi\-servers , UDP, TCP, multi\-get and binary protocol
166 Because each thread is self\-governed, memaslap can assign
167 different threads to handle different memcached servers. This is just one of
168 the ways in which memaslap testss multiple servers. The only
169 limitation is that the number of servers cannot be greater than the number
170 of threads. The other way to tests multiple servers is for replication
171 test. Each concurrency has one socket connection to each memcached server.
172 For the implementation, memaslap can set some objects to one
173 memcached server, and get these objects from the other servers.
175 By default, Memslap does single get. If the user specifies
176 multi\-get option, memaslap will collect enough get commands and
177 pack and send the commands together.
179 Memslap testss both the ASCII protocol and binary protocol,
180 but it runs on the ASCII protocol by default.
181 Memslap by default runs on the TCP protocol, but it also
182 testss UDP. Because UDP is unreliable, dropped packages and out\-of\-order
183 packages may occur. Memslap creates a memory buffer to handle
184 these problems. Memslap tries to read all the response data of
185 one command from the server and reorders the response data. If some packages
186 get lost, the waiting timeout mechanism can ensure half\-baked packages will
187 be discarded and the next command will be sent.
190 Below are some usage samples:
192 memaslap \-s 127.0.0.1:11211 \-S 5s
194 memaslap \-s 127.0.0.1:11211 \-t 2m \-v 0.2 \-e 0.05 \-b
196 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-w 40k \-S 20s \-o 0.2
198 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-T 4 \-c 128 \-d 20 \-P 40k
200 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-d 50 \-a \-n 40
202 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m
204 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m \-p 2
206 The user must specify one server at least to run memaslap. The
207 rest of the parameters have default values, as shown below:
209 Thread number = 1 Concurrency = 16
211 Run time = 600 seconds Configuration file = NULL
213 Key size = 64 Value size = 1024
215 Get/set = 9:1 Window size = 10k
217 Execute number = 0 Single get = true
219 Multi\-get = false Number of sockets of each concurrency = 1
221 Reconnect = false Data verification = false
223 Expire\-time verification = false ASCII protocol = true
225 Binary protocol = false Dumping statistic information
229 Overwrite proportion = 0% UDP = false
231 TCP = true Limit throughput = false
233 Facebook test = false Replication test = false
234 .SS Key size, value size and command distribution.
236 All the distributions are read from the configuration file specified by user
237 with “—cfg_cmd” option. If the user does not specify a configuration file,
238 memaslap will run with the default distribution (key size = 64,
239 value size = 1024, get/set = 9:1). For information on how to edit the
240 configuration file, refer to the “Configuration File” section.
242 The minimum key size is 16 bytes; the maximum key size is 250 bytes. The
243 precision of proportion is 0.001. The proportion of distribution will be
244 rounded to 3 decimal places.
246 The minimum value size is 1 bytes; the maximum value size is 1M bytes. The
247 precision of proportion is 0.001. The proportion of distribution will be
248 rounded to 3 decimal places.
249 Currently, memaslap only testss set and get commands. And it
250 testss 100% set and 100% get. For 100% get, it will preset some objects to
252 .SS Multi\-thread and concurrency
254 The high performance of memaslap benefits from the special
255 schedule of thread and concurrency. It’s important to specify the proper
256 number of them. The default number of threads is 1; the default number of
257 concurrency is 16. The user can use “—threads” and “\-\-concurrency” to
258 specify these variables.
260 If the system testss setting CPU affinity and the number of threads
261 specified by the user is greater than 1, memaslap will try to
262 bind each thread to a different CPU core. So if you want to get the best
263 performance memaslap, it is better to specify the number of
264 thread equal to the number of CPU cores. The number of threads specified by
265 the user can also be less or greater than the number of CPU cores. Because
266 of the limitation of implementation, the number of concurrencies could be
267 the multiple of the number of threads.
271 For 8 CPU cores system
276 \-\-threads=2 \-\-concurrency=128
278 \-\-threads=8 \-\-concurrency=128
280 \-\-threads=8 \-\-concurrency=256
282 \-\-threads=12 \-\-concurrency=144
286 For 16 CPU cores system
291 \-\-threads=8 \-\-concurrency=128
293 \-\-threads=16 \-\-concurrency=256
295 \-\-threads=16 \-\-concurrency=512
297 \-\-threads=24 \-\-concurrency=288
299 The memaslap performs very well, when
300 used to test the performance of memcached servers.
301 Most of the time, the bottleneck is the network or
302 the server. If for some reason the user wants to
303 limit the performance of memaslap, there
304 are two ways to do this:
306 Decrease the number of threads and concurrencies.
307 Use the option “\-\-tps” that memaslap
308 provides to limit the throughput. This option allows
309 the user to get the expected throughput. For
310 example, assume that the maximum throughput is 50
311 kops/s for a specific configuration, you can specify
312 the throughput equal to or less than the maximum
313 throughput using “\-\-tps” option.
316 Most of the time, the user does not need to specify the window size. The
317 default window size is 10k. For Schooner Memcached, the user can specify
318 different window sizes to get different cache miss rates based on the test
319 case. Memslap testss cache miss rate between 0% and 100%.
320 If you use this utility to test the performance of Schooner Memcached, you
321 can specify a proper window size to get the expected cache miss rate. The
322 formula for calculating window size is as follows:
324 Assume that the key size is 128 bytes, and the value size is 2048 bytes, and
327 1. Small cache cache_size=1M, 100% cache miss (all data get from SSD).
335 (1). cache miss rate 0%
339 (2). cache miss rate 5%
348 (1). cache miss rate 0%
358 The formula for calculating window size for cache miss rate 0%:
360 cache_size / concurrency / (key_size + value_size) * 0.5
362 The formula for calculating window size for cache miss rate 5%:
364 cache_size / concurrency / (key_size + value_size) * 0.7
367 Memslap testss both data verification and expire\-time
368 verification. The user can use "\-\-verify=" or "\-v" to specify the proportion
369 of data verification. In theory, it testss 100% data verification. The
370 user can use "\-\-exp_verify=" or "\-e" to specify the proportion of
371 expire\-time verification. In theory, it testss 100% expire\-time
372 verification. Specify the "\-\-verbose" options to get more detailed error
375 For example: \-\-exp_verify=0.01 –verify=0.1 , it means that 1% of the objects
376 set with expire\-time, 10% of the objects gotten will be verified. If the
377 objects are gotten, memaslap will verify the expire\-time and
379 .SS multi\-servers and multi\-config
381 Memslap testss multi\-servers based on self\-governed thread.
382 There is a limitation that the number of servers cannot be greater than the
383 number of threads. Memslap assigns one thread to handle one
384 server at least. The user can use the "\-\-servers=" or "\-s" option to specify
389 \-\-servers=10.1.1.1:11211,10.1.1.2:11212,10.1.1.3:11213 \-\-threads=6 \-\-concurrency=36
391 The above command means that there are 6 threads, with each thread having 6
392 concurrencies and that threads 0 and 3 handle server 0 (10.1.1.1); threads 1
393 and 4 handle server 1 (10.1.1.2); and thread 2 and 5 handle server 2
396 All the threads and concurrencies in memaslap are self\-governed.
398 So is memaslap. The user can start up several
399 memaslap instances. The user can run memaslap on different client
400 machines to communicate with the same memcached server at the same. It is
401 recommended that the user start different memaslap on different
402 machines using the same configuration.
403 .SS Run with execute number mode or time mode
405 The default memaslap runs with time mode. The default run time
406 is 10 minutes. If it times out, memaslap will exit. Do not
407 specify both execute number mode and time mode at the same time; just
412 \-\-time=30s (It means the test will run 30 seconds.)
414 \-\-execute_number=100000 (It means that after running 100000 commands, the test will exit.)
415 .SS Dump statistic information periodically.
417 The user can use "\-\-stat_freq=" or "\-S" to specify the frequency.
423 Memslap will dump the statistics of the commands (get and set) at the frequency of every 20
426 For more information on the format of dumping statistic information, refer to “Format of Output” section.
429 The user can use "\-\-division=" or "\-d" to specify multi\-get keys count.
430 Memslap by default does single get with TCP. Memslap also testss data
431 verification and expire\-time verification for multi\-get.
433 Memslap testss multi\-get with both TCP and UDP. Because of
434 the different implementation of the ASCII protocol and binary protocol,
435 there are some differences between the two. For the ASCII protocol,
436 memaslap sends one “multi\-get” to the server once. For the
437 binary protocol, memaslap sends several single get commands
438 together as “multi\-get” to the server.
441 Memslap testss both UDP and TCP. For TCP,
442 memaslap does not reconnect the memcached server if socket connections are
443 lost. If all the socket connections are lost or memcached server crashes,
444 memaslap will exit. If the user specifies the “\-\-reconnect”
445 option when socket connections are lost, it will reconnect them.
447 User can use “\-\-udp” to enable the UDP feature, but UDP comes with some
450 UDP cannot set data more than 1400 bytes.
452 UDP is not testsed by the binary protocol because the binary protocol of
453 memcached does not tests that.
455 UDP doesn’t tests reconnection.
458 Set data with TCP and multi\-get with UDP. Specify the following options:
460 "\-\-facebook \-\-division=50"
462 If you want to create thousands of TCP connections, specify the
464 "\-\-conn_sock=" option.
466 For example: \-\-facebook \-\-division=50 \-\-conn_sock=200
468 The above command means that memaslap will do facebook test,
469 each concurrency has 200 socket TCP connections and one UDP socket.
471 Memslap sets objects with the TCP socket, and multi\-gets 50
472 objects once with the UDP socket.
474 If you specify "\-\-division=50", the key size must be less that 25 bytes
475 because the UDP packet size is 1400 bytes.
478 For replication test, the user must specify at least two memcached servers.
479 The user can use “—rep_write=” option to enable feature.
483 \-\-servers=10.1.1.1:11211,10.1.1.2:11212 –rep_write=2
485 The above command means that there are 2 replication memcached servers,
486 memaslap will set objects to both server 0 and server 1, get
487 objects which are set to server 0 before from server 1, and also get objects
488 which are set to server 1 before from server 0. If server 0 crashes,
489 memaslap will only get objects from server 1. If server 0 comes
490 back to life again, memaslap will reconnect server 0. If both
491 server 0 and server 1 crash, memaslap will exit.
492 .SS Supports thousands of TCP connections
494 Start memaslap with "\-\-conn_sock=" or "\-n" to enable this
495 feature. Make sure that your system can tests opening thousands of files
496 and creating thousands of sockets. However, this feature does not tests
497 reconnection if sockets disconnect.
501 \-\-threads=8 \-\-concurrency=128 \-\-conn_sock=128
503 The above command means that memaslap starts up 8 threads, each
504 thread has 16 concurrencies, each concurrency has 128 TCP socket
505 connections, and the total number of TCP socket connections is 128 * 128 =
507 .SS Supports binary protocol
509 Start memaslap with "\-\-binary" or "\-B" options to enable this
510 feature. It testss all the above features except UDP, because the latest
511 memcached 1.3.3 does not implement binary UDP protocol.
517 Since memcached 1.3.3 doesn\(aqt implement binary UDP protocol,
518 memaslap does not tests UDP. In addition, memcached 1.3.3 does not tests
519 multi\-get. If you specify "\-\-division=50" option, it just sends 50 get
520 commands together as “mulit\-get” to the server.
521 .SH CONFIGURATION FILE
523 This section describes the format of the configuration file. By default
524 when no configuration file is specified memaslap reads the default
525 one located at ~/.memaslap.cnf.
527 Below is a sample configuration file:
531 ***************************************************************************
532 #comments should start with \(aq#\(aq
534 #start_len end_len proportion
536 #key length range from start_len to end_len
537 #start_len must be equal to or greater than 16
538 #end_len must be equal to or less than 250
539 #start_len must be equal to or greater than end_len
540 #memaslap will generate keys according to the key range
541 #proportion: indicates keys generated from one range accounts for the total
544 #example1: key range 16~100 accounts for 80%
545 # key range 101~200 accounts for 10%
546 # key range 201~250 accounts for 10%
547 # total should be 1 (0.8+0.1+0.1 = 1)
553 #example2: all keys length are 128 bytes
559 #start_len end_len proportion
561 #value length range from start_len to end_len
562 #start_len must be equal to or greater than 1
563 #end_len must be equal to or less than 1M
564 #start_len must be equal to or greater than end_len
565 #memaslap will generate values according to the value range
566 #proportion: indicates values generated from one range accounts for the
567 total generated values
569 #example1: value range 1~1000 accounts for 80%
570 # value range 1001~10000 accounts for 10%
571 # value range 10001~100000 accounts for 10%
572 # total should be 1 (0.8+0.1+0.1 = 1)
578 #example2: all value length are 128 bytes
584 #cmd_type cmd_proportion
586 #currently memaslap only testss get and set command.
592 #example: set command accounts for 50%
593 # get command accounts for 50%
594 # total should be 1 (0.5+0.5 = 1)
606 At the beginning, memaslap displays some configuration information as follows:
608 servers : 127.0.0.1:11211
618 set proportion: set_prop=0.10
620 get proportion: get_prop=0.90
627 The servers used by memaslap.
635 The number of threads memaslap runs with.
643 The number of concurrencies memaslap runs with.
651 How long to run memaslap.
659 The task window size of each concurrency.
667 The proportion of set command.
675 The proportion of get command.
679 The output of dynamic statistics is something like this:
683 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
685 Type Time(s) Ops TPS(ops/s) Net(M/s) Get_miss Min(us) Max(us)
686 Avg(us) Std_dev Geo_dist
687 Period 5 345826 69165 65.3 0 27 2198 203
689 Global 20 1257935 62896 71.8 0 26 3791 224
694 Type Time(s) Ops TPS(ops/s) Net(M/s) Get_miss Min(us) Max(us)
695 Avg(us) Std_dev Geo_dist
696 Period 5 38425 7685 7.3 0 42 628 240
698 Global 20 139780 6989 8.0 0 37 3790 253
703 Type Time(s) Ops TPS(ops/s) Net(M/s) Get_miss Min(us) Max(us)
704 Avg(us) Std_dev Geo_dist
705 Period 5 384252 76850 72.5 0 27 2198 207
707 Global 20 1397720 69886 79.7 0 26 3791 227
709 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
718 Statistics information of get command
726 Statistics information of set command
734 Statistics information of both get and set command
742 Result within a period
766 Throughput, operations/second
782 How many objects can’t be gotten
790 The minimum response time
798 The maximum response time
806 The average response time
814 Standard deviation of response time
822 Geometric distribution based on natural exponential function
826 At the end, memaslap will output something like this:
830 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
831 Get Statistics (1257956 events)
839 8: 484890 459823 12543 824
842 Set Statistics (139782 events)
850 8: 50784 65574 2064 167
853 Total Statistics (1397738 events)
861 8: 535674 525397 14607 991
871 written_bytes: 242516030
872 read_bytes: 1003702556
873 object_bytes: 152086080
878 Run time: 20.0s Ops: 1397754 TPS: 69817 Net_rate: 59.4M/s
879 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
888 Get statistics of response time
896 Set statistics of response time
904 Both get and set statistics of response time
912 The accumulated and minimum response time
920 The accumulated and maximum response time
928 The accumulated and average response time
936 Standard deviation of response time
944 Geometric distribution based on logarithm 2
952 Total get commands done
960 Total set commands done
968 How many objects can’t be gotten from server
976 How many objects need to verify but can’t get them
984 How many objects with insistent value
992 How many objects are expired but we get them
1000 How many objects are unexpired but we can’t get them
1032 How many UDP packages are disorder
1040 How many UDP packages are lost
1048 How many times UDP time out happen
1072 Throughput, operations/second
1080 The average rate of network
1086 .B \-s, \-\-servers=
1088 List one or more servers to connect. Servers count must be less than
1089 threads count. e.g.: \-\-servers=localhost:1234,localhost:11211
1091 .B \-T, \-\-threads=
1093 Number of threads to startup, better equal to CPU numbers. Default 8.
1095 .B \-c, \-\-concurrency=
1097 Number of concurrency to simulate with load. Default 128.
1099 .B \-n, \-\-conn_sock=
1101 Number of TCP socks per concurrency. Default 1.
1103 .B \-x, \-\-execute_number=
1105 Number of operations(get and set) to execute for the
1106 given test. Default 1000000.
1110 How long the test to run, suffix: s\-seconds, m\-minutes, h\-hours,
1111 d\-days e.g.: \-\-time=2h.
1113 .B \-F, \-\-cfg_cmd=
1115 Load the configure file to get command,key and value distribution list.
1117 .B \-w, \-\-win_size=
1119 Task window size of each concurrency, suffix: K, M e.g.: \-\-win_size=10k.
1122 .B \-X, \-\-fixed_size=
1124 Fixed length of value.
1128 The proportion of date verification, e.g.: \-\-verify=0.01
1130 .B \-d, \-\-division=
1132 Number of keys to multi\-get once. Default 1, means single get.
1134 .B \-S, \-\-stat_freq=
1136 Frequency of dumping statistic information. suffix: s\-seconds,
1137 m\-minutes, e.g.: \-\-resp_freq=10s.
1139 .B \-e, \-\-exp_verify=
1141 The proportion of objects with expire time, e.g.: \-\-exp_verify=0.01.
1142 Default no object with expire time
1144 .B \-o, \-\-overwrite=
1146 The proportion of objects need overwrite, e.g.: \-\-overwrite=0.01.
1147 Default never overwrite object.
1151 .B \-R, \-\-reconnect
1153 Reconnect tests, when connection is closed it will be reconnected.
1157 UDP tests, default memaslap uses TCP, TCP port and UDP port of
1158 server must be same.
1160 .B \-a, \-\-facebook
1162 Whether it enables facebook test feature, set with TCP and multi\-get with UDP.
1166 Whether it enables binary protocol. Default with ASCII protocol.
1172 Expected throughput, suffix: K, e.g.: \-\-tps=10k.
1174 .B \-p, \-\-rep_write=
1176 The first nth servers can write data, e.g.: \-\-rep_write=2.
1182 Whether it outputs detailed information when verification fails.
1186 Display this message and then exit.
1190 Display the version of the application and then exit.
1194 memaslap \-s 127.0.0.1:11211 \-S 5s
1196 memaslap \-s 127.0.0.1:11211 \-t 2m \-v 0.2 \-e 0.05 \-b
1198 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-w 40k \-S 20s \-o 0.2
1200 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-T 4 \-c 128 \-d 20 \-P 40k
1202 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-d 50 \-a \-n 40
1204 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m
1206 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m \-p 2
1209 To find out more information please check:
1210 \fI\%http://launchpad.org/libmemcached\fP
1213 Mingqiang Zhuang <\fI\%mingqiangzhuang@hengtiansoft.com\fP> (Schooner Technolgy)
1214 Brian Aker, <\fI\%brian@tangent.org\fP>
1217 \fImemcached(1)\fP \fIlibmemcached(3)\fP \fImemcached_strerror(3)\fP
1222 .\" Generated by docutils manpage writer.