jenkins-promote-staging-trunk-libmemcached-16
[awesomized/libmemcached] / docs / man / memaslap.1
1 .TH "MEMASLAP" "1" "May 03, 2012" "1.0.7" "libmemcached"
2 .SH NAME
3 memaslap \- libmemcached Documentation
4 .
5 .nr rst2man-indent-level 0
6 .
7 .de1 rstReportMargin
8 \\$1 \\n[an-margin]
9 level \\n[rst2man-indent-level]
10 level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
11 -
12 \\n[rst2man-indent0]
13 \\n[rst2man-indent1]
14 \\n[rst2man-indent2]
15 ..
16 .de1 INDENT
17 .\" .rstReportMargin pre:
18 . RS \\$1
19 . nr rst2man-indent\\n[rst2man-indent-level] \\n[an-margin]
20 . nr rst2man-indent-level +1
21 .\" .rstReportMargin post:
22 ..
23 .de UNINDENT
24 . RE
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
30 ..
31 .\" Man page generated from reStructeredText.
32 .
33 .SH SYNOPSIS
34 .sp
35 memaslap [options]
36 .INDENT 0.0
37 .TP
38 .B \-\-help
39 .UNINDENT
40 .INDENT 0.0
41 .TP
42 .B MEMCACHED_SERVERS
43 .UNINDENT
44 .SH DESCRIPTION
45 .sp
46 \fBmemaslap\fP is a load generation and benchmark tool for memcached
47 servers. It generates configurable workload such as threads, concurrencies,
48 connections, run time, overwrite, miss rate, key size, value size, get/set
49 proportion, expected throughput, and so on. Furthermore, it also testss data
50 verification, expire\-time verification, UDP, binary protocol, facebook test,
51 replication test, multi\-get and reconnection, etc.
52 .sp
53 Memaslap manages network connections like memcached with
54 libevent. Each thread of memaslap is bound with a CPU core, all
55 the threads don\(aqt communicate with each other, and there are several socket
56 connections in each thread. Each connection keeps key size distribution,
57 value size distribution, and command distribution by itself.
58 .sp
59 You can specify servers via the \fImemslap \-\-servers\fP option or via the
60 environment variable \fI\%MEMCACHED_SERVERS\fP.
61 .SH FEATURES
62 .sp
63 Memslap is developed to for the following purposes:
64 .sp
65 Manages network connections with libevent asynchronously.
66 .sp
67 Set both TCP and UDP up to use non\-blocking IO.
68 .sp
69 Improves parallelism: higher performance in multi\-threads environments.
70 .sp
71 Improves time efficiency: faster processing speed.
72 .sp
73 Generates key and value more efficiently; key size distribution and value size distribution are configurable.
74 .sp
75 Supports get, multi\-get, and set commands; command distribution is configurable.
76 .sp
77 Supports controllable miss rate and overwrite rate.
78 .sp
79 Supports data and expire\-time verification.
80 .sp
81 Supports dumping statistic information periodically.
82 .sp
83 Supports thousands of TCP connections.
84 .sp
85 Supports binary protocol.
86 .sp
87 Supports facebook test (set with TCP and multi\-get with UDP) and replication test.
88 .SH DETAILS
89 .SS Effective implementation of network.
90 .sp
91 For memaslap, both TCP and UDP use non\-blocking network IO. All
92 the network events are managed by libevent as memcached. The network module
93 of memaslap is similar to memcached. Libevent can ensure
94 memaslap can handle network very efficiently.
95 .SS Effective implementation of multi\-threads and concurrency
96 .sp
97 Memslap has the similar implementation of multi\-threads to
98 memcached. Memslap creates one or more self\-governed threads;
99 each thread is bound with one CPU core if the system testss setting CPU
100 core affinity.
101 .sp
102 In addition, each thread has a libevent to manage the events of the network;
103 each thread has one or more self\-governed concurrencies; and each
104 concurrency has one or more socket connections. All the concurrencies don’t
105 communicate with each other even though they are in the same thread.
106 .sp
107 Memslap can create thousands of socket connections, and each
108 concurrency has tens of socket connections. Each concurrency randomly or
109 sequentially selects one socket connection from its socket connection pool
110 to run, so memaslap can ensure each concurrency handles one
111 socket connection at any given time. Users can specify the number of
112 concurrency and socket connections of each concurrency according to their
113 expected workload.
114 .SS Effective implementation of generating key and value
115 .sp
116 In order to improve time efficiency and space efficiency,
117 memaslap creates a random characters table with 10M characters. All the
118 suffixes of keys and values are generated from this random characters table.
119 .sp
120 Memslap uses the offset in the character table and the length
121 of the string to identify a string. It can save much memory.
122 Each key contains two parts, a prefix and a suffix. The prefix is an
123 uint64_t, 8 bytes. In order to verify the data set before,
124 memaslap need to ensure each key is unique, so it uses the prefix to identify
125 a key. The prefix cannot include illegal characters, such as ‘r’, ‘n’,
126 ‘0’ and ‘ ‘. And memaslap has an algorithm to ensure that.
127 .sp
128 Memslap doesn’t generate all the objects (key\-value pairs) at
129 the beginning. It only generates enough objects to fill the task window
130 (default 10K objects) of each concurrency. Each object has the following
131 basic information, key prefix, key suffix offset in the character table, key
132 length, value offset in the character table, and value length.
133 .sp
134 In the work process, each concurrency sequentially or randomly selects an
135 object from the window to do set operation or get operation. At the same
136 time, each concurrency kicks objects out of its window and adds new object
137 into it.
138 .SS Simple but useful task scheduling
139 .sp
140 Memslap uses libevent to schedule all the concurrencies of
141 threads, and each concurrency schedules tasks based on the local task
142 window. Memslap assumes that if each concurrency keeps the same
143 key distribution, value distribution and commands distribution, from
144 outside, memaslap keeps all the distribution as a whole.
145 Each task window includes a lot of objects, each object stores its basic
146 information, such as key, value, expire time, and so on. At any time, all
147 the objects in the window keep the same and fixed key and value
148 distribution. If an object is overwritten, the value of the object will be
149 updated. Memslap verifies the data or expire\-time according to
150 the object information stored in the task window.
151 .sp
152 Libevent selects which concurrency to handle based on a specific network
153 event. Then the concurrency selects which command (get or set) to operate
154 based on the command distribution. If it needs to kick out an old object and
155 add a new object, in order to keep the same key and value distribution, the
156 new object must have the same key length and value length.
157 .sp
158 If memcached server has two cache layers (memory and SSD), running
159 memaslap with different window sizes can get different cache
160 miss rates. If memaslap adds enough objects into the windows at
161 the beginning, and the cache of memcached cannot store all the objects
162 initialized, then memaslap will get some objects from the second
163 cache layer. It causes the first cache layer to miss. So the user can
164 specify the window size to get the expected miss rate of the first cache
165 layer.
166 .SS Useful implementation of multi\-servers , UDP, TCP, multi\-get and binary protocol
167 .sp
168 Because each thread is self\-governed, memaslap can assign
169 different threads to handle different memcached servers. This is just one of
170 the ways in which memaslap tests multiple servers. The only
171 limitation is that the number of servers cannot be greater than the number
172 of threads. The other way to test multiple servers is for replication
173 test. Each concurrency has one socket connection to each memcached server.
174 For the implementation, memaslap can set some objects to one
175 memcached server, and get these objects from the other servers.
176 .sp
177 By default, Memslap does single get. If the user specifies
178 multi\-get option, memaslap will collect enough get commands and
179 pack and send the commands together.
180 .sp
181 Memslap testss both the ASCII protocol and binary protocol,
182 but it runs on the ASCII protocol by default.
183 Memslap by default runs on the TCP protocol, but it also
184 tests UDP. Because UDP is unreliable, dropped packages and out\-of\-order
185 packages may occur. Memslap creates a memory buffer to handle
186 these problems. Memslap tries to read all the response data of
187 one command from the server and reorders the response data. If some packages
188 get lost, the waiting timeout mechanism can ensure half\-baked packages will
189 be discarded and the next command will be sent.
190 .SH USAGE
191 .sp
192 Below are some usage samples:
193 .sp
194 memaslap \-s 127.0.0.1:11211 \-S 5s
195 .sp
196 memaslap \-s 127.0.0.1:11211 \-t 2m \-v 0.2 \-e 0.05 \-b
197 .sp
198 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-w 40k \-S 20s \-o 0.2
199 .sp
200 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-T 4 \-c 128 \-d 20 \-P 40k
201 .sp
202 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-d 50 \-a \-n 40
203 .sp
204 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m
205 .sp
206 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m \-p 2
207 .sp
208 The user must specify one server at least to run memaslap. The
209 rest of the parameters have default values, as shown below:
210 .sp
211 Thread number = 1 Concurrency = 16
212 .sp
213 Run time = 600 seconds Configuration file = NULL
214 .sp
215 Key size = 64 Value size = 1024
216 .sp
217 Get/set = 9:1 Window size = 10k
218 .sp
219 Execute number = 0 Single get = true
220 .sp
221 Multi\-get = false Number of sockets of each concurrency = 1
222 .sp
223 Reconnect = false Data verification = false
224 .sp
225 Expire\-time verification = false ASCII protocol = true
226 .sp
227 Binary protocol = false Dumping statistic information
228 .sp
229 periodically = false
230 .sp
231 Overwrite proportion = 0% UDP = false
232 .sp
233 TCP = true Limit throughput = false
234 .sp
235 Facebook test = false Replication test = false
236 .SS Key size, value size and command distribution.
237 .sp
238 All the distributions are read from the configuration file specified by user
239 with “—cfg_cmd” option. If the user does not specify a configuration file,
240 memaslap will run with the default distribution (key size = 64,
241 value size = 1024, get/set = 9:1). For information on how to edit the
242 configuration file, refer to the “Configuration File” section.
243 .sp
244 The minimum key size is 16 bytes; the maximum key size is 250 bytes. The
245 precision of proportion is 0.001. The proportion of distribution will be
246 rounded to 3 decimal places.
247 .sp
248 The minimum value size is 1 bytes; the maximum value size is 1M bytes. The
249 precision of proportion is 0.001. The proportion of distribution will be
250 rounded to 3 decimal places.
251 Currently, memaslap only testss set and get commands. And it
252 testss 100% set and 100% get. For 100% get, it will preset some objects to
253 the server.
254 .SS Multi\-thread and concurrency
255 .sp
256 The high performance of memaslap benefits from the special
257 schedule of thread and concurrency. It’s important to specify the proper
258 number of them. The default number of threads is 1; the default number of
259 concurrency is 16. The user can use “—threads” and “\-\-concurrency” to
260 specify these variables.
261 .sp
262 If the system tests setting CPU affinity and the number of threads
263 specified by the user is greater than 1, memaslap will try to
264 bind each thread to a different CPU core. So if you want to get the best
265 performance memaslap, it is better to specify the number of
266 thread equal to the number of CPU cores. The number of threads specified by
267 the user can also be less or greater than the number of CPU cores. Because
268 of the limitation of implementation, the number of concurrencies could be
269 the multiple of the number of threads.
270 .INDENT 0.0
271 .IP 1. 3
272 For 8 CPU cores system
273 .UNINDENT
274 .sp
275 For example:
276 .sp
277 \-\-threads=2 \-\-concurrency=128
278 .sp
279 \-\-threads=8 \-\-concurrency=128
280 .sp
281 \-\-threads=8 \-\-concurrency=256
282 .sp
283 \-\-threads=12 \-\-concurrency=144
284 .INDENT 0.0
285 .IP 2. 3
286 For 16 CPU cores system
287 .UNINDENT
288 .sp
289 For example:
290 .sp
291 \-\-threads=8 \-\-concurrency=128
292 .sp
293 \-\-threads=16 \-\-concurrency=256
294 .sp
295 \-\-threads=16 \-\-concurrency=512
296 .sp
297 \-\-threads=24 \-\-concurrency=288
298 .sp
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:
305 .sp
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.
314 .SS Window size
315 .sp
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:
323 .sp
324 Assume that the key size is 128 bytes, and the value size is 2048 bytes, and
325 concurrency=128.
326 .sp
327 1. Small cache cache_size=1M, 100% cache miss (all data get from SSD).
328 win_size=10k
329 .INDENT 0.0
330 .IP 2. 3
331 cache_size=4G
332 .UNINDENT
333 .sp
334 (1). cache miss rate 0%
335 .sp
336 win_size=8k
337 .sp
338 (2). cache miss rate 5%
339 .sp
340 win_size=11k
341 .INDENT 0.0
342 .IP 3. 3
343 cache_size=16G
344 .UNINDENT
345 .sp
346 (1). cache miss rate 0%
347 .sp
348 win_size=32k
349 .sp
350 (2). cache miss
351 .sp
352 rate 5%
353 .sp
354 win_size=46k
355 .sp
356 The formula for calculating window size for cache miss rate 0%:
357 .sp
358 cache_size / concurrency / (key_size + value_size) * 0.5
359 .sp
360 The formula for calculating window size for cache miss rate 5%:
361 .sp
362 cache_size / concurrency / (key_size + value_size) * 0.7
363 .SS Verification
364 .sp
365 Memslap testss both data verification and expire\-time
366 verification. The user can use "\-\-verify=" or "\-v" to specify the proportion
367 of data verification. In theory, it testss 100% data verification. The
368 user can use "\-\-exp_verify=" or "\-e" to specify the proportion of
369 expire\-time verification. In theory, it testss 100% expire\-time
370 verification. Specify the "\-\-verbose" options to get more detailed error
371 information.
372 .sp
373 For example: \-\-exp_verify=0.01 –verify=0.1 , it means that 1% of the objects
374 set with expire\-time, 10% of the objects gotten will be verified. If the
375 objects are gotten, memaslap will verify the expire\-time and
376 value.
377 .SS multi\-servers and multi\-config
378 .sp
379 Memslap testss multi\-servers based on self\-governed thread.
380 There is a limitation that the number of servers cannot be greater than the
381 number of threads. Memslap assigns one thread to handle one
382 server at least. The user can use the "\-\-servers=" or "\-s" option to specify
383 multi\-servers.
384 .sp
385 For example:
386 .sp
387 \-\-servers=10.1.1.1:11211,10.1.1.2:11212,10.1.1.3:11213 \-\-threads=6 \-\-concurrency=36
388 .sp
389 The above command means that there are 6 threads, with each thread having 6
390 concurrencies and that threads 0 and 3 handle server 0 (10.1.1.1); threads 1
391 and 4 handle server 1 (10.1.1.2); and thread 2 and 5 handle server 2
392 (10.1.1.3).
393 .sp
394 All the threads and concurrencies in memaslap are self\-governed.
395 .sp
396 So is memaslap. The user can start up several
397 memaslap instances. The user can run memaslap on different client
398 machines to communicate with the same memcached server at the same. It is
399 recommended that the user start different memaslap on different
400 machines using the same configuration.
401 .SS Run with execute number mode or time mode
402 .sp
403 The default memaslap runs with time mode. The default run time
404 is 10 minutes. If it times out, memaslap will exit. Do not
405 specify both execute number mode and time mode at the same time; just
406 specify one instead.
407 .sp
408 For example:
409 .sp
410 \-\-time=30s (It means the test will run 30 seconds.)
411 .sp
412 \-\-execute_number=100000 (It means that after running 100000 commands, the test will exit.)
413 .SS Dump statistic information periodically.
414 .sp
415 The user can use "\-\-stat_freq=" or "\-S" to specify the frequency.
416 .sp
417 For example:
418 .sp
419 \-\-stat_freq=20s
420 .sp
421 Memslap will dump the statistics of the commands (get and set) at the frequency of every 20
422 seconds.
423 .sp
424 For more information on the format of dumping statistic information, refer to “Format of Output” section.
425 .SS Multi\-get
426 .sp
427 The user can use "\-\-division=" or "\-d" to specify multi\-get keys count.
428 Memslap by default does single get with TCP. Memslap also testss data
429 verification and expire\-time verification for multi\-get.
430 .sp
431 Memslap testss multi\-get with both TCP and UDP. Because of
432 the different implementation of the ASCII protocol and binary protocol,
433 there are some differences between the two. For the ASCII protocol,
434 memaslap sends one “multi\-get” to the server once. For the
435 binary protocol, memaslap sends several single get commands
436 together as “multi\-get” to the server.
437 .SS UDP and TCP
438 .sp
439 Memslap testss both UDP and TCP. For TCP,
440 memaslap does not reconnect the memcached server if socket connections are
441 lost. If all the socket connections are lost or memcached server crashes,
442 memaslap will exit. If the user specifies the “\-\-reconnect”
443 option when socket connections are lost, it will reconnect them.
444 .sp
445 User can use “\-\-udp” to enable the UDP feature, but UDP comes with some
446 limitations:
447 .sp
448 UDP cannot set data more than 1400 bytes.
449 .sp
450 UDP is not testsed by the binary protocol because the binary protocol of
451 memcached does not tests that.
452 .sp
453 UDP doesn’t tests reconnection.
454 .SS Facebook test
455 .sp
456 Set data with TCP and multi\-get with UDP. Specify the following options:
457 .sp
458 "\-\-facebook \-\-division=50"
459 .sp
460 If you want to create thousands of TCP connections, specify the
461 .sp
462 "\-\-conn_sock=" option.
463 .sp
464 For example: \-\-facebook \-\-division=50 \-\-conn_sock=200
465 .sp
466 The above command means that memaslap will do facebook test,
467 each concurrency has 200 socket TCP connections and one UDP socket.
468 .sp
469 Memslap sets objects with the TCP socket, and multi\-gets 50
470 objects once with the UDP socket.
471 .sp
472 If you specify "\-\-division=50", the key size must be less that 25 bytes
473 because the UDP packet size is 1400 bytes.
474 .SS Replication test
475 .sp
476 For replication test, the user must specify at least two memcached servers.
477 The user can use “—rep_write=” option to enable feature.
478 .sp
479 For example:
480 .sp
481 \-\-servers=10.1.1.1:11211,10.1.1.2:11212 –rep_write=2
482 .sp
483 The above command means that there are 2 replication memcached servers,
484 memaslap will set objects to both server 0 and server 1, get
485 objects which are set to server 0 before from server 1, and also get objects
486 which are set to server 1 before from server 0. If server 0 crashes,
487 memaslap will only get objects from server 1. If server 0 comes
488 back to life again, memaslap will reconnect server 0. If both
489 server 0 and server 1 crash, memaslap will exit.
490 .SS Supports thousands of TCP connections
491 .sp
492 Start memaslap with "\-\-conn_sock=" or "\-n" to enable this
493 feature. Make sure that your system can tests opening thousands of files
494 and creating thousands of sockets. However, this feature does not tests
495 reconnection if sockets disconnect.
496 .sp
497 For example:
498 .sp
499 \-\-threads=8 \-\-concurrency=128 \-\-conn_sock=128
500 .sp
501 The above command means that memaslap starts up 8 threads, each
502 thread has 16 concurrencies, each concurrency has 128 TCP socket
503 connections, and the total number of TCP socket connections is 128 * 128 =
504 16384.
505 .SS Supports binary protocol
506 .sp
507 Start memaslap with "\-\-binary" or "\-B" options to enable this
508 feature. It testss all the above features except UDP, because the latest
509 memcached 1.3.3 does not implement binary UDP protocol.
510 .sp
511 For example:
512 .sp
513 \-\-binary
514 .sp
515 Since memcached 1.3.3 doesn\(aqt implement binary UDP protocol,
516 memaslap does not tests UDP. In addition, memcached 1.3.3 does not tests
517 multi\-get. If you specify "\-\-division=50" option, it just sends 50 get
518 commands together as “mulit\-get” to the server.
519 .SH CONFIGURATION FILE
520 .sp
521 This section describes the format of the configuration file. By default
522 when no configuration file is specified memaslap reads the default
523 one located at ~/.memaslap.cnf.
524 .sp
525 Below is a sample configuration file:
526 .sp
527 .nf
528 .ft C
529 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
530 #comments should start with \(aq#\(aq
531 #key
532 #start_len end_len proportion
533 #
534 #key length range from start_len to end_len
535 #start_len must be equal to or greater than 16
536 #end_len must be equal to or less than 250
537 #start_len must be equal to or greater than end_len
538 #memaslap will generate keys according to the key range
539 #proportion: indicates keys generated from one range accounts for the total
540 generated keys
541 #
542 #example1: key range 16~100 accounts for 80%
543 # key range 101~200 accounts for 10%
544 # key range 201~250 accounts for 10%
545 # total should be 1 (0.8+0.1+0.1 = 1)
546 #
547 # 16 100 0.8
548 # 101 200 0.1
549 # 201 249 0.1
550 #
551 #example2: all keys length are 128 bytes
552 #
553 # 128 128 1
554 key
555 128 128 1
556 #value
557 #start_len end_len proportion
558 #
559 #value length range from start_len to end_len
560 #start_len must be equal to or greater than 1
561 #end_len must be equal to or less than 1M
562 #start_len must be equal to or greater than end_len
563 #memaslap will generate values according to the value range
564 #proportion: indicates values generated from one range accounts for the
565 total generated values
566 #
567 #example1: value range 1~1000 accounts for 80%
568 # value range 1001~10000 accounts for 10%
569 # value range 10001~100000 accounts for 10%
570 # total should be 1 (0.8+0.1+0.1 = 1)
571 #
572 # 1 1000 0.8
573 # 1001 10000 0.1
574 # 10001 100000 0.1
575 #
576 #example2: all value length are 128 bytes
577 #
578 # 128 128 1
579 value
580 2048 2048 1
581 #cmd
582 #cmd_type cmd_proportion
583 #
584 #currently memaslap only testss get and set command.
585 #
586 #cmd_type
587 #set 0
588 #get 1
589 #
590 #example: set command accounts for 50%
591 # get command accounts for 50%
592 # total should be 1 (0.5+0.5 = 1)
593 #
594 # cmd
595 # 0 0.5
596 # 1 0.5
597 cmd
598 0 0.1
599 1.0 0.9
600 .ft P
601 .fi
602 .SH FORMAT OF OUTPUT
603 .sp
604 At the beginning, memaslap displays some configuration information as follows:
605 .sp
606 servers : 127.0.0.1:11211
607 .sp
608 threads count: 1
609 .sp
610 concurrency: 16
611 .sp
612 run time: 20s
613 .sp
614 windows size: 10k
615 .sp
616 set proportion: set_prop=0.10
617 .sp
618 get proportion: get_prop=0.90
619 .SS Where
620 .sp
621 servers : "servers"
622 .INDENT 0.0
623 .INDENT 3.5
624 The servers used by memaslap.
625 .UNINDENT
626 .UNINDENT
627 .sp
628 threads count
629 .INDENT 0.0
630 .INDENT 3.5
631 The number of threads memaslap runs with.
632 .UNINDENT
633 .UNINDENT
634 .sp
635 concurrency
636 .INDENT 0.0
637 .INDENT 3.5
638 The number of concurrencies memaslap runs with.
639 .UNINDENT
640 .UNINDENT
641 .sp
642 run time
643 .INDENT 0.0
644 .INDENT 3.5
645 How long to run memaslap.
646 .UNINDENT
647 .UNINDENT
648 .sp
649 windows size
650 .INDENT 0.0
651 .INDENT 3.5
652 The task window size of each concurrency.
653 .UNINDENT
654 .UNINDENT
655 .sp
656 set proportion
657 .INDENT 0.0
658 .INDENT 3.5
659 The proportion of set command.
660 .UNINDENT
661 .UNINDENT
662 .sp
663 get proportion
664 .INDENT 0.0
665 .INDENT 3.5
666 The proportion of get command.
667 .UNINDENT
668 .UNINDENT
669 .sp
670 The output of dynamic statistics is something like this:
671 .sp
672 .nf
673 .ft C
674 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
675 Get Statistics
676 Type Time(s) Ops TPS(ops/s) Net(M/s) Get_miss Min(us) Max(us)
677 Avg(us) Std_dev Geo_dist
678 Period 5 345826 69165 65.3 0 27 2198 203
679 95.43 177.29
680 Global 20 1257935 62896 71.8 0 26 3791 224
681 117.79 192.60
682
683
684 Set Statistics
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 38425 7685 7.3 0 42 628 240
688 88.05 220.21
689 Global 20 139780 6989 8.0 0 37 3790 253
690 117.93 224.83
691
692
693 Total Statistics
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 384252 76850 72.5 0 27 2198 207
697 94.72 181.18
698 Global 20 1397720 69886 79.7 0 26 3791 227
699 117.93 195.60
700 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
701 .ft P
702 .fi
703 .SS Where
704 .sp
705 Get Statistics
706 .INDENT 0.0
707 .INDENT 3.5
708 Statistics information of get command
709 .UNINDENT
710 .UNINDENT
711 .sp
712 Set Statistics
713 .INDENT 0.0
714 .INDENT 3.5
715 Statistics information of set command
716 .UNINDENT
717 .UNINDENT
718 .sp
719 Total Statistics
720 .INDENT 0.0
721 .INDENT 3.5
722 Statistics information of both get and set command
723 .UNINDENT
724 .UNINDENT
725 .sp
726 Period
727 .INDENT 0.0
728 .INDENT 3.5
729 Result within a period
730 .UNINDENT
731 .UNINDENT
732 .sp
733 Global
734 .INDENT 0.0
735 .INDENT 3.5
736 Accumulated results
737 .UNINDENT
738 .UNINDENT
739 .sp
740 Ops
741 .INDENT 0.0
742 .INDENT 3.5
743 Total operations
744 .UNINDENT
745 .UNINDENT
746 .sp
747 TPS
748 .INDENT 0.0
749 .INDENT 3.5
750 Throughput, operations/second
751 .UNINDENT
752 .UNINDENT
753 .sp
754 Net
755 .INDENT 0.0
756 .INDENT 3.5
757 The rate of network
758 .UNINDENT
759 .UNINDENT
760 .sp
761 Get_miss
762 .INDENT 0.0
763 .INDENT 3.5
764 How many objects can’t be gotten
765 .UNINDENT
766 .UNINDENT
767 .sp
768 Min
769 .INDENT 0.0
770 .INDENT 3.5
771 The minimum response time
772 .UNINDENT
773 .UNINDENT
774 .sp
775 Max
776 .INDENT 0.0
777 .INDENT 3.5
778 The maximum response time
779 .UNINDENT
780 .UNINDENT
781 .sp
782 Avg:
783 .INDENT 0.0
784 .INDENT 3.5
785 The average response time
786 .UNINDENT
787 .UNINDENT
788 .sp
789 Std_dev
790 .INDENT 0.0
791 .INDENT 3.5
792 Standard deviation of response time
793 .UNINDENT
794 .UNINDENT
795 .sp
796 Geo_dist
797 .INDENT 0.0
798 .INDENT 3.5
799 Geometric distribution based on natural exponential function
800 .UNINDENT
801 .UNINDENT
802 .sp
803 At the end, memaslap will output something like this:
804 .sp
805 .nf
806 .ft C
807 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
808 Get Statistics (1257956 events)
809 Min: 26
810 Max: 3791
811 Avg: 224
812 Geo: 192.60
813 Std: 116.23
814 Log2 Dist:
815 4: 0 10 84490 215345
816 8: 484890 459823 12543 824
817 12: 31
818
819 Set Statistics (139782 events)
820 Min: 37
821 Max: 3790
822 Avg: 253
823 Geo: 224.84
824 Std: 116.83
825 Log2 Dist:
826 4: 0 0 4200 16988
827 8: 50784 65574 2064 167
828 12: 5
829
830 Total Statistics (1397738 events)
831 Min: 26
832 Max: 3791
833 Avg: 227
834 Geo: 195.60
835 Std: 116.60
836 Log2 Dist:
837 4: 0 10 88690 232333
838 8: 535674 525397 14607 991
839 12: 36
840
841 cmd_get: 1257969
842 cmd_set: 139785
843 get_misses: 0
844 verify_misses: 0
845 verify_failed: 0
846 expired_get: 0
847 unexpired_unget: 0
848 written_bytes: 242516030
849 read_bytes: 1003702556
850 object_bytes: 152086080
851 packet_disorder: 0
852 packet_drop: 0
853 udp_timeout: 0
854
855 Run time: 20.0s Ops: 1397754 TPS: 69817 Net_rate: 59.4M/s
856 \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
857 .ft P
858 .fi
859 .SS Where
860 .sp
861 Get Statistics
862 .INDENT 0.0
863 .INDENT 3.5
864 Get statistics of response time
865 .UNINDENT
866 .UNINDENT
867 .sp
868 Set Statistics
869 .INDENT 0.0
870 .INDENT 3.5
871 Set statistics of response time
872 .UNINDENT
873 .UNINDENT
874 .sp
875 Total Statistics
876 .INDENT 0.0
877 .INDENT 3.5
878 Both get and set statistics of response time
879 .UNINDENT
880 .UNINDENT
881 .sp
882 Min
883 .INDENT 0.0
884 .INDENT 3.5
885 The accumulated and minimum response time
886 .UNINDENT
887 .UNINDENT
888 .sp
889 Max
890 .INDENT 0.0
891 .INDENT 3.5
892 The accumulated and maximum response time
893 .UNINDENT
894 .UNINDENT
895 .sp
896 Avg
897 .INDENT 0.0
898 .INDENT 3.5
899 The accumulated and average response time
900 .UNINDENT
901 .UNINDENT
902 .sp
903 Std
904 .INDENT 0.0
905 .INDENT 3.5
906 Standard deviation of response time
907 .UNINDENT
908 .UNINDENT
909 .sp
910 Log2 Dist
911 .INDENT 0.0
912 .INDENT 3.5
913 Geometric distribution based on logarithm 2
914 .UNINDENT
915 .UNINDENT
916 .sp
917 cmd_get
918 .INDENT 0.0
919 .INDENT 3.5
920 Total get commands done
921 .UNINDENT
922 .UNINDENT
923 .sp
924 cmd_set
925 .INDENT 0.0
926 .INDENT 3.5
927 Total set commands done
928 .UNINDENT
929 .UNINDENT
930 .sp
931 get_misses
932 .INDENT 0.0
933 .INDENT 3.5
934 How many objects can’t be gotten from server
935 .UNINDENT
936 .UNINDENT
937 .sp
938 verify_misses
939 .INDENT 0.0
940 .INDENT 3.5
941 How many objects need to verify but can’t get them
942 .UNINDENT
943 .UNINDENT
944 .sp
945 verify_failed
946 .INDENT 0.0
947 .INDENT 3.5
948 How many objects with insistent value
949 .UNINDENT
950 .UNINDENT
951 .sp
952 expired_get
953 .INDENT 0.0
954 .INDENT 3.5
955 How many objects are expired but we get them
956 .UNINDENT
957 .UNINDENT
958 .sp
959 unexpired_unget
960 .INDENT 0.0
961 .INDENT 3.5
962 How many objects are unexpired but we can’t get them
963 .UNINDENT
964 .UNINDENT
965 .sp
966 written_bytes
967 .INDENT 0.0
968 .INDENT 3.5
969 Total written bytes
970 .UNINDENT
971 .UNINDENT
972 .sp
973 read_bytes
974 .INDENT 0.0
975 .INDENT 3.5
976 Total read bytes
977 .UNINDENT
978 .UNINDENT
979 .sp
980 object_bytes
981 .INDENT 0.0
982 .INDENT 3.5
983 Total object bytes
984 .UNINDENT
985 .UNINDENT
986 .sp
987 packet_disorder
988 .INDENT 0.0
989 .INDENT 3.5
990 How many UDP packages are disorder
991 .UNINDENT
992 .UNINDENT
993 .sp
994 packet_drop
995 .INDENT 0.0
996 .INDENT 3.5
997 How many UDP packages are lost
998 .UNINDENT
999 .UNINDENT
1000 .sp
1001 udp_timeout
1002 .INDENT 0.0
1003 .INDENT 3.5
1004 How many times UDP time out happen
1005 .UNINDENT
1006 .UNINDENT
1007 .sp
1008 Run time
1009 .INDENT 0.0
1010 .INDENT 3.5
1011 Total run time
1012 .UNINDENT
1013 .UNINDENT
1014 .sp
1015 Ops
1016 .INDENT 0.0
1017 .INDENT 3.5
1018 Total operations
1019 .UNINDENT
1020 .UNINDENT
1021 .sp
1022 TPS
1023 .INDENT 0.0
1024 .INDENT 3.5
1025 Throughput, operations/second
1026 .UNINDENT
1027 .UNINDENT
1028 .sp
1029 Net_rate
1030 .INDENT 0.0
1031 .INDENT 3.5
1032 The average rate of network
1033 .UNINDENT
1034 .UNINDENT
1035 .SH OPTIONS
1036 .INDENT 0.0
1037 .TP
1038 .B \-s, \-\-servers=
1039 List one or more servers to connect. Servers count must be less than
1040 threads count. e.g.: \-\-servers=localhost:1234,localhost:11211
1041 .TP
1042 .B \-T, \-\-threads=
1043 Number of threads to startup, better equal to CPU numbers. Default 8.
1044 .TP
1045 .B \-c, \-\-concurrency=
1046 Number of concurrency to simulate with load. Default 128.
1047 .TP
1048 .B \-n, \-\-conn_sock=
1049 Number of TCP socks per concurrency. Default 1.
1050 .TP
1051 .B \-x, \-\-execute_number=
1052 Number of operations(get and set) to execute for the
1053 given test. Default 1000000.
1054 .TP
1055 .B \-t, \-\-time=
1056 How long the test to run, suffix: s\-seconds, m\-minutes, h\-hours,
1057 d\-days e.g.: \-\-time=2h.
1058 .TP
1059 .B \-F, \-\-cfg_cmd=
1060 Load the configure file to get command,key and value distribution list.
1061 .TP
1062 .B \-w, \-\-win_size=
1063 Task window size of each concurrency, suffix: K, M e.g.: \-\-win_size=10k.
1064 Default 10k.
1065 .TP
1066 .B \-X, \-\-fixed_size=
1067 Fixed length of value.
1068 .TP
1069 .B \-v, \-\-verify=
1070 The proportion of date verification, e.g.: \-\-verify=0.01
1071 .TP
1072 .B \-d, \-\-division=
1073 Number of keys to multi\-get once. Default 1, means single get.
1074 .TP
1075 .B \-S, \-\-stat_freq=
1076 Frequency of dumping statistic information. suffix: s\-seconds,
1077 m\-minutes, e.g.: \-\-resp_freq=10s.
1078 .TP
1079 .B \-e, \-\-exp_verify=
1080 The proportion of objects with expire time, e.g.: \-\-exp_verify=0.01.
1081 Default no object with expire time
1082 .TP
1083 .B \-o, \-\-overwrite=
1084 The proportion of objects need overwrite, e.g.: \-\-overwrite=0.01.
1085 Default never overwrite object.
1086 .UNINDENT
1087 .INDENT 0.0
1088 .TP
1089 .B \-R, \-\-reconnect
1090 Reconnect tests, when connection is closed it will be reconnected.
1091 .TP
1092 .B \-U, \-\-udp
1093 UDP tests, default memaslap uses TCP, TCP port and UDP port of
1094 server must be same.
1095 .TP
1096 .B \-a, \-\-facebook
1097 Whether it enables facebook test feature, set with TCP and multi\-get with UDP.
1098 .TP
1099 .B \-B, \-\-binary
1100 Whether it enables binary protocol. Default with ASCII protocol.
1101 .UNINDENT
1102 .INDENT 0.0
1103 .TP
1104 .B \-P, \-\-tps=
1105 Expected throughput, suffix: K, e.g.: \-\-tps=10k.
1106 .TP
1107 .B \-p, \-\-rep_write=
1108 The first nth servers can write data, e.g.: \-\-rep_write=2.
1109 .UNINDENT
1110 .INDENT 0.0
1111 .TP
1112 .B \-b, \-\-verbose
1113 Whether it outputs detailed information when verification fails.
1114 .TP
1115 .B \-h, \-\-help
1116 Display this message and then exit.
1117 .TP
1118 .B \-V, \-\-version
1119 Display the version of the application and then exit.
1120 .UNINDENT
1121 .SH EXAMPLES
1122 .sp
1123 memaslap \-s 127.0.0.1:11211 \-S 5s
1124 .sp
1125 memaslap \-s 127.0.0.1:11211 \-t 2m \-v 0.2 \-e 0.05 \-b
1126 .sp
1127 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-w 40k \-S 20s \-o 0.2
1128 .sp
1129 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-T 4 \-c 128 \-d 20 \-P 40k
1130 .sp
1131 memaslap \-s 127.0.0.1:11211 \-F config \-t 2m \-d 50 \-a \-n 40
1132 .sp
1133 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m
1134 .sp
1135 memaslap \-s 127.0.0.1:11211,127.0.0.1:11212 \-F config \-t 2m \-p 2
1136 .SH HOME
1137 .sp
1138 To find out more information please check:
1139 \fI\%http://libmemcached.org/\fP
1140 .SH AUTHORS
1141 .sp
1142 Mingqiang Zhuang <\fI\%mingqiangzhuang@hengtiansoft.com\fP> (Schooner Technolgy)
1143 Brian Aker, <\fI\%brian@tangent.org\fP>
1144 .SH SEE ALSO
1145 .sp
1146 \fImemcached(1)\fP \fIlibmemcached(3)\fP
1147 .SH AUTHOR
1148 Brian Aker
1149 .SH COPYRIGHT
1150 2011, Brian Aker DataDifferential, http://datadifferential.com/
1151 .\" Generated by docutils manpage writer.
1152 .\"
1153 .