Commit | Line | Data |
---|---|---|
71bfa161 JA |
1 | Table of contents |
2 | ----------------- | |
3 | ||
4 | 1. Overview | |
5 | 2. How fio works | |
6 | 3. Running fio | |
7 | 4. Job file format | |
8 | 5. Detailed list of parameters | |
9 | 6. Normal output | |
10 | 7. Terse output | |
11 | ||
12 | ||
13 | 1.0 Overview and history | |
14 | ------------------------ | |
15 | fio was originally written to save me the hassle of writing special test | |
16 | case programs when I wanted to test a specific workload, either for | |
17 | performance reasons or to find/reproduce a bug. The process of writing | |
18 | such a test app can be tiresome, especially if you have to do it often. | |
19 | Hence I needed a tool that would be able to simulate a given io workload | |
20 | without resorting to writing a tailored test case again and again. | |
21 | ||
22 | A test work load is difficult to define, though. There can be any number | |
23 | of processes or threads involved, and they can each be using their own | |
24 | way of generating io. You could have someone dirtying large amounts of | |
25 | memory in an memory mapped file, or maybe several threads issuing | |
26 | reads using asynchronous io. fio needed to be flexible enough to | |
27 | simulate both of these cases, and many more. | |
28 | ||
29 | 2.0 How fio works | |
30 | ----------------- | |
31 | The first step in getting fio to simulate a desired io workload, is | |
32 | writing a job file describing that specific setup. A job file may contain | |
33 | any number of threads and/or files - the typical contents of the job file | |
34 | is a global section defining shared parameters, and one or more job | |
35 | sections describing the jobs involved. When run, fio parses this file | |
36 | and sets everything up as described. If we break down a job from top to | |
37 | bottom, it contains the following basic parameters: | |
38 | ||
39 | IO type Defines the io pattern issued to the file(s). | |
40 | We may only be reading sequentially from this | |
41 | file(s), or we may be writing randomly. Or even | |
42 | mixing reads and writes, sequentially or randomly. | |
43 | ||
44 | Block size In how large chunks are we issuing io? This may be | |
45 | a single value, or it may describe a range of | |
46 | block sizes. | |
47 | ||
48 | IO size How much data are we going to be reading/writing. | |
49 | ||
50 | IO engine How do we issue io? We could be memory mapping the | |
51 | file, we could be using regular read/write, we | |
d0ff85df | 52 | could be using splice, async io, syslet, or even |
71bfa161 JA |
53 | SG (SCSI generic sg). |
54 | ||
6c219763 | 55 | IO depth If the io engine is async, how large a queuing |
71bfa161 JA |
56 | depth do we want to maintain? |
57 | ||
58 | IO type Should we be doing buffered io, or direct/raw io? | |
59 | ||
60 | Num files How many files are we spreading the workload over. | |
61 | ||
62 | Num threads How many threads or processes should we spread | |
63 | this workload over. | |
64 | ||
65 | The above are the basic parameters defined for a workload, in addition | |
66 | there's a multitude of parameters that modify other aspects of how this | |
67 | job behaves. | |
68 | ||
69 | ||
70 | 3.0 Running fio | |
71 | --------------- | |
72 | See the README file for command line parameters, there are only a few | |
73 | of them. | |
74 | ||
75 | Running fio is normally the easiest part - you just give it the job file | |
76 | (or job files) as parameters: | |
77 | ||
78 | $ fio job_file | |
79 | ||
80 | and it will start doing what the job_file tells it to do. You can give | |
81 | more than one job file on the command line, fio will serialize the running | |
82 | of those files. Internally that is the same as using the 'stonewall' | |
83 | parameter described the the parameter section. | |
84 | ||
b4692828 JA |
85 | If the job file contains only one job, you may as well just give the |
86 | parameters on the command line. The command line parameters are identical | |
87 | to the job parameters, with a few extra that control global parameters | |
88 | (see README). For example, for the job file parameter iodepth=2, the | |
c2b1e753 JA |
89 | mirror command line option would be --iodepth 2 or --iodepth=2. You can |
90 | also use the command line for giving more than one job entry. For each | |
91 | --name option that fio sees, it will start a new job with that name. | |
92 | Command line entries following a --name entry will apply to that job, | |
93 | until there are no more entries or a new --name entry is seen. This is | |
94 | similar to the job file options, where each option applies to the current | |
95 | job until a new [] job entry is seen. | |
b4692828 | 96 | |
71bfa161 JA |
97 | fio does not need to run as root, except if the files or devices specified |
98 | in the job section requires that. Some other options may also be restricted, | |
6c219763 | 99 | such as memory locking, io scheduler switching, and decreasing the nice value. |
71bfa161 JA |
100 | |
101 | ||
102 | 4.0 Job file format | |
103 | ------------------- | |
104 | As previously described, fio accepts one or more job files describing | |
105 | what it is supposed to do. The job file format is the classic ini file, | |
106 | where the names enclosed in [] brackets define the job name. You are free | |
107 | to use any ascii name you want, except 'global' which has special meaning. | |
108 | A global section sets defaults for the jobs described in that file. A job | |
109 | may override a global section parameter, and a job file may even have | |
110 | several global sections if so desired. A job is only affected by a global | |
65db0851 JA |
111 | section residing above it. If the first character in a line is a ';' or a |
112 | '#', the entire line is discarded as a comment. | |
71bfa161 | 113 | |
3c54bc46 | 114 | So let's look at a really simple job file that defines two processes, each |
71bfa161 JA |
115 | randomly reading from a 128MiB file. |
116 | ||
117 | ; -- start job file -- | |
118 | [global] | |
119 | rw=randread | |
120 | size=128m | |
121 | ||
122 | [job1] | |
123 | ||
124 | [job2] | |
125 | ||
126 | ; -- end job file -- | |
127 | ||
128 | As you can see, the job file sections themselves are empty as all the | |
129 | described parameters are shared. As no filename= option is given, fio | |
c2b1e753 JA |
130 | makes up a filename for each of the jobs as it sees fit. On the command |
131 | line, this job would look as follows: | |
132 | ||
133 | $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2 | |
134 | ||
71bfa161 | 135 | |
3c54bc46 | 136 | Let's look at an example that has a number of processes writing randomly |
71bfa161 JA |
137 | to files. |
138 | ||
139 | ; -- start job file -- | |
140 | [random-writers] | |
141 | ioengine=libaio | |
142 | iodepth=4 | |
143 | rw=randwrite | |
144 | bs=32k | |
145 | direct=0 | |
146 | size=64m | |
147 | numjobs=4 | |
148 | ||
149 | ; -- end job file -- | |
150 | ||
151 | Here we have no global section, as we only have one job defined anyway. | |
152 | We want to use async io here, with a depth of 4 for each file. We also | |
153 | increased the buffer size used to 32KiB and define numjobs to 4 to | |
154 | fork 4 identical jobs. The result is 4 processes each randomly writing | |
b4692828 JA |
155 | to their own 64MiB file. Instead of using the above job file, you could |
156 | have given the parameters on the command line. For this case, you would | |
157 | specify: | |
158 | ||
159 | $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4 | |
71bfa161 | 160 | |
3c54bc46 AC |
161 | fio also supports environment variable expansion in job files. Any |
162 | substring of the form "${VARNAME}" as part of an option value (in other | |
163 | words, on the right of the `='), will be expanded to the value of the | |
164 | environment variable called VARNAME. If no such environment variable | |
165 | is defined, or VARNAME is the empty string, the empty string will be | |
166 | substituted. | |
167 | ||
168 | As an example, let's look at a sample fio invocation and job file: | |
169 | ||
170 | $ SIZE=64m NUMJOBS=4 fio jobfile.fio | |
171 | ||
172 | ; -- start job file -- | |
173 | [random-writers] | |
174 | rw=randwrite | |
175 | size=${SIZE} | |
176 | numjobs=${NUMJOBS} | |
177 | ; -- end job file -- | |
178 | ||
179 | This will expand to the following equivalent job file at runtime: | |
180 | ||
181 | ; -- start job file -- | |
182 | [random-writers] | |
183 | rw=randwrite | |
184 | size=64m | |
185 | numjobs=4 | |
186 | ; -- end job file -- | |
187 | ||
71bfa161 JA |
188 | fio ships with a few example job files, you can also look there for |
189 | inspiration. | |
190 | ||
191 | ||
192 | 5.0 Detailed list of parameters | |
193 | ------------------------------- | |
194 | ||
195 | This section describes in details each parameter associated with a job. | |
196 | Some parameters take an option of a given type, such as an integer or | |
197 | a string. The following types are used: | |
198 | ||
199 | str String. This is a sequence of alpha characters. | |
6d16ecb6 | 200 | int Integer. A whole number value, can be negative. If prefixed with |
bf9a3edb | 201 | 0x, the integer is assumed to be of base 16 (hexadecimal). |
e417fd66 JA |
202 | time Integer with possible time postfix. In seconds unless otherwise |
203 | specified, use eg 10m for 10 minutes. Accepts s/m/h for seconds, | |
204 | minutes, and hours. | |
71bfa161 JA |
205 | siint SI integer. A whole number value, which may contain a postfix |
206 | describing the base of the number. Accepted postfixes are k/m/g, | |
6c219763 | 207 | meaning kilo, mega, and giga. So if you want to specify 4096, |
71bfa161 JA |
208 | you could either write out '4096' or just give 4k. The postfixes |
209 | signify base 2 values, so 1024 is 1k and 1024k is 1m and so on. | |
43159d18 | 210 | If the option accepts an upper and lower range, use a colon ':' |
bf9a3edb | 211 | or minus '-' to separate such values. See irange. |
71bfa161 JA |
212 | bool Boolean. Usually parsed as an integer, however only defined for |
213 | true and false (1 and 0). | |
214 | irange Integer range with postfix. Allows value range to be given, such | |
bf9a3edb | 215 | as 1024-4096. A colon may also be used as the separator, eg |
0c9baf91 JA |
216 | 1k:4k. If the option allows two sets of ranges, they can be |
217 | specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see | |
218 | siint. | |
71bfa161 JA |
219 | |
220 | With the above in mind, here follows the complete list of fio job | |
221 | parameters. | |
222 | ||
223 | name=str ASCII name of the job. This may be used to override the | |
224 | name printed by fio for this job. Otherwise the job | |
c2b1e753 | 225 | name is used. On the command line this parameter has the |
6c219763 | 226 | special purpose of also signaling the start of a new |
c2b1e753 | 227 | job. |
71bfa161 | 228 | |
61697c37 JA |
229 | description=str Text description of the job. Doesn't do anything except |
230 | dump this text description when this job is run. It's | |
231 | not parsed. | |
232 | ||
71bfa161 JA |
233 | directory=str Prefix filenames with this directory. Used to places files |
234 | in a different location than "./". | |
235 | ||
236 | filename=str Fio normally makes up a filename based on the job name, | |
237 | thread number, and file number. If you want to share | |
238 | files between threads in a job or several jobs, specify | |
ed92ac0c | 239 | a filename for each of them to override the default. If |
414c2a3e JA |
240 | the ioengine used is 'net', the filename is the host, port, |
241 | and protocol to use in the format of =host/port/protocol. | |
242 | See ioengine=net for more. If the ioengine is file based, you | |
243 | can specify a number of files by separating the names with a | |
244 | ':' colon. So if you wanted a job to open /dev/sda and /dev/sdb | |
245 | as the two working files, you would use | |
246 | filename=/dev/sda:/dev/sdb. '-' is a reserved name, meaning | |
247 | stdin or stdout. Which of the two depends on the read/write | |
248 | direction set. | |
71bfa161 | 249 | |
bbf6b540 JA |
250 | opendir=str Tell fio to recursively add any file it can find in this |
251 | directory and down the file system tree. | |
252 | ||
4d4e80f2 JA |
253 | lockfile=str Fio defaults to not doing any locking files before it does |
254 | IO to them. If a file or file descriptor is shared, fio | |
255 | can serialize IO to that file to make the end result | |
256 | consistent. This is usual for emulating real workloads that | |
257 | share files. The lock modes are: | |
258 | ||
259 | none No locking. The default. | |
260 | exclusive Only one thread/process may do IO, | |
261 | excluding all others. | |
262 | readwrite Read-write locking on the file. Many | |
263 | readers may access the file at the | |
264 | same time, but writes get exclusive | |
265 | access. | |
266 | ||
267 | The option may be post-fixed with a lock batch number. If | |
268 | set, then each thread/process may do that amount of IOs to | |
bf9a3edb | 269 | the file before giving up the lock. Since lock acquisition is |
4d4e80f2 | 270 | expensive, batching the lock/unlocks will speed up IO. |
29c1349f | 271 | |
d3aad8f2 | 272 | readwrite=str |
71bfa161 JA |
273 | rw=str Type of io pattern. Accepted values are: |
274 | ||
275 | read Sequential reads | |
276 | write Sequential writes | |
277 | randwrite Random writes | |
278 | randread Random reads | |
279 | rw Sequential mixed reads and writes | |
280 | randrw Random mixed reads and writes | |
281 | ||
282 | For the mixed io types, the default is to split them 50/50. | |
283 | For certain types of io the result may still be skewed a bit, | |
211097b2 JA |
284 | since the speed may be different. It is possible to specify |
285 | a number of IO's to do before getting a new offset - this | |
286 | is only useful for random IO, where fio would normally | |
287 | generate a new random offset for every IO. If you append | |
288 | eg 8 to randread, you would get a new random offset for | |
289 | every 8 IO's. The result would be a seek for only every 8 | |
290 | IO's, instead of for every IO. Use rw=randread:8 to specify | |
291 | that. | |
71bfa161 | 292 | |
ee738499 JA |
293 | randrepeat=bool For random IO workloads, seed the generator in a predictable |
294 | way so that results are repeatable across repetitions. | |
295 | ||
d2f3ac35 JA |
296 | fadvise_hint=bool By default, fio will use fadvise() to advise the kernel |
297 | on what IO patterns it is likely to issue. Sometimes you | |
298 | want to test specific IO patterns without telling the | |
299 | kernel about it, in which case you can disable this option. | |
300 | If set, fio will use POSIX_FADV_SEQUENTIAL for sequential | |
301 | IO and POSIX_FADV_RANDOM for random IO. | |
302 | ||
7616cafe JA |
303 | size=siint The total size of file io for this job. Fio will run until |
304 | this many bytes has been transferred, unless runtime is | |
305 | limited by other options (such as 'runtime', for instance). | |
306 | Unless specific nr_files and filesize options are given, | |
307 | fio will divide this size between the available files | |
308 | specified by the job. | |
71bfa161 | 309 | |
9c60ce64 JA |
310 | filesize=siint Individual file sizes. May be a range, in which case fio |
311 | will select sizes for files at random within the given range | |
312 | and limited to 'size' in total (if that is given). If not | |
313 | given, each created file is the same size. | |
314 | ||
aa31f1f1 SL |
315 | fill_device=bool Sets size to something really large and waits for ENOSPC (no |
316 | space left on device) as the terminating condition. Only makes | |
317 | sense with sequential write. | |
318 | ||
d3aad8f2 | 319 | blocksize=siint |
f90eff5a JA |
320 | bs=siint The block size used for the io units. Defaults to 4k. Values |
321 | can be given for both read and writes. If a single siint is | |
322 | given, it will apply to both. If a second siint is specified | |
323 | after a comma, it will apply to writes only. In other words, | |
324 | the format is either bs=read_and_write or bs=read,write. | |
325 | bs=4k,8k will thus use 4k blocks for reads, and 8k blocks | |
787f7e95 JA |
326 | for writes. If you only wish to set the write size, you |
327 | can do so by passing an empty read size - bs=,8k will set | |
328 | 8k for writes and leave the read default value. | |
a00735e6 | 329 | |
d3aad8f2 | 330 | blocksize_range=irange |
71bfa161 JA |
331 | bsrange=irange Instead of giving a single block size, specify a range |
332 | and fio will mix the issued io block sizes. The issued | |
333 | io unit will always be a multiple of the minimum value | |
f90eff5a JA |
334 | given (also see bs_unaligned). Applies to both reads and |
335 | writes, however a second range can be given after a comma. | |
336 | See bs=. | |
a00735e6 | 337 | |
564ca972 JA |
338 | bssplit=str Sometimes you want even finer grained control of the |
339 | block sizes issued, not just an even split between them. | |
340 | This option allows you to weight various block sizes, | |
341 | so that you are able to define a specific amount of | |
342 | block sizes issued. The format for this option is: | |
343 | ||
344 | bssplit=blocksize/percentage:blocksize/percentage | |
345 | ||
346 | for as many block sizes as needed. So if you want to define | |
347 | a workload that has 50% 64k blocks, 10% 4k blocks, and | |
348 | 40% 32k blocks, you would write: | |
349 | ||
350 | bssplit=4k/10:64k/50:32k/40 | |
351 | ||
352 | Ordering does not matter. If the percentage is left blank, | |
353 | fio will fill in the remaining values evenly. So a bssplit | |
354 | option like this one: | |
355 | ||
356 | bssplit=4k/50:1k/:32k/ | |
357 | ||
358 | would have 50% 4k ios, and 25% 1k and 32k ios. The percentages | |
359 | always add up to 100, if bssplit is given a range that adds | |
360 | up to more, it will error out. | |
361 | ||
d3aad8f2 | 362 | blocksize_unaligned |
690adba3 JA |
363 | bs_unaligned If this option is given, any byte size value within bsrange |
364 | may be used as a block range. This typically wont work with | |
365 | direct IO, as that normally requires sector alignment. | |
71bfa161 | 366 | |
e9459e5a JA |
367 | zero_buffers If this option is given, fio will init the IO buffers to |
368 | all zeroes. The default is to fill them with random data. | |
369 | ||
5973cafb JA |
370 | refill_buffers If this option is given, fio will refill the IO buffers |
371 | on every submit. The default is to only fill it at init | |
372 | time and reuse that data. Only makes sense if zero_buffers | |
41ccd845 JA |
373 | isn't specified, naturally. If data verification is enabled, |
374 | refill_buffers is also automatically enabled. | |
5973cafb | 375 | |
71bfa161 JA |
376 | nrfiles=int Number of files to use for this job. Defaults to 1. |
377 | ||
390b1537 JA |
378 | openfiles=int Number of files to keep open at the same time. Defaults to |
379 | the same as nrfiles, can be set smaller to limit the number | |
380 | simultaneous opens. | |
381 | ||
5af1c6f3 JA |
382 | file_service_type=str Defines how fio decides which file from a job to |
383 | service next. The following types are defined: | |
384 | ||
385 | random Just choose a file at random. | |
386 | ||
387 | roundrobin Round robin over open files. This | |
388 | is the default. | |
389 | ||
1907dbc6 JA |
390 | The string can have a number appended, indicating how |
391 | often to switch to a new file. So if option random:4 is | |
392 | given, fio will switch to a new random file after 4 ios | |
393 | have been issued. | |
394 | ||
71bfa161 JA |
395 | ioengine=str Defines how the job issues io to the file. The following |
396 | types are defined: | |
397 | ||
398 | sync Basic read(2) or write(2) io. lseek(2) is | |
399 | used to position the io location. | |
400 | ||
a31041ea | 401 | psync Basic pread(2) or pwrite(2) io. |
402 | ||
e05af9e5 | 403 | vsync Basic readv(2) or writev(2) IO. |
1d2af02a | 404 | |
15d182aa JA |
405 | libaio Linux native asynchronous io. Note that Linux |
406 | may only support queued behaviour with | |
407 | non-buffered IO (set direct=1 or buffered=0). | |
71bfa161 JA |
408 | |
409 | posixaio glibc posix asynchronous io. | |
410 | ||
417f0068 JA |
411 | solarisaio Solaris native asynchronous io. |
412 | ||
71bfa161 JA |
413 | mmap File is memory mapped and data copied |
414 | to/from using memcpy(3). | |
415 | ||
416 | splice splice(2) is used to transfer the data and | |
417 | vmsplice(2) to transfer data from user | |
418 | space to the kernel. | |
419 | ||
d0ff85df JA |
420 | syslet-rw Use the syslet system calls to make |
421 | regular read/write async. | |
422 | ||
71bfa161 | 423 | sg SCSI generic sg v3 io. May either be |
6c219763 | 424 | synchronous using the SG_IO ioctl, or if |
71bfa161 JA |
425 | the target is an sg character device |
426 | we use read(2) and write(2) for asynchronous | |
427 | io. | |
428 | ||
a94ea28b JA |
429 | null Doesn't transfer any data, just pretends |
430 | to. This is mainly used to exercise fio | |
431 | itself and for debugging/testing purposes. | |
432 | ||
ed92ac0c JA |
433 | net Transfer over the network to given host:port. |
434 | 'filename' must be set appropriately to | |
414c2a3e | 435 | filename=host/port/protocol regardless of send |
ed92ac0c | 436 | or receive, if the latter only the port |
414c2a3e JA |
437 | argument is used. 'host' may be an IP address |
438 | or hostname, port is the port number to be used, | |
439 | and protocol may be 'udp' or 'tcp'. If no | |
440 | protocol is given, TCP is used. | |
ed92ac0c | 441 | |
9cce02e8 JA |
442 | netsplice Like net, but uses splice/vmsplice to |
443 | map data and send/receive. | |
444 | ||
53aec0a4 | 445 | cpuio Doesn't transfer any data, but burns CPU |
ba0fbe10 JA |
446 | cycles according to the cpuload= and |
447 | cpucycle= options. Setting cpuload=85 | |
448 | will cause that job to do nothing but burn | |
36ecec83 GP |
449 | 85% of the CPU. In case of SMP machines, |
450 | use numjobs=<no_of_cpu> to get desired CPU | |
451 | usage, as the cpuload only loads a single | |
452 | CPU at the desired rate. | |
ba0fbe10 | 453 | |
e9a1806f JA |
454 | guasi The GUASI IO engine is the Generic Userspace |
455 | Asyncronous Syscall Interface approach | |
456 | to async IO. See | |
457 | ||
458 | http://www.xmailserver.org/guasi-lib.html | |
459 | ||
460 | for more info on GUASI. | |
461 | ||
8a7bd877 JA |
462 | external Prefix to specify loading an external |
463 | IO engine object file. Append the engine | |
464 | filename, eg ioengine=external:/tmp/foo.o | |
465 | to load ioengine foo.o in /tmp. | |
466 | ||
71bfa161 JA |
467 | iodepth=int This defines how many io units to keep in flight against |
468 | the file. The default is 1 for each file defined in this | |
469 | job, can be overridden with a larger value for higher | |
470 | concurrency. | |
471 | ||
4950421a | 472 | iodepth_batch_submit=int |
cb5ab512 | 473 | iodepth_batch=int This defines how many pieces of IO to submit at once. |
89e820f6 JA |
474 | It defaults to 1 which means that we submit each IO |
475 | as soon as it is available, but can be raised to submit | |
476 | bigger batches of IO at the time. | |
cb5ab512 | 477 | |
4950421a JA |
478 | iodepth_batch_complete=int This defines how many pieces of IO to retrieve |
479 | at once. It defaults to 1 which means that we'll ask | |
480 | for a minimum of 1 IO in the retrieval process from | |
481 | the kernel. The IO retrieval will go on until we | |
482 | hit the limit set by iodepth_low. If this variable is | |
483 | set to 0, then fio will always check for completed | |
484 | events before queuing more IO. This helps reduce | |
485 | IO latency, at the cost of more retrieval system calls. | |
486 | ||
e916b390 JA |
487 | iodepth_low=int The low water mark indicating when to start filling |
488 | the queue again. Defaults to the same as iodepth, meaning | |
489 | that fio will attempt to keep the queue full at all times. | |
490 | If iodepth is set to eg 16 and iodepth_low is set to 4, then | |
491 | after fio has filled the queue of 16 requests, it will let | |
492 | the depth drain down to 4 before starting to fill it again. | |
493 | ||
71bfa161 | 494 | direct=bool If value is true, use non-buffered io. This is usually |
76a43db4 JA |
495 | O_DIRECT. |
496 | ||
497 | buffered=bool If value is true, use buffered io. This is the opposite | |
498 | of the 'direct' option. Defaults to true. | |
71bfa161 JA |
499 | |
500 | offset=siint Start io at the given offset in the file. The data before | |
501 | the given offset will not be touched. This effectively | |
502 | caps the file size at real_size - offset. | |
503 | ||
504 | fsync=int If writing to a file, issue a sync of the dirty data | |
505 | for every number of blocks given. For example, if you give | |
506 | 32 as a parameter, fio will sync the file for every 32 | |
507 | writes issued. If fio is using non-buffered io, we may | |
508 | not sync the file. The exception is the sg io engine, which | |
6c219763 | 509 | synchronizes the disk cache anyway. |
71bfa161 | 510 | |
5036fc1e JA |
511 | overwrite=bool If true, writes to a file will always overwrite existing |
512 | data. If the file doesn't already exist, it will be | |
513 | created before the write phase begins. If the file exists | |
514 | and is large enough for the specified write phase, nothing | |
515 | will be done. | |
71bfa161 JA |
516 | |
517 | end_fsync=bool If true, fsync file contents when the job exits. | |
518 | ||
ebb1415f JA |
519 | fsync_on_close=bool If true, fio will fsync() a dirty file on close. |
520 | This differs from end_fsync in that it will happen on every | |
521 | file close, not just at the end of the job. | |
522 | ||
71bfa161 JA |
523 | rwmixread=int How large a percentage of the mix should be reads. |
524 | ||
525 | rwmixwrite=int How large a percentage of the mix should be writes. If both | |
526 | rwmixread and rwmixwrite is given and the values do not add | |
527 | up to 100%, the latter of the two will be used to override | |
528 | the first. | |
529 | ||
bb8895e0 JA |
530 | norandommap Normally fio will cover every block of the file when doing |
531 | random IO. If this option is given, fio will just get a | |
532 | new random offset without looking at past io history. This | |
533 | means that some blocks may not be read or written, and that | |
534 | some blocks may be read/written more than once. This option | |
8347239a JA |
535 | is mutually exclusive with verify= if and only if multiple |
536 | blocksizes (via bsrange=) are used, since fio only tracks | |
537 | complete rewrites of blocks. | |
bb8895e0 | 538 | |
2b386d25 JA |
539 | softrandommap See norandommap. If fio runs with the random block map enabled |
540 | and it fails to allocate the map, if this option is set it | |
541 | will continue without a random block map. As coverage will | |
542 | not be as complete as with random maps, this option is | |
543 | disabled by default. | |
544 | ||
71bfa161 JA |
545 | nice=int Run the job with the given nice value. See man nice(2). |
546 | ||
547 | prio=int Set the io priority value of this job. Linux limits us to | |
548 | a positive value between 0 and 7, with 0 being the highest. | |
549 | See man ionice(1). | |
550 | ||
551 | prioclass=int Set the io priority class. See man ionice(1). | |
552 | ||
553 | thinktime=int Stall the job x microseconds after an io has completed before | |
554 | issuing the next. May be used to simulate processing being | |
48097d5c JA |
555 | done by an application. See thinktime_blocks and |
556 | thinktime_spin. | |
557 | ||
558 | thinktime_spin=int | |
559 | Only valid if thinktime is set - pretend to spend CPU time | |
560 | doing something with the data received, before falling back | |
561 | to sleeping for the rest of the period specified by | |
562 | thinktime. | |
9c1f7434 JA |
563 | |
564 | thinktime_blocks | |
565 | Only valid if thinktime is set - control how many blocks | |
566 | to issue, before waiting 'thinktime' usecs. If not set, | |
567 | defaults to 1 which will make fio wait 'thinktime' usecs | |
568 | after every block. | |
71bfa161 JA |
569 | |
570 | rate=int Cap the bandwidth used by this job to this number of KiB/sec. | |
571 | ||
572 | ratemin=int Tell fio to do whatever it can to maintain at least this | |
4e991c23 JA |
573 | bandwidth. Failing to meet this requirement, will cause |
574 | the job to exit. | |
575 | ||
576 | rate_iops=int Cap the bandwidth to this number of IOPS. Basically the same | |
577 | as rate, just specified independently of bandwidth. If the | |
578 | job is given a block size range instead of a fixed value, | |
579 | the smallest block size is used as the metric. | |
580 | ||
581 | rate_iops_min=int If fio doesn't meet this rate of IO, it will cause | |
582 | the job to exit. | |
71bfa161 JA |
583 | |
584 | ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number | |
6c219763 | 585 | of milliseconds. |
71bfa161 JA |
586 | |
587 | cpumask=int Set the CPU affinity of this job. The parameter given is a | |
a08bc17f JA |
588 | bitmask of allowed CPU's the job may run on. So if you want |
589 | the allowed CPUs to be 1 and 5, you would pass the decimal | |
590 | value of (1 << 1 | 1 << 5), or 34. See man | |
7dbb6eba | 591 | sched_setaffinity(2). This may not work on all supported |
b0ea08ce JA |
592 | operating systems or kernel versions. This option doesn't |
593 | work well for a higher CPU count than what you can store in | |
594 | an integer mask, so it can only control cpus 1-32. For | |
595 | boxes with larger CPU counts, use cpus_allowed. | |
71bfa161 | 596 | |
d2e268b0 JA |
597 | cpus_allowed=str Controls the same options as cpumask, but it allows a text |
598 | setting of the permitted CPUs instead. So to use CPUs 1 and | |
62a7273d JA |
599 | 5, you would specify cpus_allowed=1,5. This options also |
600 | allows a range of CPUs. Say you wanted a binding to CPUs | |
601 | 1, 5, and 8-15, you would set cpus_allowed=1,5,8-15. | |
d2e268b0 | 602 | |
e417fd66 | 603 | startdelay=time Start this job the specified number of seconds after fio |
71bfa161 JA |
604 | has started. Only useful if the job file contains several |
605 | jobs, and you want to delay starting some jobs to a certain | |
606 | time. | |
607 | ||
e417fd66 | 608 | runtime=time Tell fio to terminate processing after the specified number |
71bfa161 JA |
609 | of seconds. It can be quite hard to determine for how long |
610 | a specified job will run, so this parameter is handy to | |
611 | cap the total runtime to a given time. | |
612 | ||
cf4464ca | 613 | time_based If set, fio will run for the duration of the runtime |
bf9a3edb | 614 | specified even if the file(s) are completely read or |
cf4464ca JA |
615 | written. It will simply loop over the same workload |
616 | as many times as the runtime allows. | |
617 | ||
e417fd66 | 618 | ramp_time=time If set, fio will run the specified workload for this amount |
721938ae JA |
619 | of time before logging any performance numbers. Useful for |
620 | letting performance settle before logging results, thus | |
b29ee5b3 JA |
621 | minimizing the runtime required for stable results. Note |
622 | that the ramp_time is considered lead in time for a job, | |
623 | thus it will increase the total runtime if a special timeout | |
624 | or runtime is specified. | |
721938ae | 625 | |
71bfa161 JA |
626 | invalidate=bool Invalidate the buffer/page cache parts for this file prior |
627 | to starting io. Defaults to true. | |
628 | ||
629 | sync=bool Use sync io for buffered writes. For the majority of the | |
630 | io engines, this means using O_SYNC. | |
631 | ||
d3aad8f2 | 632 | iomem=str |
71bfa161 JA |
633 | mem=str Fio can use various types of memory as the io unit buffer. |
634 | The allowed values are: | |
635 | ||
636 | malloc Use memory from malloc(3) as the buffers. | |
637 | ||
638 | shm Use shared memory as the buffers. Allocated | |
639 | through shmget(2). | |
640 | ||
74b025b0 JA |
641 | shmhuge Same as shm, but use huge pages as backing. |
642 | ||
313cb206 JA |
643 | mmap Use mmap to allocate buffers. May either be |
644 | anonymous memory, or can be file backed if | |
645 | a filename is given after the option. The | |
646 | format is mem=mmap:/path/to/file. | |
71bfa161 | 647 | |
d0bdaf49 JA |
648 | mmaphuge Use a memory mapped huge file as the buffer |
649 | backing. Append filename after mmaphuge, ala | |
650 | mem=mmaphuge:/hugetlbfs/file | |
651 | ||
71bfa161 | 652 | The area allocated is a function of the maximum allowed |
5394ae5f JA |
653 | bs size for the job, multiplied by the io depth given. Note |
654 | that for shmhuge and mmaphuge to work, the system must have | |
655 | free huge pages allocated. This can normally be checked | |
656 | and set by reading/writing /proc/sys/vm/nr_hugepages on a | |
657 | Linux system. Fio assumes a huge page is 4MiB in size. So | |
658 | to calculate the number of huge pages you need for a given | |
659 | job file, add up the io depth of all jobs (normally one unless | |
660 | iodepth= is used) and multiply by the maximum bs set. Then | |
661 | divide that number by the huge page size. You can see the | |
662 | size of the huge pages in /proc/meminfo. If no huge pages | |
663 | are allocated by having a non-zero number in nr_hugepages, | |
56bb17f2 | 664 | using mmaphuge or shmhuge will fail. Also see hugepage-size. |
5394ae5f JA |
665 | |
666 | mmaphuge also needs to have hugetlbfs mounted and the file | |
667 | location should point there. So if it's mounted in /huge, | |
668 | you would use mem=mmaphuge:/huge/somefile. | |
71bfa161 | 669 | |
56bb17f2 JA |
670 | hugepage-size=siint |
671 | Defines the size of a huge page. Must at least be equal | |
672 | to the system setting, see /proc/meminfo. Defaults to 4MiB. | |
c51074e7 JA |
673 | Should probably always be a multiple of megabytes, so using |
674 | hugepage-size=Xm is the preferred way to set this to avoid | |
675 | setting a non-pow-2 bad value. | |
56bb17f2 | 676 | |
71bfa161 JA |
677 | exitall When one job finishes, terminate the rest. The default is |
678 | to wait for each job to finish, sometimes that is not the | |
679 | desired action. | |
680 | ||
681 | bwavgtime=int Average the calculated bandwidth over the given time. Value | |
6c219763 | 682 | is specified in milliseconds. |
71bfa161 JA |
683 | |
684 | create_serialize=bool If true, serialize the file creating for the jobs. | |
685 | This may be handy to avoid interleaving of data | |
686 | files, which may greatly depend on the filesystem | |
687 | used and even the number of processors in the system. | |
688 | ||
689 | create_fsync=bool fsync the data file after creation. This is the | |
690 | default. | |
691 | ||
e545a6ce | 692 | unlink=bool Unlink the job files when done. Not the default, as repeated |
bf9a3edb JA |
693 | runs of that job would then waste time recreating the file |
694 | set again and again. | |
71bfa161 JA |
695 | |
696 | loops=int Run the specified number of iterations of this job. Used | |
697 | to repeat the same workload a given number of times. Defaults | |
698 | to 1. | |
699 | ||
68e1f29a | 700 | do_verify=bool Run the verify phase after a write phase. Only makes sense if |
e84c73a8 SL |
701 | verify is set. Defaults to 1. |
702 | ||
71bfa161 JA |
703 | verify=str If writing to a file, fio can verify the file contents |
704 | after each iteration of the job. The allowed values are: | |
705 | ||
706 | md5 Use an md5 sum of the data area and store | |
707 | it in the header of each block. | |
708 | ||
17dc34df JA |
709 | crc64 Use an experimental crc64 sum of the data |
710 | area and store it in the header of each | |
711 | block. | |
712 | ||
bac39e0e JA |
713 | crc32c Use a crc32c sum of the data area and store |
714 | it in the header of each block. | |
715 | ||
3845591f JA |
716 | crc32c-intel Use hardware assisted crc32c calcuation |
717 | provided on SSE4.2 enabled processors. | |
718 | ||
71bfa161 JA |
719 | crc32 Use a crc32 sum of the data area and store |
720 | it in the header of each block. | |
721 | ||
969f7ed3 JA |
722 | crc16 Use a crc16 sum of the data area and store |
723 | it in the header of each block. | |
724 | ||
17dc34df JA |
725 | crc7 Use a crc7 sum of the data area and store |
726 | it in the header of each block. | |
727 | ||
cd14cc10 JA |
728 | sha512 Use sha512 as the checksum function. |
729 | ||
730 | sha256 Use sha256 as the checksum function. | |
731 | ||
7437ee87 SL |
732 | meta Write extra information about each io |
733 | (timestamp, block number etc.). The block | |
734 | number is verified. | |
735 | ||
36690c9b JA |
736 | null Only pretend to verify. Useful for testing |
737 | internals with ioengine=null, not for much | |
738 | else. | |
739 | ||
6c219763 | 740 | This option can be used for repeated burn-in tests of a |
71bfa161 JA |
741 | system to make sure that the written data is also |
742 | correctly read back. | |
743 | ||
160b966d JA |
744 | verifysort=bool If set, fio will sort written verify blocks when it deems |
745 | it faster to read them back in a sorted manner. This is | |
746 | often the case when overwriting an existing file, since | |
747 | the blocks are already laid out in the file system. You | |
748 | can ignore this option unless doing huge amounts of really | |
749 | fast IO where the red-black tree sorting CPU time becomes | |
750 | significant. | |
3f9f4e26 | 751 | |
a59e170d | 752 | verify_offset=siint Swap the verification header with data somewhere else |
546a9142 SL |
753 | in the block before writing. Its swapped back before |
754 | verifying. | |
755 | ||
a59e170d | 756 | verify_interval=siint Write the verification header at a finer granularity |
3f9f4e26 SL |
757 | than the blocksize. It will be written for chunks the |
758 | size of header_interval. blocksize should divide this | |
759 | evenly. | |
90059d65 | 760 | |
e28218f3 SL |
761 | verify_pattern=int If set, fio will fill the io buffers with this |
762 | pattern. Fio defaults to filling with totally random | |
763 | bytes, but sometimes it's interesting to fill with a known | |
764 | pattern for io verification purposes. Depending on the | |
765 | width of the pattern, fio will fill 1/2/3/4 bytes of the | |
766 | buffer at the time. The verify_pattern cannot be larger than | |
767 | a 32-bit quantity. | |
768 | ||
68e1f29a | 769 | verify_fatal=bool Normally fio will keep checking the entire contents |
a12a3b4d JA |
770 | before quitting on a block verification failure. If this |
771 | option is set, fio will exit the job on the first observed | |
772 | failure. | |
160b966d | 773 | |
71bfa161 JA |
774 | stonewall Wait for preceeding jobs in the job file to exit, before |
775 | starting this one. Can be used to insert serialization | |
b3d62a75 JA |
776 | points in the job file. A stone wall also implies starting |
777 | a new reporting group. | |
778 | ||
779 | new_group Start a new reporting group. If this option isn't given, | |
780 | jobs in a file will be part of the same reporting group | |
bf9a3edb | 781 | unless separated by a stone wall (or if it's a group |
b3d62a75 | 782 | by itself, with the numjobs option). |
71bfa161 JA |
783 | |
784 | numjobs=int Create the specified number of clones of this job. May be | |
785 | used to setup a larger number of threads/processes doing | |
fa28c85a JA |
786 | the same thing. We regard that grouping of jobs as a |
787 | specific group. | |
788 | ||
789 | group_reporting If 'numjobs' is set, it may be interesting to display | |
790 | statistics for the group as a whole instead of for each | |
791 | individual job. This is especially true of 'numjobs' is | |
792 | large, looking at individual thread/process output quickly | |
793 | becomes unwieldy. If 'group_reporting' is specified, fio | |
794 | will show the final report per-group instead of per-job. | |
71bfa161 JA |
795 | |
796 | thread fio defaults to forking jobs, however if this option is | |
797 | given, fio will use pthread_create(3) to create threads | |
798 | instead. | |
799 | ||
800 | zonesize=siint Divide a file into zones of the specified size. See zoneskip. | |
801 | ||
802 | zoneskip=siint Skip the specified number of bytes when zonesize data has | |
803 | been read. The two zone options can be used to only do | |
804 | io on zones of a file. | |
805 | ||
076efc7c JA |
806 | write_iolog=str Write the issued io patterns to the specified file. See |
807 | read_iolog. | |
71bfa161 | 808 | |
076efc7c | 809 | read_iolog=str Open an iolog with the specified file name and replay the |
71bfa161 | 810 | io patterns it contains. This can be used to store a |
6df8adaa JA |
811 | workload and replay it sometime later. The iolog given |
812 | may also be a blktrace binary file, which allows fio | |
813 | to replay a workload captured by blktrace. See blktrace | |
814 | for how to capture such logging data. For blktrace replay, | |
815 | the file needs to be turned into a blkparse binary data | |
816 | file first (blktrace <device> -d file_for_fio.bin). | |
71bfa161 | 817 | |
e3cedca7 | 818 | write_bw_log=str If given, write a bandwidth log of the jobs in this job |
71bfa161 | 819 | file. Can be used to store data of the bandwidth of the |
e0da9bc2 JA |
820 | jobs in their lifetime. The included fio_generate_plots |
821 | script uses gnuplot to turn these text files into nice | |
e3cedca7 JA |
822 | graphs. See write_log_log for behaviour of given |
823 | filename. For this option, the postfix is _bw.log. | |
71bfa161 | 824 | |
e3cedca7 JA |
825 | write_lat_log=str Same as write_bw_log, except that this option stores io |
826 | completion latencies instead. If no filename is given | |
827 | with this option, the default filename of "jobname_type.log" | |
828 | is used. Even if the filename is given, fio will still | |
829 | append the type of log. So if one specifies | |
830 | ||
831 | write_lat_log=foo | |
832 | ||
833 | The actual log names will be foo_clat.log and foo_slat.log. | |
834 | This helps fio_generate_plot fine the logs automatically. | |
71bfa161 JA |
835 | |
836 | lockmem=siint Pin down the specified amount of memory with mlock(2). Can | |
837 | potentially be used instead of removing memory or booting | |
838 | with less memory to simulate a smaller amount of memory. | |
839 | ||
840 | exec_prerun=str Before running this job, issue the command specified | |
841 | through system(3). | |
842 | ||
843 | exec_postrun=str After the job completes, issue the command specified | |
844 | though system(3). | |
845 | ||
846 | ioscheduler=str Attempt to switch the device hosting the file to the specified | |
847 | io scheduler before running. | |
848 | ||
849 | cpuload=int If the job is a CPU cycle eater, attempt to use the specified | |
850 | percentage of CPU cycles. | |
851 | ||
852 | cpuchunks=int If the job is a CPU cycle eater, split the load into | |
6c219763 | 853 | cycles of the given time. In milliseconds. |
71bfa161 | 854 | |
0a839f30 JA |
855 | disk_util=bool Generate disk utilization statistics, if the platform |
856 | supports it. Defaults to on. | |
857 | ||
9520ebb9 JA |
858 | disable_clat=bool Disable measurements of completion latency numbers. Useful |
859 | only for cutting back the number of calls to gettimeofday, | |
860 | as that does impact performance at really high IOPS rates. | |
861 | Note that to really get rid of a large amount of these | |
862 | calls, this option must be used with disable_slat and | |
863 | disable_bw as well. | |
864 | ||
865 | disable_slat=bool Disable measurements of submission latency numbers. See | |
866 | disable_clat. | |
867 | ||
868 | disable_bw=bool Disable measurements of throughput/bandwidth numbers. See | |
869 | disable_clat. | |
870 | ||
993bf48b JA |
871 | gtod_reduce=bool Enable all of the gettimeofday() reducing options |
872 | (disable_clat, disable_slat, disable_bw) plus reduce | |
873 | precision of the timeout somewhat to really shrink | |
874 | the gettimeofday() call count. With this option enabled, | |
875 | we only do about 0.4% of the gtod() calls we would have | |
876 | done if all time keeping was enabled. | |
877 | ||
be4ecfdf JA |
878 | gtod_cpu=int Sometimes it's cheaper to dedicate a single thread of |
879 | execution to just getting the current time. Fio (and | |
880 | databases, for instance) are very intensive on gettimeofday() | |
881 | calls. With this option, you can set one CPU aside for | |
882 | doing nothing but logging current time to a shared memory | |
883 | location. Then the other threads/processes that run IO | |
884 | workloads need only copy that segment, instead of entering | |
885 | the kernel with a gettimeofday() call. The CPU set aside | |
886 | for doing these time calls will be excluded from other | |
887 | uses. Fio will manually clear it from the CPU mask of other | |
888 | jobs. | |
889 | ||
71bfa161 JA |
890 | |
891 | 6.0 Interpreting the output | |
892 | --------------------------- | |
893 | ||
894 | fio spits out a lot of output. While running, fio will display the | |
895 | status of the jobs created. An example of that would be: | |
896 | ||
73c8b082 | 897 | Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s] |
71bfa161 JA |
898 | |
899 | The characters inside the square brackets denote the current status of | |
900 | each thread. The possible values (in typical life cycle order) are: | |
901 | ||
902 | Idle Run | |
903 | ---- --- | |
904 | P Thread setup, but not started. | |
905 | C Thread created. | |
906 | I Thread initialized, waiting. | |
907 | R Running, doing sequential reads. | |
908 | r Running, doing random reads. | |
909 | W Running, doing sequential writes. | |
910 | w Running, doing random writes. | |
911 | M Running, doing mixed sequential reads/writes. | |
912 | m Running, doing mixed random reads/writes. | |
913 | F Running, currently waiting for fsync() | |
914 | V Running, doing verification of written data. | |
915 | E Thread exited, not reaped by main thread yet. | |
916 | _ Thread reaped. | |
917 | ||
918 | The other values are fairly self explanatory - number of threads | |
c9f60304 JA |
919 | currently running and doing io, rate of io since last check (read speed |
920 | listed first, then write speed), and the estimated completion percentage | |
921 | and time for the running group. It's impossible to estimate runtime of | |
922 | the following groups (if any). | |
71bfa161 JA |
923 | |
924 | When fio is done (or interrupted by ctrl-c), it will show the data for | |
925 | each thread, group of threads, and disks in that order. For each data | |
926 | direction, the output looks like: | |
927 | ||
928 | Client1 (g=0): err= 0: | |
929 | write: io= 32MiB, bw= 666KiB/s, runt= 50320msec | |
6104ddb6 JA |
930 | slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92 |
931 | clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82 | |
932 | bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68 | |
e7823a94 | 933 | cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17 |
71619dc2 | 934 | IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0% |
838bc709 JA |
935 | submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% |
936 | complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% | |
30061b97 | 937 | issued r/w: total=0/32768, short=0/0 |
8abdce66 JA |
938 | lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%, |
939 | lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0% | |
71bfa161 JA |
940 | |
941 | The client number is printed, along with the group id and error of that | |
942 | thread. Below is the io statistics, here for writes. In the order listed, | |
943 | they denote: | |
944 | ||
945 | io= Number of megabytes io performed | |
946 | bw= Average bandwidth rate | |
947 | runt= The runtime of that thread | |
72fbda2a | 948 | slat= Submission latency (avg being the average, stdev being the |
71bfa161 JA |
949 | standard deviation). This is the time it took to submit |
950 | the io. For sync io, the slat is really the completion | |
8a35c71e | 951 | latency, since queue/complete is one operation there. This |
bf9a3edb | 952 | value can be in milliseconds or microseconds, fio will choose |
8a35c71e | 953 | the most appropriate base and print that. In the example |
bf9a3edb | 954 | above, milliseconds is the best scale. |
71bfa161 JA |
955 | clat= Completion latency. Same names as slat, this denotes the |
956 | time from submission to completion of the io pieces. For | |
957 | sync io, clat will usually be equal (or very close) to 0, | |
958 | as the time from submit to complete is basically just | |
959 | CPU time (io has already been done, see slat explanation). | |
960 | bw= Bandwidth. Same names as the xlat stats, but also includes | |
961 | an approximate percentage of total aggregate bandwidth | |
962 | this thread received in this group. This last value is | |
963 | only really useful if the threads in this group are on the | |
964 | same disk, since they are then competing for disk access. | |
965 | cpu= CPU usage. User and system time, along with the number | |
e7823a94 JA |
966 | of context switches this thread went through, usage of |
967 | system and user time, and finally the number of major | |
968 | and minor page faults. | |
71619dc2 JA |
969 | IO depths= The distribution of io depths over the job life time. The |
970 | numbers are divided into powers of 2, so for example the | |
971 | 16= entries includes depths up to that value but higher | |
972 | than the previous entry. In other words, it covers the | |
973 | range from 16 to 31. | |
838bc709 JA |
974 | IO submit= How many pieces of IO were submitting in a single submit |
975 | call. Each entry denotes that amount and below, until | |
976 | the previous entry - eg, 8=100% mean that we submitted | |
977 | anywhere in between 5-8 ios per submit call. | |
978 | IO complete= Like the above submit number, but for completions instead. | |
30061b97 JA |
979 | IO issued= The number of read/write requests issued, and how many |
980 | of them were short. | |
ec118304 JA |
981 | IO latencies= The distribution of IO completion latencies. This is the |
982 | time from when IO leaves fio and when it gets completed. | |
983 | The numbers follow the same pattern as the IO depths, | |
984 | meaning that 2=1.6% means that 1.6% of the IO completed | |
8abdce66 JA |
985 | within 2 msecs, 20=12.8% means that 12.8% of the IO |
986 | took more than 10 msecs, but less than (or equal to) 20 msecs. | |
71bfa161 JA |
987 | |
988 | After each client has been listed, the group statistics are printed. They | |
989 | will look like this: | |
990 | ||
991 | Run status group 0 (all jobs): | |
992 | READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec | |
993 | WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec | |
994 | ||
995 | For each data direction, it prints: | |
996 | ||
997 | io= Number of megabytes io performed. | |
998 | aggrb= Aggregate bandwidth of threads in this group. | |
999 | minb= The minimum average bandwidth a thread saw. | |
1000 | maxb= The maximum average bandwidth a thread saw. | |
1001 | mint= The smallest runtime of the threads in that group. | |
1002 | maxt= The longest runtime of the threads in that group. | |
1003 | ||
1004 | And finally, the disk statistics are printed. They will look like this: | |
1005 | ||
1006 | Disk stats (read/write): | |
1007 | sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00% | |
1008 | ||
1009 | Each value is printed for both reads and writes, with reads first. The | |
1010 | numbers denote: | |
1011 | ||
1012 | ios= Number of ios performed by all groups. | |
1013 | merge= Number of merges io the io scheduler. | |
1014 | ticks= Number of ticks we kept the disk busy. | |
1015 | io_queue= Total time spent in the disk queue. | |
1016 | util= The disk utilization. A value of 100% means we kept the disk | |
1017 | busy constantly, 50% would be a disk idling half of the time. | |
1018 | ||
1019 | ||
1020 | 7.0 Terse output | |
1021 | ---------------- | |
1022 | ||
1023 | For scripted usage where you typically want to generate tables or graphs | |
6af019c9 | 1024 | of the results, fio can output the results in a semicolon separated format. |
71bfa161 JA |
1025 | The format is one long line of values, such as: |
1026 | ||
6af019c9 JA |
1027 | client1;0;0;1906777;1090804;1790;0;0;0.000000;0.000000;0;0;0.000000;0.000000;929380;1152890;25.510151%;1078276.333333;128948.113404;0;0;0;0;0;0.000000;0.000000;0;0;0.000000;0.000000;0;0;0.000000%;0.000000;0.000000;100.000000%;0.000000%;324;100.0%;0.0%;0.0%;0.0%;0.0%;0.0%;0.0%;100.0%;0.0%;0.0%;0.0%;0.0%;0.0% |
1028 | ;0.0%;0.0%;0.0%;0.0%;0.0% | |
71bfa161 | 1029 | |
6820cb3b JA |
1030 | To enable terse output, use the --minimal command line option. |
1031 | ||
71bfa161 JA |
1032 | Split up, the format is as follows: |
1033 | ||
1034 | jobname, groupid, error | |
1035 | READ status: | |
1036 | KiB IO, bandwidth (KiB/sec), runtime (msec) | |
1037 | Submission latency: min, max, mean, deviation | |
1038 | Completion latency: min, max, mean, deviation | |
6c219763 | 1039 | Bw: min, max, aggregate percentage of total, mean, deviation |
71bfa161 JA |
1040 | WRITE status: |
1041 | KiB IO, bandwidth (KiB/sec), runtime (msec) | |
1042 | Submission latency: min, max, mean, deviation | |
1043 | Completion latency: min, max, mean, deviation | |
6c219763 | 1044 | Bw: min, max, aggregate percentage of total, mean, deviation |
046ee302 | 1045 | CPU usage: user, system, context switches, major faults, minor faults |
2270890c JA |
1046 | IO depths: <=1, 2, 4, 8, 16, 32, >=64 |
1047 | IO latencies: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000 | |
1048 | Text description | |
71bfa161 | 1049 |