blob: bafad93fbdacc050d7c3b694556859af7b3cd9eb [file] [log] [blame]
Jens Axboe71bfa162006-10-25 11:08:19 +02001Table of contents
2-----------------
3
41. Overview
52. How fio works
63. Running fio
74. Job file format
85. Detailed list of parameters
96. Normal output
107. Terse output
Paul Dubs25c8b9d2011-07-21 17:26:02 +0200118. Trace file format
Bruce Cran43f09da2013-02-24 11:09:11 +0000129. CPU idleness profiling
Jens Axboe71bfa162006-10-25 11:08:19 +020013
141.0 Overview and history
15------------------------
16fio was originally written to save me the hassle of writing special test
17case programs when I wanted to test a specific workload, either for
18performance reasons or to find/reproduce a bug. The process of writing
19such a test app can be tiresome, especially if you have to do it often.
20Hence I needed a tool that would be able to simulate a given io workload
21without resorting to writing a tailored test case again and again.
22
23A test work load is difficult to define, though. There can be any number
24of processes or threads involved, and they can each be using their own
25way of generating io. You could have someone dirtying large amounts of
26memory in an memory mapped file, or maybe several threads issuing
27reads using asynchronous io. fio needed to be flexible enough to
28simulate both of these cases, and many more.
29
302.0 How fio works
31-----------------
32The first step in getting fio to simulate a desired io workload, is
33writing a job file describing that specific setup. A job file may contain
34any number of threads and/or files - the typical contents of the job file
35is a global section defining shared parameters, and one or more job
36sections describing the jobs involved. When run, fio parses this file
37and sets everything up as described. If we break down a job from top to
38bottom, it contains the following basic parameters:
39
40 IO type Defines the io pattern issued to the file(s).
41 We may only be reading sequentially from this
42 file(s), or we may be writing randomly. Or even
43 mixing reads and writes, sequentially or randomly.
44
45 Block size In how large chunks are we issuing io? This may be
46 a single value, or it may describe a range of
47 block sizes.
48
49 IO size How much data are we going to be reading/writing.
50
51 IO engine How do we issue io? We could be memory mapping the
52 file, we could be using regular read/write, we
Jens Axboed0ff85d2007-02-14 01:19:41 +010053 could be using splice, async io, syslet, or even
Jens Axboe71bfa162006-10-25 11:08:19 +020054 SG (SCSI generic sg).
55
Jens Axboe6c219762006-11-03 15:51:45 +010056 IO depth If the io engine is async, how large a queuing
Jens Axboe71bfa162006-10-25 11:08:19 +020057 depth do we want to maintain?
58
59 IO type Should we be doing buffered io, or direct/raw io?
60
61 Num files How many files are we spreading the workload over.
62
63 Num threads How many threads or processes should we spread
64 this workload over.
Bruce Cran66c098b2012-11-27 12:16:07 +000065
Jens Axboe71bfa162006-10-25 11:08:19 +020066The above are the basic parameters defined for a workload, in addition
67there's a multitude of parameters that modify other aspects of how this
68job behaves.
69
70
713.0 Running fio
72---------------
73See the README file for command line parameters, there are only a few
74of them.
75
76Running fio is normally the easiest part - you just give it the job file
77(or job files) as parameters:
78
79$ fio job_file
80
81and it will start doing what the job_file tells it to do. You can give
82more than one job file on the command line, fio will serialize the running
83of those files. Internally that is the same as using the 'stonewall'
84parameter described the the parameter section.
85
Jens Axboeb4692822006-10-27 13:43:22 +020086If the job file contains only one job, you may as well just give the
87parameters on the command line. The command line parameters are identical
88to the job parameters, with a few extra that control global parameters
89(see README). For example, for the job file parameter iodepth=2, the
Jens Axboec2b1e752006-10-30 09:03:13 +010090mirror command line option would be --iodepth 2 or --iodepth=2. You can
91also use the command line for giving more than one job entry. For each
92--name option that fio sees, it will start a new job with that name.
93Command line entries following a --name entry will apply to that job,
94until there are no more entries or a new --name entry is seen. This is
95similar to the job file options, where each option applies to the current
96job until a new [] job entry is seen.
Jens Axboeb4692822006-10-27 13:43:22 +020097
Jens Axboe71bfa162006-10-25 11:08:19 +020098fio does not need to run as root, except if the files or devices specified
99in the job section requires that. Some other options may also be restricted,
Jens Axboe6c219762006-11-03 15:51:45 +0100100such as memory locking, io scheduler switching, and decreasing the nice value.
Jens Axboe71bfa162006-10-25 11:08:19 +0200101
102
1034.0 Job file format
104-------------------
105As previously described, fio accepts one or more job files describing
106what it is supposed to do. The job file format is the classic ini file,
107where the names enclosed in [] brackets define the job name. You are free
108to use any ascii name you want, except 'global' which has special meaning.
109A global section sets defaults for the jobs described in that file. A job
110may override a global section parameter, and a job file may even have
111several global sections if so desired. A job is only affected by a global
Jens Axboe65db0852007-02-20 10:22:01 +0100112section residing above it. If the first character in a line is a ';' or a
113'#', the entire line is discarded as a comment.
Jens Axboe71bfa162006-10-25 11:08:19 +0200114
Aaron Carroll3c54bc42008-10-07 11:25:38 +0200115So let's look at a really simple job file that defines two processes, each
Jens Axboeb22989b2009-07-17 22:29:23 +0200116randomly reading from a 128MB file.
Jens Axboe71bfa162006-10-25 11:08:19 +0200117
118; -- start job file --
119[global]
120rw=randread
121size=128m
122
123[job1]
124
125[job2]
126
127; -- end job file --
128
129As you can see, the job file sections themselves are empty as all the
130described parameters are shared. As no filename= option is given, fio
Jens Axboec2b1e752006-10-30 09:03:13 +0100131makes up a filename for each of the jobs as it sees fit. On the command
132line, this job would look as follows:
133
134$ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
135
Jens Axboe71bfa162006-10-25 11:08:19 +0200136
Aaron Carroll3c54bc42008-10-07 11:25:38 +0200137Let's look at an example that has a number of processes writing randomly
Jens Axboe71bfa162006-10-25 11:08:19 +0200138to files.
139
140; -- start job file --
141[random-writers]
142ioengine=libaio
143iodepth=4
144rw=randwrite
145bs=32k
146direct=0
147size=64m
148numjobs=4
149
150; -- end job file --
151
152Here we have no global section, as we only have one job defined anyway.
153We want to use async io here, with a depth of 4 for each file. We also
Jens Axboeb22989b2009-07-17 22:29:23 +0200154increased the buffer size used to 32KB and define numjobs to 4 to
Jens Axboe71bfa162006-10-25 11:08:19 +0200155fork 4 identical jobs. The result is 4 processes each randomly writing
Jens Axboeb22989b2009-07-17 22:29:23 +0200156to their own 64MB file. Instead of using the above job file, you could
Jens Axboeb4692822006-10-27 13:43:22 +0200157have given the parameters on the command line. For this case, you would
158specify:
159
160$ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
Jens Axboe71bfa162006-10-25 11:08:19 +0200161
Jens Axboe74929ac2009-08-05 11:42:37 +02001624.1 Environment variables
163-------------------------
164
Aaron Carroll3c54bc42008-10-07 11:25:38 +0200165fio also supports environment variable expansion in job files. Any
166substring of the form "${VARNAME}" as part of an option value (in other
167words, on the right of the `='), will be expanded to the value of the
168environment variable called VARNAME. If no such environment variable
169is defined, or VARNAME is the empty string, the empty string will be
170substituted.
171
172As an example, let's look at a sample fio invocation and job file:
173
174$ SIZE=64m NUMJOBS=4 fio jobfile.fio
175
176; -- start job file --
177[random-writers]
178rw=randwrite
179size=${SIZE}
180numjobs=${NUMJOBS}
181; -- end job file --
182
183This will expand to the following equivalent job file at runtime:
184
185; -- start job file --
186[random-writers]
187rw=randwrite
188size=64m
189numjobs=4
190; -- end job file --
191
Jens Axboe71bfa162006-10-25 11:08:19 +0200192fio ships with a few example job files, you can also look there for
193inspiration.
194
Jens Axboe74929ac2009-08-05 11:42:37 +02001954.2 Reserved keywords
196---------------------
197
198Additionally, fio has a set of reserved keywords that will be replaced
199internally with the appropriate value. Those keywords are:
200
201$pagesize The architecture page size of the running system
202$mb_memory Megabytes of total memory in the system
203$ncpus Number of online available CPUs
204
205These can be used on the command line or in the job file, and will be
206automatically substituted with the current system values when the job
Jens Axboe892a6ff2009-11-13 12:19:49 +0100207is run. Simple math is also supported on these keywords, so you can
208perform actions like:
209
210size=8*$mb_memory
211
212and get that properly expanded to 8 times the size of memory in the
213machine.
Jens Axboe74929ac2009-08-05 11:42:37 +0200214
Jens Axboe71bfa162006-10-25 11:08:19 +0200215
2165.0 Detailed list of parameters
217-------------------------------
218
219This section describes in details each parameter associated with a job.
220Some parameters take an option of a given type, such as an integer or
221a string. The following types are used:
222
223str String. This is a sequence of alpha characters.
Jens Axboeb09da8f2009-07-17 23:16:17 +0200224time Integer with possible time suffix. In seconds unless otherwise
Jens Axboee417fd62008-09-11 09:27:15 +0200225 specified, use eg 10m for 10 minutes. Accepts s/m/h for seconds,
226 minutes, and hours.
Jens Axboeb09da8f2009-07-17 23:16:17 +0200227int SI integer. A whole number value, which may contain a suffix
228 describing the base of the number. Accepted suffixes are k/m/g/t/p,
229 meaning kilo, mega, giga, tera, and peta. The suffix is not case
Jens Axboe57fc29f2010-06-23 22:24:07 +0200230 sensitive, and you may also include trailing 'b' (eg 'kb' is the same
231 as 'k'). So if you want to specify 4096, you could either write
Jens Axboeb09da8f2009-07-17 23:16:17 +0200232 out '4096' or just give 4k. The suffixes signify base 2 values, so
Jens Axboe57fc29f2010-06-23 22:24:07 +0200233 1024 is 1k and 1024k is 1m and so on, unless the suffix is explicitly
234 set to a base 10 value using 'kib', 'mib', 'gib', etc. If that is the
235 case, then 1000 is used as the multiplier. This can be handy for
236 disks, since manufacturers generally use base 10 values when listing
237 the capacity of a drive. If the option accepts an upper and lower
238 range, use a colon ':' or minus '-' to separate such values. May also
239 include a prefix to indicate numbers base. If 0x is used, the number
240 is assumed to be hexadecimal. See irange.
Jens Axboe71bfa162006-10-25 11:08:19 +0200241bool Boolean. Usually parsed as an integer, however only defined for
242 true and false (1 and 0).
Jens Axboeb09da8f2009-07-17 23:16:17 +0200243irange Integer range with suffix. Allows value range to be given, such
Jens Axboebf9a3ed2008-06-05 11:53:08 +0200244 as 1024-4096. A colon may also be used as the separator, eg
Jens Axboe0c9baf92007-01-11 15:59:26 +0100245 1k:4k. If the option allows two sets of ranges, they can be
246 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
Jens Axboef7fa2652009-03-09 14:20:20 +0100247 int.
Yu-ju Hong83349192011-08-13 00:53:44 +0200248float_list A list of floating numbers, separated by a ':' character.
Jens Axboe71bfa162006-10-25 11:08:19 +0200249
250With the above in mind, here follows the complete list of fio job
251parameters.
252
253name=str ASCII name of the job. This may be used to override the
254 name printed by fio for this job. Otherwise the job
Jens Axboec2b1e752006-10-30 09:03:13 +0100255 name is used. On the command line this parameter has the
Jens Axboe6c219762006-11-03 15:51:45 +0100256 special purpose of also signaling the start of a new
Jens Axboec2b1e752006-10-30 09:03:13 +0100257 job.
Jens Axboe71bfa162006-10-25 11:08:19 +0200258
Jens Axboe61697c32007-02-05 15:04:46 +0100259description=str Text description of the job. Doesn't do anything except
260 dump this text description when this job is run. It's
261 not parsed.
262
Randy Dunlap37760412009-05-13 07:51:05 +0200263directory=str Prefix filenames with this directory. Used to place files
Jens Axboe71bfa162006-10-25 11:08:19 +0200264 in a different location than "./".
265
266filename=str Fio normally makes up a filename based on the job name,
267 thread number, and file number. If you want to share
268 files between threads in a job or several jobs, specify
Jens Axboeed92ac02007-02-06 14:43:52 +0100269 a filename for each of them to override the default. If
Jens Axboe414c2a32009-01-16 13:21:15 +0100270 the ioengine used is 'net', the filename is the host, port,
Jens Axboe0fd666b2011-10-06 20:08:53 +0200271 and protocol to use in the format of =host,port,protocol.
Jens Axboe414c2a32009-01-16 13:21:15 +0100272 See ioengine=net for more. If the ioengine is file based, you
273 can specify a number of files by separating the names with a
274 ':' colon. So if you wanted a job to open /dev/sda and /dev/sdb
275 as the two working files, you would use
Jens Axboe30a45882013-01-30 12:53:55 +0100276 filename=/dev/sda:/dev/sdb. On Windows, disk devices are
277 accessed as \\.\PhysicalDrive0 for the first device,
278 \\.\PhysicalDrive1 for the second etc. Note: Windows and
279 FreeBSD prevent write access to areas of the disk containing
280 in-use data (e.g. filesystems).
281 If the wanted filename does need to include a colon, then
282 escape that with a '\' character. For instance, if the filename
283 is "/dev/dsk/foo@3,0:c", then you would use
284 filename="/dev/dsk/foo@3,0\:c". '-' is a reserved name, meaning
285 stdin or stdout. Which of the two depends on the read/write
286 direction set.
Jens Axboe71bfa162006-10-25 11:08:19 +0200287
Jens Axboede98bd32013-04-05 11:09:20 +0200288filename_format=str
289 If sharing multiple files between jobs, it is usually necessary
290 to have fio generate the exact names that you want. By default,
291 fio will name a file based on the default file format
292 specification of jobname.jobnumber.filenumber. With this
293 option, that can be customized. Fio will recognize and replace
294 the following keywords in this string:
295
296 $jobname
297 The name of the worker thread or process.
298
299 $jobnum
300 The incremental number of the worker thread or
301 process.
302
303 $filenum
304 The incremental number of the file for that worker
305 thread or process.
306
307 To have dependent jobs share a set of files, this option can
308 be set to have fio generate filenames that are shared between
309 the two. For instance, if testfiles.$filenum is specified,
310 file number 4 for any job will be named testfiles.4. The
311 default of $jobname.$jobnum.$filenum will be used if
312 no other format specifier is given.
313
Jens Axboebbf6b542007-03-13 15:28:55 +0100314opendir=str Tell fio to recursively add any file it can find in this
315 directory and down the file system tree.
316
Randy Dunlap37760412009-05-13 07:51:05 +0200317lockfile=str Fio defaults to not locking any files before it does
Jens Axboe4d4e80f2008-03-04 10:18:56 +0100318 IO to them. If a file or file descriptor is shared, fio
319 can serialize IO to that file to make the end result
320 consistent. This is usual for emulating real workloads that
321 share files. The lock modes are:
Jens Axboe29c13492008-03-01 19:25:20 +0100322
Jens Axboe4d4e80f2008-03-04 10:18:56 +0100323 none No locking. The default.
324 exclusive Only one thread/process may do IO,
325 excluding all others.
326 readwrite Read-write locking on the file. Many
327 readers may access the file at the
328 same time, but writes get exclusive
329 access.
330
Jens Axboed3aad8f2007-03-15 14:12:05 +0100331readwrite=str
Jens Axboe71bfa162006-10-25 11:08:19 +0200332rw=str Type of io pattern. Accepted values are:
333
334 read Sequential reads
335 write Sequential writes
336 randwrite Random writes
337 randread Random reads
Jens Axboe10b023d2012-03-23 13:40:06 +0100338 rw,readwrite Sequential mixed reads and writes
Jens Axboe71bfa162006-10-25 11:08:19 +0200339 randrw Random mixed reads and writes
340
341 For the mixed io types, the default is to split them 50/50.
342 For certain types of io the result may still be skewed a bit,
Jens Axboe211097b2007-03-22 18:56:45 +0100343 since the speed may be different. It is possible to specify
Jens Axboe38dad622010-07-20 14:46:00 -0600344 a number of IO's to do before getting a new offset, this is
345 one by appending a ':<nr>' to the end of the string given.
346 For a random read, it would look like 'rw=randread:8' for
Jens Axboe059b0802011-08-25 09:09:37 +0200347 passing in an offset modifier with a value of 8. If the
Lucian Adrian Grijincuddb754d2012-04-05 18:18:35 -0600348 suffix is used with a sequential IO pattern, then the value
Jens Axboe059b0802011-08-25 09:09:37 +0200349 specified will be added to the generated offset for each IO.
350 For instance, using rw=write:4k will skip 4k for every
351 write. It turns sequential IO into sequential IO with holes.
352 See the 'rw_sequencer' option.
Jens Axboe38dad622010-07-20 14:46:00 -0600353
354rw_sequencer=str If an offset modifier is given by appending a number to
355 the rw=<str> line, then this option controls how that
356 number modifies the IO offset being generated. Accepted
357 values are:
358
359 sequential Generate sequential offset
360 identical Generate the same offset
361
362 'sequential' is only useful for random IO, where fio would
363 normally generate a new random offset for every IO. If you
364 append eg 8 to randread, you would get a new random offset for
Jens Axboe211097b2007-03-22 18:56:45 +0100365 every 8 IO's. The result would be a seek for only every 8
366 IO's, instead of for every IO. Use rw=randread:8 to specify
Jens Axboe38dad622010-07-20 14:46:00 -0600367 that. As sequential IO is already sequential, setting
368 'sequential' for that would not result in any differences.
369 'identical' behaves in a similar fashion, except it sends
370 the same offset 8 number of times before generating a new
371 offset.
Jens Axboe71bfa162006-10-25 11:08:19 +0200372
Jens Axboe90fef2d2009-07-17 22:33:32 +0200373kb_base=int The base unit for a kilobyte. The defacto base is 2^10, 1024.
374 Storage manufacturers like to use 10^3 or 1000 as a base
375 ten unit instead, for obvious reasons. Allow values are
376 1024 or 1000, with 1024 being the default.
377
Jens Axboe771e58b2013-01-30 12:56:23 +0100378unified_rw_reporting=bool Fio normally reports statistics on a per
379 data direction basis, meaning that read, write, and trim are
380 accounted and reported separately. If this option is set,
381 the fio will sum the results and report them as "mixed"
382 instead.
383
Jens Axboeee738492007-01-10 11:23:16 +0100384randrepeat=bool For random IO workloads, seed the generator in a predictable
385 way so that results are repeatable across repetitions.
386
Jens Axboe04778ba2014-01-10 20:57:01 -0700387randseed=int Seed the random number generators based on this seed value, to
388 be able to control what sequence of output is being generated.
389 If not set, the random sequence depends on the randrepeat
390 setting.
391
Jens Axboe2615cc42011-03-28 09:35:09 +0200392use_os_rand=bool Fio can either use the random generator supplied by the OS
393 to generator random offsets, or it can use it's own internal
394 generator (based on Tausworthe). Default is to use the
395 internal generator, which is often of better quality and
396 faster.
397
Eric Gourioua596f042011-06-17 09:11:45 +0200398fallocate=str Whether pre-allocation is performed when laying down files.
399 Accepted values are:
400
401 none Do not pre-allocate space
402 posix Pre-allocate via posix_fallocate()
403 keep Pre-allocate via fallocate() with
404 FALLOC_FL_KEEP_SIZE set
405 0 Backward-compatible alias for 'none'
406 1 Backward-compatible alias for 'posix'
407
408 May not be available on all supported platforms. 'keep' is only
409 available on Linux.If using ZFS on Solaris this must be set to
410 'none' because ZFS doesn't support it. Default: 'posix'.
Jens Axboe7bc8c2c2010-01-28 11:31:31 +0100411
Jens Axboed2f3ac32007-03-22 19:24:09 +0100412fadvise_hint=bool By default, fio will use fadvise() to advise the kernel
413 on what IO patterns it is likely to issue. Sometimes you
414 want to test specific IO patterns without telling the
415 kernel about it, in which case you can disable this option.
416 If set, fio will use POSIX_FADV_SEQUENTIAL for sequential
417 IO and POSIX_FADV_RANDOM for random IO.
418
Jens Axboef7fa2652009-03-09 14:20:20 +0100419size=int The total size of file io for this job. Fio will run until
Jens Axboe7616caf2007-05-25 09:26:05 +0200420 this many bytes has been transferred, unless runtime is
421 limited by other options (such as 'runtime', for instance).
Randy Dunlap37760412009-05-13 07:51:05 +0200422 Unless specific nrfiles and filesize options are given,
Jens Axboe7616caf2007-05-25 09:26:05 +0200423 fio will divide this size between the available files
Jens Axboed6667262010-06-25 11:32:48 +0200424 specified by the job. If not set, fio will use the full
425 size of the given files or devices. If the the files
Jens Axboe7bb59102011-07-12 19:47:03 +0200426 do not exist, size must be given. It is also possible to
427 give size as a percentage between 1 and 100. If size=20%
428 is given, fio will use 20% of the full size of the given
429 files or devices.
Jens Axboe71bfa162006-10-25 11:08:19 +0200430
Jens Axboef7fa2652009-03-09 14:20:20 +0100431filesize=int Individual file sizes. May be a range, in which case fio
Jens Axboe9c60ce62007-03-15 09:14:47 +0100432 will select sizes for files at random within the given range
433 and limited to 'size' in total (if that is given). If not
434 given, each created file is the same size.
435
Jens Axboe74586c12011-01-20 10:16:03 -0700436fill_device=bool
437fill_fs=bool Sets size to something really large and waits for ENOSPC (no
Shawn Lewisaa31f1f2008-01-11 09:45:11 +0100438 space left on device) as the terminating condition. Only makes
Jens Axboede98bd32013-04-05 11:09:20 +0200439 sense with sequential write. For a read workload, the mount
Jens Axboe4f124322011-01-19 15:35:26 -0700440 point will be filled first then IO started on the result. This
441 option doesn't make sense if operating on a raw device node,
442 since the size of that is already known by the file system.
443 Additionally, writing beyond end-of-device will not return
444 ENOSPC there.
Shawn Lewisaa31f1f2008-01-11 09:45:11 +0100445
Jens Axboef7fa2652009-03-09 14:20:20 +0100446blocksize=int
447bs=int The block size used for the io units. Defaults to 4k. Values
448 can be given for both read and writes. If a single int is
449 given, it will apply to both. If a second int is specified
Jens Axboef90eff52006-11-06 11:08:21 +0100450 after a comma, it will apply to writes only. In other words,
Jens Axboed9472272013-07-25 10:20:45 -0600451 the format is either bs=read_and_write or bs=read,write,trim.
452 bs=4k,8k will thus use 4k blocks for reads, 8k blocks for
453 writes, and 8k for trims. You can terminate the list with
454 a trailing comma. bs=4k,8k, would use the default value for
455 trims.. If you only wish to set the write size, you
Jens Axboe787f7e92006-11-06 13:26:29 +0100456 can do so by passing an empty read size - bs=,8k will set
457 8k for writes and leave the read default value.
Jens Axboea00735e2006-11-03 08:58:08 +0100458
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100459blockalign=int
460ba=int At what boundary to align random IO offsets. Defaults to
461 the same as 'blocksize' the minimum blocksize given.
462 Minimum alignment is typically 512b for using direct IO,
463 though it usually depends on the hardware block size. This
464 option is mutually exclusive with using a random map for
465 files, so it will turn off that option.
466
Jens Axboed3aad8f2007-03-15 14:12:05 +0100467blocksize_range=irange
Jens Axboe71bfa162006-10-25 11:08:19 +0200468bsrange=irange Instead of giving a single block size, specify a range
469 and fio will mix the issued io block sizes. The issued
470 io unit will always be a multiple of the minimum value
Jens Axboef90eff52006-11-06 11:08:21 +0100471 given (also see bs_unaligned). Applies to both reads and
472 writes, however a second range can be given after a comma.
473 See bs=.
Jens Axboea00735e2006-11-03 08:58:08 +0100474
Jens Axboe564ca972007-12-14 12:21:19 +0100475bssplit=str Sometimes you want even finer grained control of the
476 block sizes issued, not just an even split between them.
477 This option allows you to weight various block sizes,
478 so that you are able to define a specific amount of
479 block sizes issued. The format for this option is:
480
481 bssplit=blocksize/percentage:blocksize/percentage
482
483 for as many block sizes as needed. So if you want to define
484 a workload that has 50% 64k blocks, 10% 4k blocks, and
485 40% 32k blocks, you would write:
486
487 bssplit=4k/10:64k/50:32k/40
488
489 Ordering does not matter. If the percentage is left blank,
490 fio will fill in the remaining values evenly. So a bssplit
491 option like this one:
492
493 bssplit=4k/50:1k/:32k/
494
495 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages
496 always add up to 100, if bssplit is given a range that adds
497 up to more, it will error out.
498
Jens Axboe720e84a2009-04-21 08:29:55 +0200499 bssplit also supports giving separate splits to reads and
500 writes. The format is identical to what bs= accepts. You
501 have to separate the read and write parts with a comma. So
502 if you want a workload that has 50% 2k reads and 50% 4k reads,
503 while having 90% 4k writes and 10% 8k writes, you would
504 specify:
505
506 bssplit=2k/50:4k/50,4k/90,8k/10
507
Jens Axboed3aad8f2007-03-15 14:12:05 +0100508blocksize_unaligned
Jens Axboe690adba2006-10-30 15:25:09 +0100509bs_unaligned If this option is given, any byte size value within bsrange
510 may be used as a block range. This typically wont work with
511 direct IO, as that normally requires sector alignment.
Jens Axboe71bfa162006-10-25 11:08:19 +0200512
Jens Axboe6aca9b32013-07-25 12:45:26 -0600513bs_is_seq_rand If this option is set, fio will use the normal read,write
514 blocksize settings as sequential,random instead. Any random
515 read or write will use the WRITE blocksize settings, and any
516 sequential read or write will use the READ blocksize setting.
517
Jens Axboee9459e52007-04-17 15:46:32 +0200518zero_buffers If this option is given, fio will init the IO buffers to
519 all zeroes. The default is to fill them with random data.
520
Jens Axboe5973caf2008-05-21 19:52:35 +0200521refill_buffers If this option is given, fio will refill the IO buffers
522 on every submit. The default is to only fill it at init
523 time and reuse that data. Only makes sense if zero_buffers
Jens Axboe41ccd842008-05-22 09:17:33 +0200524 isn't specified, naturally. If data verification is enabled,
525 refill_buffers is also automatically enabled.
Jens Axboe5973caf2008-05-21 19:52:35 +0200526
Jens Axboefd684182011-09-19 09:24:44 +0200527scramble_buffers=bool If refill_buffers is too costly and the target is
528 using data deduplication, then setting this option will
529 slightly modify the IO buffer contents to defeat normal
530 de-dupe attempts. This is not enough to defeat more clever
531 block compression attempts, but it will stop naive dedupe of
532 blocks. Default: true.
533
Jens Axboec5751c62012-03-15 15:02:56 +0100534buffer_compress_percentage=int If this is set, then fio will attempt to
535 provide IO buffer content (on WRITEs) that compress to
536 the specified level. Fio does this by providing a mix of
537 random data and zeroes. Note that this is per block size
538 unit, for file/disk wide compression level that matches
539 this setting, you'll also want to set refill_buffers.
540
541buffer_compress_chunk=int See buffer_compress_percentage. This
542 setting allows fio to manage how big the ranges of random
543 data and zeroed data is. Without this set, fio will
544 provide buffer_compress_percentage of blocksize random
545 data, followed by the remaining zeroed. With this set
546 to some chunk size smaller than the block size, fio can
547 alternate random and zeroed data throughout the IO
548 buffer.
549
Jens Axboece35b1e2014-01-14 15:35:58 -0700550buffer_pattern=str If set, fio will fill the io buffers with this pattern.
551 If not set, the contents of io buffers is defined by the other
552 options related to buffer contents. The setting can be any
553 pattern of bytes, and can be prefixed with 0x for hex values.
554
Jens Axboe71bfa162006-10-25 11:08:19 +0200555nrfiles=int Number of files to use for this job. Defaults to 1.
556
Jens Axboe390b1532007-03-09 13:03:00 +0100557openfiles=int Number of files to keep open at the same time. Defaults to
558 the same as nrfiles, can be set smaller to limit the number
559 simultaneous opens.
560
Jens Axboe5af1c6f2007-03-01 10:06:10 +0100561file_service_type=str Defines how fio decides which file from a job to
562 service next. The following types are defined:
563
564 random Just choose a file at random.
565
566 roundrobin Round robin over open files. This
567 is the default.
568
Jens Axboea086c252009-03-04 08:27:37 +0100569 sequential Finish one file before moving on to
570 the next. Multiple files can still be
571 open depending on 'openfiles'.
572
Jens Axboe1907dbc2007-03-12 11:44:28 +0100573 The string can have a number appended, indicating how
574 often to switch to a new file. So if option random:4 is
575 given, fio will switch to a new random file after 4 ios
576 have been issued.
577
Jens Axboe71bfa162006-10-25 11:08:19 +0200578ioengine=str Defines how the job issues io to the file. The following
579 types are defined:
580
581 sync Basic read(2) or write(2) io. lseek(2) is
582 used to position the io location.
583
gurudas paia31041e2007-10-23 15:12:30 +0200584 psync Basic pread(2) or pwrite(2) io.
585
Gurudas Paie05af9e2008-02-06 11:16:15 +0100586 vsync Basic readv(2) or writev(2) IO.
Jens Axboe1d2af022008-02-04 10:59:07 +0100587
Jens Axboea46c5e02013-05-16 20:38:09 +0200588 psyncv Basic preadv(2) or pwritev(2) IO.
589
Jens Axboe15d182a2009-01-16 19:15:07 +0100590 libaio Linux native asynchronous io. Note that Linux
591 may only support queued behaviour with
592 non-buffered IO (set direct=1 or buffered=0).
Steven Langde890a12011-11-09 14:03:34 +0100593 This engine defines engine specific options.
Jens Axboe71bfa162006-10-25 11:08:19 +0200594
595 posixaio glibc posix asynchronous io.
596
Jens Axboe417f0062008-06-02 11:59:30 +0200597 solarisaio Solaris native asynchronous io.
598
Bruce Cran03e20d62011-01-02 20:14:54 +0100599 windowsaio Windows native asynchronous io.
600
Jens Axboe71bfa162006-10-25 11:08:19 +0200601 mmap File is memory mapped and data copied
602 to/from using memcpy(3).
603
604 splice splice(2) is used to transfer the data and
605 vmsplice(2) to transfer data from user
606 space to the kernel.
607
Jens Axboed0ff85d2007-02-14 01:19:41 +0100608 syslet-rw Use the syslet system calls to make
609 regular read/write async.
610
Jens Axboe71bfa162006-10-25 11:08:19 +0200611 sg SCSI generic sg v3 io. May either be
Jens Axboe6c219762006-11-03 15:51:45 +0100612 synchronous using the SG_IO ioctl, or if
Jens Axboe71bfa162006-10-25 11:08:19 +0200613 the target is an sg character device
614 we use read(2) and write(2) for asynchronous
615 io.
616
Jens Axboea94ea282006-11-24 12:37:34 +0100617 null Doesn't transfer any data, just pretends
618 to. This is mainly used to exercise fio
619 itself and for debugging/testing purposes.
620
Jens Axboeed92ac02007-02-06 14:43:52 +0100621 net Transfer over the network to given host:port.
Steven Langde890a12011-11-09 14:03:34 +0100622 Depending on the protocol used, the hostname,
623 port, listen and filename options are used to
624 specify what sort of connection to make, while
625 the protocol option determines which protocol
626 will be used.
627 This engine defines engine specific options.
Jens Axboeed92ac02007-02-06 14:43:52 +0100628
Jens Axboe9cce02e2007-06-22 15:42:21 +0200629 netsplice Like net, but uses splice/vmsplice to
630 map data and send/receive.
Steven Langde890a12011-11-09 14:03:34 +0100631 This engine defines engine specific options.
Jens Axboe9cce02e2007-06-22 15:42:21 +0200632
gurudas pai53aec0a2007-10-05 13:20:18 +0200633 cpuio Doesn't transfer any data, but burns CPU
Jens Axboeba0fbe12007-03-09 14:34:23 +0100634 cycles according to the cpuload= and
635 cpucycle= options. Setting cpuload=85
636 will cause that job to do nothing but burn
Gurudas Pai36ecec82008-02-08 08:50:14 +0100637 85% of the CPU. In case of SMP machines,
638 use numjobs=<no_of_cpu> to get desired CPU
639 usage, as the cpuload only loads a single
640 CPU at the desired rate.
Jens Axboeba0fbe12007-03-09 14:34:23 +0100641
Jens Axboee9a18062007-03-21 08:51:56 +0100642 guasi The GUASI IO engine is the Generic Userspace
643 Asyncronous Syscall Interface approach
644 to async IO. See
645
646 http://www.xmailserver.org/guasi-lib.html
647
648 for more info on GUASI.
649
ren yufei21b8aee2011-08-01 10:01:57 +0200650 rdma The RDMA I/O engine supports both RDMA
Bart Van Asscheeb52fa32011-08-15 09:01:05 +0200651 memory semantics (RDMA_WRITE/RDMA_READ) and
652 channel semantics (Send/Recv) for the
653 InfiniBand, RoCE and iWARP protocols.
ren yufei21b8aee2011-08-01 10:01:57 +0200654
Dmitry Monakhovd54fce82012-09-20 15:37:17 +0400655 falloc IO engine that does regular fallocate to
656 simulate data transfer as fio ioengine.
657 DDIR_READ does fallocate(,mode = keep_size,)
Jens Axboe0981fd72012-09-20 19:23:02 +0200658 DDIR_WRITE does fallocate(,mode = 0)
Dmitry Monakhovd54fce82012-09-20 15:37:17 +0400659 DDIR_TRIM does fallocate(,mode = punch_hole)
660
661 e4defrag IO engine that does regular EXT4_IOC_MOVE_EXT
662 ioctls to simulate defragment activity in
663 request to DDIR_WRITE event
Jens Axboe0981fd72012-09-20 19:23:02 +0200664
Jens Axboe8a7bd872007-02-28 11:12:25 +0100665 external Prefix to specify loading an external
666 IO engine object file. Append the engine
667 filename, eg ioengine=external:/tmp/foo.o
668 to load ioengine foo.o in /tmp.
669
Jens Axboe71bfa162006-10-25 11:08:19 +0200670iodepth=int This defines how many io units to keep in flight against
671 the file. The default is 1 for each file defined in this
672 job, can be overridden with a larger value for higher
Jens Axboeee72ca02010-12-02 20:05:37 +0100673 concurrency. Note that increasing iodepth beyond 1 will not
674 affect synchronous ioengines (except for small degress when
Bruce Cran9b836562011-01-08 19:49:54 +0100675 verify_async is in use). Even async engines may impose OS
Jens Axboeee72ca02010-12-02 20:05:37 +0100676 restrictions causing the desired depth not to be achieved.
677 This may happen on Linux when using libaio and not setting
678 direct=1, since buffered IO is not async on that OS. Keep an
679 eye on the IO depth distribution in the fio output to verify
680 that the achieved depth is as expected. Default: 1.
Jens Axboe71bfa162006-10-25 11:08:19 +0200681
Jens Axboe49504212008-06-05 09:03:30 +0200682iodepth_batch_submit=int
Jens Axboecb5ab512007-02-26 12:57:09 +0100683iodepth_batch=int This defines how many pieces of IO to submit at once.
Jens Axboe89e820f2008-01-18 10:30:07 +0100684 It defaults to 1 which means that we submit each IO
685 as soon as it is available, but can be raised to submit
686 bigger batches of IO at the time.
Jens Axboecb5ab512007-02-26 12:57:09 +0100687
Jens Axboe49504212008-06-05 09:03:30 +0200688iodepth_batch_complete=int This defines how many pieces of IO to retrieve
689 at once. It defaults to 1 which means that we'll ask
690 for a minimum of 1 IO in the retrieval process from
691 the kernel. The IO retrieval will go on until we
692 hit the limit set by iodepth_low. If this variable is
693 set to 0, then fio will always check for completed
694 events before queuing more IO. This helps reduce
695 IO latency, at the cost of more retrieval system calls.
696
Jens Axboee916b392007-02-20 14:37:26 +0100697iodepth_low=int The low water mark indicating when to start filling
698 the queue again. Defaults to the same as iodepth, meaning
699 that fio will attempt to keep the queue full at all times.
700 If iodepth is set to eg 16 and iodepth_low is set to 4, then
701 after fio has filled the queue of 16 requests, it will let
702 the depth drain down to 4 before starting to fill it again.
703
Jens Axboe71bfa162006-10-25 11:08:19 +0200704direct=bool If value is true, use non-buffered io. This is usually
Bruce Cran9b836562011-01-08 19:49:54 +0100705 O_DIRECT. Note that ZFS on Solaris doesn't support direct io.
Bruce Cran93bcfd22012-02-20 20:18:19 +0100706 On Windows the synchronous ioengines don't support direct io.
Jens Axboe76a43db2007-01-11 13:24:44 +0100707
Chris Masond01612f2013-11-15 15:52:58 -0700708atomic=bool If value is true, attempt to use atomic direct IO. Atomic
709 writes are guaranteed to be stable once acknowledged by
710 the operating system. Only Linux supports O_ATOMIC right
711 now.
712
Jens Axboe76a43db2007-01-11 13:24:44 +0100713buffered=bool If value is true, use buffered io. This is the opposite
714 of the 'direct' option. Defaults to true.
Jens Axboe71bfa162006-10-25 11:08:19 +0200715
Jens Axboef7fa2652009-03-09 14:20:20 +0100716offset=int Start io at the given offset in the file. The data before
Jens Axboe71bfa162006-10-25 11:08:19 +0200717 the given offset will not be touched. This effectively
718 caps the file size at real_size - offset.
719
Dan Ehrenberg214ac7e2012-03-15 14:44:26 +0100720offset_increment=int If this is provided, then the real offset becomes
721 the offset + offset_increment * thread_number, where the
722 thread number is a counter that starts at 0 and is incremented
723 for each job. This option is useful if there are several jobs
724 which are intended to operate on a file in parallel in disjoint
725 segments, with even spacing between the starting points.
726
Jens Axboeddf24e42013-08-09 12:53:44 -0600727number_ios=int Fio will normally perform IOs until it has exhausted the size
728 of the region set by size=, or if it exhaust the allocated
729 time (or hits an error condition). With this setting, the
730 range/size can be set independently of the number of IOs to
731 perform. When fio reaches this number, it will exit normally
732 and report status.
733
Jens Axboe71bfa162006-10-25 11:08:19 +0200734fsync=int If writing to a file, issue a sync of the dirty data
735 for every number of blocks given. For example, if you give
736 32 as a parameter, fio will sync the file for every 32
737 writes issued. If fio is using non-buffered io, we may
738 not sync the file. The exception is the sg io engine, which
Jens Axboe6c219762006-11-03 15:51:45 +0100739 synchronizes the disk cache anyway.
Jens Axboe71bfa162006-10-25 11:08:19 +0200740
Jens Axboee76b1da2010-03-09 20:49:54 +0100741fdatasync=int Like fsync= but uses fdatasync() to only sync data and not
Jens Axboe5f9099e2009-06-16 22:40:26 +0200742 metadata blocks.
Bruce Cran93bcfd22012-02-20 20:18:19 +0100743 In FreeBSD and Windows there is no fdatasync(), this falls back to
Joshua Aunee72fa4d2010-02-11 00:59:18 -0700744 using fsync()
Jens Axboe5f9099e2009-06-16 22:40:26 +0200745
Jens Axboee76b1da2010-03-09 20:49:54 +0100746sync_file_range=str:val Use sync_file_range() for every 'val' number of
747 write operations. Fio will track range of writes that
748 have happened since the last sync_file_range() call. 'str'
749 can currently be one or more of:
750
751 wait_before SYNC_FILE_RANGE_WAIT_BEFORE
752 write SYNC_FILE_RANGE_WRITE
753 wait_after SYNC_FILE_RANGE_WAIT_AFTER
754
755 So if you do sync_file_range=wait_before,write:8, fio would
756 use SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE for
757 every 8 writes. Also see the sync_file_range(2) man page.
758 This option is Linux specific.
759
Jens Axboe5036fc12008-04-15 09:20:46 +0200760overwrite=bool If true, writes to a file will always overwrite existing
761 data. If the file doesn't already exist, it will be
762 created before the write phase begins. If the file exists
763 and is large enough for the specified write phase, nothing
764 will be done.
Jens Axboe71bfa162006-10-25 11:08:19 +0200765
Jens Axboedbd11ea2013-01-13 17:16:46 +0100766end_fsync=bool If true, fsync file contents when a write stage has completed.
Jens Axboe71bfa162006-10-25 11:08:19 +0200767
Jens Axboeebb14152007-03-13 14:42:15 +0100768fsync_on_close=bool If true, fio will fsync() a dirty file on close.
769 This differs from end_fsync in that it will happen on every
770 file close, not just at the end of the job.
771
Jens Axboe71bfa162006-10-25 11:08:19 +0200772rwmixread=int How large a percentage of the mix should be reads.
773
774rwmixwrite=int How large a percentage of the mix should be writes. If both
775 rwmixread and rwmixwrite is given and the values do not add
776 up to 100%, the latter of the two will be used to override
Jens Axboec35dd7a2009-06-10 08:39:16 +0200777 the first. This may interfere with a given rate setting,
778 if fio is asked to limit reads or writes to a certain rate.
779 If that is the case, then the distribution may be skewed.
Jens Axboe71bfa162006-10-25 11:08:19 +0200780
Jens Axboe92d42d62012-11-15 15:38:32 -0700781random_distribution=str:float By default, fio will use a completely uniform
782 random distribution when asked to perform random IO. Sometimes
783 it is useful to skew the distribution in specific ways,
784 ensuring that some parts of the data is more hot than others.
785 fio includes the following distribution models:
786
787 random Uniform random distribution
788 zipf Zipf distribution
789 pareto Pareto distribution
790
791 When using a zipf or pareto distribution, an input value
792 is also needed to define the access pattern. For zipf, this
793 is the zipf theta. For pareto, it's the pareto power. Fio
794 includes a test program, genzipf, that can be used visualize
795 what the given input values will yield in terms of hit rates.
796 If you wanted to use zipf with a theta of 1.2, you would use
797 random_distribution=zipf:1.2 as the option. If a non-uniform
798 model is used, fio will disable use of the random map.
799
Jens Axboe211c9b82013-04-26 08:56:17 -0600800percentage_random=int For a random workload, set how big a percentage should
801 be random. This defaults to 100%, in which case the workload
802 is fully random. It can be set from anywhere from 0 to 100.
803 Setting it to 0 would make the workload fully sequential. Any
804 setting in between will result in a random mix of sequential
Jens Axboed9472272013-07-25 10:20:45 -0600805 and random IO, at the given percentages. It is possible to
806 set different values for reads, writes, and trim. To do so,
807 simply use a comma separated list. See blocksize.
Jens Axboe211c9b82013-04-26 08:56:17 -0600808
Jens Axboebb8895e2006-10-30 15:14:48 +0100809norandommap Normally fio will cover every block of the file when doing
810 random IO. If this option is given, fio will just get a
811 new random offset without looking at past io history. This
812 means that some blocks may not be read or written, and that
813 some blocks may be read/written more than once. This option
Jens Axboe83472392009-02-19 21:32:12 +0100814 is mutually exclusive with verify= if and only if multiple
815 blocksizes (via bsrange=) are used, since fio only tracks
816 complete rewrites of blocks.
Jens Axboebb8895e2006-10-30 15:14:48 +0100817
Jens Axboe0408c202011-08-08 09:07:28 +0200818softrandommap=bool See norandommap. If fio runs with the random block map
819 enabled and it fails to allocate the map, if this option is
820 set it will continue without a random block map. As coverage
821 will not be as complete as with random maps, this option is
Jens Axboe2b386d22008-03-26 10:32:57 +0100822 disabled by default.
823
Jens Axboee8b19612012-12-05 10:28:08 +0100824random_generator=str Fio supports the following engines for generating
825 IO offsets for random IO:
826
827 tausworthe Strong 2^88 cycle random number generator
828 lfsr Linear feedback shift register generator
829
830 Tausworthe is a strong random number generator, but it
831 requires tracking on the side if we want to ensure that
832 blocks are only read or written once. LFSR guarantees
833 that we never generate the same offset twice, and it's
834 also less computationally expensive. It's not a true
835 random generator, however, though for IO purposes it's
836 typically good enough. LFSR only works with single
837 block sizes, not with workloads that use multiple block
838 sizes. If used with such a workload, fio may read or write
839 some blocks multiple times.
Bruce Cran43f09da2013-02-24 11:09:11 +0000840
Jens Axboe71bfa162006-10-25 11:08:19 +0200841nice=int Run the job with the given nice value. See man nice(2).
842
843prio=int Set the io priority value of this job. Linux limits us to
844 a positive value between 0 and 7, with 0 being the highest.
845 See man ionice(1).
846
847prioclass=int Set the io priority class. See man ionice(1).
848
849thinktime=int Stall the job x microseconds after an io has completed before
850 issuing the next. May be used to simulate processing being
Jens Axboe48097d52007-02-17 06:30:44 +0100851 done by an application. See thinktime_blocks and
852 thinktime_spin.
853
854thinktime_spin=int
855 Only valid if thinktime is set - pretend to spend CPU time
856 doing something with the data received, before falling back
857 to sleeping for the rest of the period specified by
858 thinktime.
Jens Axboe9c1f7432007-01-03 20:43:19 +0100859
Jens Axboe4d01ece2013-05-17 12:47:11 +0200860thinktime_blocks=int
Jens Axboe9c1f7432007-01-03 20:43:19 +0100861 Only valid if thinktime is set - control how many blocks
862 to issue, before waiting 'thinktime' usecs. If not set,
863 defaults to 1 which will make fio wait 'thinktime' usecs
Jens Axboe4d01ece2013-05-17 12:47:11 +0200864 after every block. This effectively makes any queue depth
865 setting redundant, since no more than 1 IO will be queued
866 before we have to complete it and do our thinktime. In
867 other words, this setting effectively caps the queue depth
868 if the latter is larger.
Jens Axboe71bfa162006-10-25 11:08:19 +0200869
Jens Axboe581e7142009-06-09 12:47:16 +0200870rate=int Cap the bandwidth used by this job. The number is in bytes/sec,
Jens Axboeb09da8f2009-07-17 23:16:17 +0200871 the normal suffix rules apply. You can use rate=500k to limit
Jens Axboe581e7142009-06-09 12:47:16 +0200872 reads and writes to 500k each, or you can specify read and
873 writes separately. Using rate=1m,500k would limit reads to
874 1MB/sec and writes to 500KB/sec. Capping only reads or
875 writes can be done with rate=,500k or rate=500k,. The former
876 will only limit writes (to 500KB/sec), the latter will only
877 limit reads.
Jens Axboe71bfa162006-10-25 11:08:19 +0200878
879ratemin=int Tell fio to do whatever it can to maintain at least this
Jens Axboe4e991c22007-03-15 11:41:11 +0100880 bandwidth. Failing to meet this requirement, will cause
Jens Axboe581e7142009-06-09 12:47:16 +0200881 the job to exit. The same format as rate is used for
882 read vs write separation.
Jens Axboe4e991c22007-03-15 11:41:11 +0100883
884rate_iops=int Cap the bandwidth to this number of IOPS. Basically the same
885 as rate, just specified independently of bandwidth. If the
886 job is given a block size range instead of a fixed value,
Jens Axboe581e7142009-06-09 12:47:16 +0200887 the smallest block size is used as the metric. The same format
Anatol Pomozovde8f6de2013-09-26 16:31:34 -0700888 as rate is used for read vs write separation.
Jens Axboe4e991c22007-03-15 11:41:11 +0100889
890rate_iops_min=int If fio doesn't meet this rate of IO, it will cause
Jens Axboe581e7142009-06-09 12:47:16 +0200891 the job to exit. The same format as rate is used for read vs
Anatol Pomozovde8f6de2013-09-26 16:31:34 -0700892 write separation.
Jens Axboe71bfa162006-10-25 11:08:19 +0200893
Jens Axboe3e260a42013-12-09 12:38:53 -0700894latency_target=int If set, fio will attempt to find the max performance
895 point that the given workload will run at while maintaining a
896 latency below this target. The values is given in microseconds.
897 See latency_window and latency_percentile
898
899latency_window=int Used with latency_target to specify the sample window
900 that the job is run at varying queue depths to test the
901 performance. The value is given in microseconds.
902
903latency_percentile=float The percentage of IOs that must fall within the
904 criteria specified by latency_target and latency_window. If not
905 set, this defaults to 100.0, meaning that all IOs must be equal
906 or below to the value set by latency_target.
907
Jens Axboe15501532012-10-24 16:37:45 +0200908max_latency=int If set, fio will exit the job if it exceeds this maximum
909 latency. It will exit with an ETIME error.
910
Jens Axboe71bfa162006-10-25 11:08:19 +0200911ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
Jens Axboe6c219762006-11-03 15:51:45 +0100912 of milliseconds.
Jens Axboe71bfa162006-10-25 11:08:19 +0200913
914cpumask=int Set the CPU affinity of this job. The parameter given is a
Jens Axboea08bc172007-06-13 21:00:46 +0200915 bitmask of allowed CPU's the job may run on. So if you want
916 the allowed CPUs to be 1 and 5, you would pass the decimal
917 value of (1 << 1 | 1 << 5), or 34. See man
Jens Axboe7dbb6eb2007-05-22 09:13:31 +0200918 sched_setaffinity(2). This may not work on all supported
Jens Axboeb0ea08c2008-12-05 12:57:11 +0100919 operating systems or kernel versions. This option doesn't
920 work well for a higher CPU count than what you can store in
921 an integer mask, so it can only control cpus 1-32. For
922 boxes with larger CPU counts, use cpus_allowed.
Jens Axboe71bfa162006-10-25 11:08:19 +0200923
Jens Axboed2e268b2007-06-15 10:33:49 +0200924cpus_allowed=str Controls the same options as cpumask, but it allows a text
925 setting of the permitted CPUs instead. So to use CPUs 1 and
Jens Axboe62a72732008-12-08 11:37:01 +0100926 5, you would specify cpus_allowed=1,5. This options also
927 allows a range of CPUs. Say you wanted a binding to CPUs
928 1, 5, and 8-15, you would set cpus_allowed=1,5,8-15.
Jens Axboed2e268b2007-06-15 10:33:49 +0200929
Yufei Rend0b937e2012-10-19 23:11:52 -0400930numa_cpu_nodes=str Set this job running on spcified NUMA nodes' CPUs. The
931 arguments allow comma delimited list of cpu numbers,
932 A-B ranges, or 'all'. Note, to enable numa options support,
Jens Axboe67bf9822013-01-10 11:23:19 +0100933 fio must be built on a system with libnuma-dev(el) installed.
Yufei Rend0b937e2012-10-19 23:11:52 -0400934
935numa_mem_policy=str Set this job's memory policy and corresponding NUMA
936 nodes. Format of the argements:
937 <mode>[:<nodelist>]
938 `mode' is one of the following memory policy:
939 default, prefer, bind, interleave, local
940 For `default' and `local' memory policy, no node is
941 needed to be specified.
942 For `prefer', only one node is allowed.
943 For `bind' and `interleave', it allow comma delimited
944 list of numbers, A-B ranges, or 'all'.
945
Jens Axboee417fd62008-09-11 09:27:15 +0200946startdelay=time Start this job the specified number of seconds after fio
Jens Axboe71bfa162006-10-25 11:08:19 +0200947 has started. Only useful if the job file contains several
948 jobs, and you want to delay starting some jobs to a certain
949 time.
950
Jens Axboee417fd62008-09-11 09:27:15 +0200951runtime=time Tell fio to terminate processing after the specified number
Jens Axboe71bfa162006-10-25 11:08:19 +0200952 of seconds. It can be quite hard to determine for how long
953 a specified job will run, so this parameter is handy to
954 cap the total runtime to a given time.
955
Jens Axboecf4464c2007-04-17 20:14:42 +0200956time_based If set, fio will run for the duration of the runtime
Jens Axboebf9a3ed2008-06-05 11:53:08 +0200957 specified even if the file(s) are completely read or
Jens Axboecf4464c2007-04-17 20:14:42 +0200958 written. It will simply loop over the same workload
959 as many times as the runtime allows.
960
Jens Axboee417fd62008-09-11 09:27:15 +0200961ramp_time=time If set, fio will run the specified workload for this amount
Jens Axboe721938a2008-09-10 09:46:16 +0200962 of time before logging any performance numbers. Useful for
963 letting performance settle before logging results, thus
Jens Axboeb29ee5b2008-09-11 10:17:26 +0200964 minimizing the runtime required for stable results. Note
965 that the ramp_time is considered lead in time for a job,
966 thus it will increase the total runtime if a special timeout
967 or runtime is specified.
Jens Axboe721938a2008-09-10 09:46:16 +0200968
Jens Axboe71bfa162006-10-25 11:08:19 +0200969invalidate=bool Invalidate the buffer/page cache parts for this file prior
970 to starting io. Defaults to true.
971
972sync=bool Use sync io for buffered writes. For the majority of the
973 io engines, this means using O_SYNC.
974
Jens Axboed3aad8f2007-03-15 14:12:05 +0100975iomem=str
Jens Axboe71bfa162006-10-25 11:08:19 +0200976mem=str Fio can use various types of memory as the io unit buffer.
977 The allowed values are:
978
979 malloc Use memory from malloc(3) as the buffers.
980
981 shm Use shared memory as the buffers. Allocated
982 through shmget(2).
983
Jens Axboe74b025b2006-12-19 15:18:14 +0100984 shmhuge Same as shm, but use huge pages as backing.
985
Jens Axboe313cb202006-12-21 09:50:00 +0100986 mmap Use mmap to allocate buffers. May either be
987 anonymous memory, or can be file backed if
988 a filename is given after the option. The
989 format is mem=mmap:/path/to/file.
Jens Axboe71bfa162006-10-25 11:08:19 +0200990
Jens Axboed0bdaf42006-12-20 14:40:44 +0100991 mmaphuge Use a memory mapped huge file as the buffer
992 backing. Append filename after mmaphuge, ala
993 mem=mmaphuge:/hugetlbfs/file
994
Jens Axboe71bfa162006-10-25 11:08:19 +0200995 The area allocated is a function of the maximum allowed
Jens Axboe5394ae52006-12-20 20:15:41 +0100996 bs size for the job, multiplied by the io depth given. Note
997 that for shmhuge and mmaphuge to work, the system must have
998 free huge pages allocated. This can normally be checked
999 and set by reading/writing /proc/sys/vm/nr_hugepages on a
Jens Axboeb22989b2009-07-17 22:29:23 +02001000 Linux system. Fio assumes a huge page is 4MB in size. So
Jens Axboe5394ae52006-12-20 20:15:41 +01001001 to calculate the number of huge pages you need for a given
1002 job file, add up the io depth of all jobs (normally one unless
1003 iodepth= is used) and multiply by the maximum bs set. Then
1004 divide that number by the huge page size. You can see the
1005 size of the huge pages in /proc/meminfo. If no huge pages
1006 are allocated by having a non-zero number in nr_hugepages,
Jens Axboe56bb17f2006-12-20 20:27:36 +01001007 using mmaphuge or shmhuge will fail. Also see hugepage-size.
Jens Axboe5394ae52006-12-20 20:15:41 +01001008
1009 mmaphuge also needs to have hugetlbfs mounted and the file
1010 location should point there. So if it's mounted in /huge,
1011 you would use mem=mmaphuge:/huge/somefile.
Jens Axboe71bfa162006-10-25 11:08:19 +02001012
Jens Axboed529ee12009-07-01 10:33:03 +02001013iomem_align=int This indiciates the memory alignment of the IO memory buffers.
1014 Note that the given alignment is applied to the first IO unit
1015 buffer, if using iodepth the alignment of the following buffers
1016 are given by the bs used. In other words, if using a bs that is
1017 a multiple of the page sized in the system, all buffers will
1018 be aligned to this value. If using a bs that is not page
1019 aligned, the alignment of subsequent IO memory buffers is the
1020 sum of the iomem_align and bs used.
1021
Jens Axboef7fa2652009-03-09 14:20:20 +01001022hugepage-size=int
Jens Axboe56bb17f2006-12-20 20:27:36 +01001023 Defines the size of a huge page. Must at least be equal
Jens Axboeb22989b2009-07-17 22:29:23 +02001024 to the system setting, see /proc/meminfo. Defaults to 4MB.
Jens Axboec51074e2006-12-20 20:28:33 +01001025 Should probably always be a multiple of megabytes, so using
1026 hugepage-size=Xm is the preferred way to set this to avoid
1027 setting a non-pow-2 bad value.
Jens Axboe56bb17f2006-12-20 20:27:36 +01001028
Jens Axboe71bfa162006-10-25 11:08:19 +02001029exitall When one job finishes, terminate the rest. The default is
1030 to wait for each job to finish, sometimes that is not the
1031 desired action.
1032
1033bwavgtime=int Average the calculated bandwidth over the given time. Value
Jens Axboe6c219762006-11-03 15:51:45 +01001034 is specified in milliseconds.
Jens Axboe71bfa162006-10-25 11:08:19 +02001035
Jens Axboec8eeb9d2011-10-05 14:02:22 +02001036iopsavgtime=int Average the calculated IOPS over the given time. Value
1037 is specified in milliseconds.
1038
Jens Axboe71bfa162006-10-25 11:08:19 +02001039create_serialize=bool If true, serialize the file creating for the jobs.
1040 This may be handy to avoid interleaving of data
1041 files, which may greatly depend on the filesystem
1042 used and even the number of processors in the system.
1043
1044create_fsync=bool fsync the data file after creation. This is the
1045 default.
1046
Jens Axboe814452b2009-03-04 12:53:13 +01001047create_on_open=bool Don't pre-setup the files for IO, just create open()
1048 when it's time to do IO to that file.
1049
Jens Axboe25460cf2012-05-02 13:58:02 +02001050create_only=bool If true, fio will only run the setup phase of the job.
1051 If files need to be laid out or updated on disk, only
1052 that will be done. The actual job contents are not
1053 executed.
1054
Zhang, Yanminafad68f2009-05-20 11:30:55 +02001055pre_read=bool If this is given, files will be pre-read into memory before
Jens Axboe34f1c042009-06-02 14:19:25 +02001056 starting the given IO operation. This will also clear
1057 the 'invalidate' flag, since it is pointless to pre-read
Jens Axboe9c0d2242009-07-01 12:26:28 +02001058 and then drop the cache. This will only work for IO engines
1059 that are seekable, since they allow you to read the same data
1060 multiple times. Thus it will not work on eg network or splice
1061 IO.
Zhang, Yanminafad68f2009-05-20 11:30:55 +02001062
Jens Axboee545a6c2007-01-14 00:00:29 +01001063unlink=bool Unlink the job files when done. Not the default, as repeated
Jens Axboebf9a3ed2008-06-05 11:53:08 +02001064 runs of that job would then waste time recreating the file
1065 set again and again.
Jens Axboe71bfa162006-10-25 11:08:19 +02001066
1067loops=int Run the specified number of iterations of this job. Used
1068 to repeat the same workload a given number of times. Defaults
1069 to 1.
1070
Juan Casse62167762013-09-17 14:06:13 -07001071verify_only Do not perform specified workload---only verify data still
1072 matches previous invocation of this workload. This option
1073 allows one to check data multiple times at a later date
1074 without overwriting it. This option makes sense only for
1075 workloads that write data, and does not support workloads
1076 with the time_based option set.
1077
Jens Axboe68e1f292007-08-10 10:32:14 +02001078do_verify=bool Run the verify phase after a write phase. Only makes sense if
Shawn Lewise84c73a2007-08-02 22:19:32 +02001079 verify is set. Defaults to 1.
1080
Jens Axboe71bfa162006-10-25 11:08:19 +02001081verify=str If writing to a file, fio can verify the file contents
1082 after each iteration of the job. The allowed values are:
1083
1084 md5 Use an md5 sum of the data area and store
1085 it in the header of each block.
1086
Jens Axboe17dc34d2007-07-27 15:36:02 +02001087 crc64 Use an experimental crc64 sum of the data
1088 area and store it in the header of each
1089 block.
1090
Jens Axboebac39e02008-06-11 20:46:19 +02001091 crc32c Use a crc32c sum of the data area and store
1092 it in the header of each block.
1093
Jens Axboe38455912008-08-04 15:35:26 +02001094 crc32c-intel Use hardware assisted crc32c calcuation
Jens Axboe0539d752010-06-21 15:22:56 +02001095 provided on SSE4.2 enabled processors. Falls
1096 back to regular software crc32c, if not
1097 supported by the system.
Jens Axboe38455912008-08-04 15:35:26 +02001098
Jens Axboe71bfa162006-10-25 11:08:19 +02001099 crc32 Use a crc32 sum of the data area and store
1100 it in the header of each block.
1101
Jens Axboe969f7ed2007-07-27 09:07:17 +02001102 crc16 Use a crc16 sum of the data area and store
1103 it in the header of each block.
1104
Jens Axboe17dc34d2007-07-27 15:36:02 +02001105 crc7 Use a crc7 sum of the data area and store
1106 it in the header of each block.
1107
Jens Axboe844ea602014-02-20 13:21:45 -08001108 xxhash Use xxhash as the checksum function. Generally
1109 the fastest software checksum that fio
1110 supports.
1111
Jens Axboecd14cc12007-07-30 10:59:33 +02001112 sha512 Use sha512 as the checksum function.
1113
1114 sha256 Use sha256 as the checksum function.
1115
Jens Axboe7c353ce2009-08-09 22:40:33 +02001116 sha1 Use optimized sha1 as the checksum function.
1117
Shawn Lewis7437ee82007-08-02 21:05:58 +02001118 meta Write extra information about each io
1119 (timestamp, block number etc.). The block
Juan Casse62167762013-09-17 14:06:13 -07001120 number is verified. The io sequence number is
1121 verified for workloads that write data.
1122 See also verify_pattern.
Shawn Lewis7437ee82007-08-02 21:05:58 +02001123
Jens Axboe36690c92007-03-26 10:23:34 +02001124 null Only pretend to verify. Useful for testing
1125 internals with ioengine=null, not for much
1126 else.
1127
Jens Axboe6c219762006-11-03 15:51:45 +01001128 This option can be used for repeated burn-in tests of a
Jens Axboe71bfa162006-10-25 11:08:19 +02001129 system to make sure that the written data is also
Jens Axboeb892dc02009-09-05 20:37:35 +02001130 correctly read back. If the data direction given is
1131 a read or random read, fio will assume that it should
1132 verify a previously written file. If the data direction
1133 includes any form of write, the verify will be of the
1134 newly written data.
Jens Axboe71bfa162006-10-25 11:08:19 +02001135
Jens Axboe160b9662007-03-27 10:59:49 +02001136verifysort=bool If set, fio will sort written verify blocks when it deems
1137 it faster to read them back in a sorted manner. This is
1138 often the case when overwriting an existing file, since
1139 the blocks are already laid out in the file system. You
1140 can ignore this option unless doing huge amounts of really
1141 fast IO where the red-black tree sorting CPU time becomes
1142 significant.
Shawn Lewis3f9f4e22007-07-28 21:10:37 +02001143
Jens Axboef7fa2652009-03-09 14:20:20 +01001144verify_offset=int Swap the verification header with data somewhere else
Shawn Lewis546a9142007-07-28 21:11:37 +02001145 in the block before writing. Its swapped back before
1146 verifying.
1147
Jens Axboef7fa2652009-03-09 14:20:20 +01001148verify_interval=int Write the verification header at a finer granularity
Shawn Lewis3f9f4e22007-07-28 21:10:37 +02001149 than the blocksize. It will be written for chunks the
1150 size of header_interval. blocksize should divide this
1151 evenly.
Jens Axboe90059d62007-07-30 09:33:12 +02001152
Radha Ramachandran0e92f872009-10-27 20:14:27 +01001153verify_pattern=str If set, fio will fill the io buffers with this
Shawn Lewise28218f2008-01-16 11:01:33 +01001154 pattern. Fio defaults to filling with totally random
1155 bytes, but sometimes it's interesting to fill with a known
1156 pattern for io verification purposes. Depending on the
1157 width of the pattern, fio will fill 1/2/3/4 bytes of the
Radha Ramachandran0e92f872009-10-27 20:14:27 +01001158 buffer at the time(it can be either a decimal or a hex number).
1159 The verify_pattern if larger than a 32-bit quantity has to
Jens Axboe996093b2010-06-24 08:37:13 +02001160 be a hex number that starts with either "0x" or "0X". Use
1161 with verify=meta.
Shawn Lewise28218f2008-01-16 11:01:33 +01001162
Jens Axboe68e1f292007-08-10 10:32:14 +02001163verify_fatal=bool Normally fio will keep checking the entire contents
Jens Axboea12a3b42007-08-09 10:20:54 +02001164 before quitting on a block verification failure. If this
1165 option is set, fio will exit the job on the first observed
1166 failure.
Jens Axboee8462bd2009-07-06 12:59:04 +02001167
Jens Axboeb463e932011-01-12 09:03:23 +01001168verify_dump=bool If set, dump the contents of both the original data
1169 block and the data block we read off disk to files. This
1170 allows later analysis to inspect just what kind of data
Jens Axboeef71e312011-10-25 22:43:36 +02001171 corruption occurred. Off by default.
Jens Axboeb463e932011-01-12 09:03:23 +01001172
Jens Axboee8462bd2009-07-06 12:59:04 +02001173verify_async=int Fio will normally verify IO inline from the submitting
1174 thread. This option takes an integer describing how many
1175 async offload threads to create for IO verification instead,
1176 causing fio to offload the duty of verifying IO contents
Jens Axboec85c3242009-07-06 14:12:57 +02001177 to one or more separate threads. If using this offload
1178 option, even sync IO engines can benefit from using an
1179 iodepth setting higher than 1, as it allows them to have
1180 IO in flight while verifies are running.
Jens Axboee8462bd2009-07-06 12:59:04 +02001181
1182verify_async_cpus=str Tell fio to set the given CPU affinity on the
1183 async IO verification threads. See cpus_allowed for the
1184 format used.
Jens Axboe6f874182010-06-21 12:53:26 +02001185
1186verify_backlog=int Fio will normally verify the written contents of a
1187 job that utilizes verify once that job has completed. In
1188 other words, everything is written then everything is read
1189 back and verified. You may want to verify continually
1190 instead for a variety of reasons. Fio stores the meta data
1191 associated with an IO block in memory, so for large
1192 verify workloads, quite a bit of memory would be used up
1193 holding this meta data. If this option is enabled, fio
Jens Axboef42195a2010-10-26 08:10:58 -06001194 will write only N blocks before verifying these blocks.
1195
Jens Axboe6f874182010-06-21 12:53:26 +02001196verify_backlog_batch=int Control how many blocks fio will verify
1197 if verify_backlog is set. If not set, will default to
1198 the value of verify_backlog (meaning the entire queue
Jens Axboef42195a2010-10-26 08:10:58 -06001199 is read back and verified). If verify_backlog_batch is
1200 less than verify_backlog then not all blocks will be verified,
1201 if verify_backlog_batch is larger than verify_backlog, some
1202 blocks will be verified more than once.
Bruce Cran66c098b2012-11-27 12:16:07 +00001203
Jens Axboed3923652011-08-03 12:38:39 +02001204stonewall
Anatol Pomozovde8f6de2013-09-26 16:31:34 -07001205wait_for_previous Wait for preceding jobs in the job file to exit, before
Jens Axboe71bfa162006-10-25 11:08:19 +02001206 starting this one. Can be used to insert serialization
Jens Axboeb3d62a72007-03-20 14:23:26 +01001207 points in the job file. A stone wall also implies starting
1208 a new reporting group.
1209
Akash Vermaabcab6a2012-10-04 15:58:28 -07001210new_group Start a new reporting group. See: group_reporting.
Jens Axboe71bfa162006-10-25 11:08:19 +02001211
1212numjobs=int Create the specified number of clones of this job. May be
1213 used to setup a larger number of threads/processes doing
Akash Vermaabcab6a2012-10-04 15:58:28 -07001214 the same thing. Each thread is reported separately; to see
1215 statistics for all clones as a whole, use group_reporting in
1216 conjunction with new_group.
Jens Axboefa28c852007-03-06 15:40:49 +01001217
Akash Vermaabcab6a2012-10-04 15:58:28 -07001218group_reporting It may sometimes be interesting to display statistics for
Jens Axboe04b2f792012-10-10 09:09:59 -06001219 groups of jobs as a whole instead of for each individual job.
1220 This is especially true if 'numjobs' is used; looking at
1221 individual thread/process output quickly becomes unwieldy.
1222 To see the final report per-group instead of per-job, use
1223 'group_reporting'. Jobs in a file will be part of the same
1224 reporting group, unless if separated by a stonewall, or by
1225 using 'new_group'.
Jens Axboe71bfa162006-10-25 11:08:19 +02001226
1227thread fio defaults to forking jobs, however if this option is
1228 given, fio will use pthread_create(3) to create threads
1229 instead.
1230
Jens Axboef7fa2652009-03-09 14:20:20 +01001231zonesize=int Divide a file into zones of the specified size. See zoneskip.
Jens Axboe71bfa162006-10-25 11:08:19 +02001232
Jens Axboef7fa2652009-03-09 14:20:20 +01001233zoneskip=int Skip the specified number of bytes when zonesize data has
Jens Axboe71bfa162006-10-25 11:08:19 +02001234 been read. The two zone options can be used to only do
1235 io on zones of a file.
1236
Jens Axboe076efc72006-10-27 11:24:25 +02001237write_iolog=str Write the issued io patterns to the specified file. See
Stefan Hajnoczi5b42a482011-01-08 20:28:41 +01001238 read_iolog. Specify a separate file for each job, otherwise
1239 the iologs will be interspersed and the file may be corrupt.
Jens Axboe71bfa162006-10-25 11:08:19 +02001240
Jens Axboe076efc72006-10-27 11:24:25 +02001241read_iolog=str Open an iolog with the specified file name and replay the
Jens Axboe71bfa162006-10-25 11:08:19 +02001242 io patterns it contains. This can be used to store a
Jens Axboe6df8ada2007-05-15 13:23:19 +02001243 workload and replay it sometime later. The iolog given
1244 may also be a blktrace binary file, which allows fio
1245 to replay a workload captured by blktrace. See blktrace
1246 for how to capture such logging data. For blktrace replay,
1247 the file needs to be turned into a blkparse binary data
Jens Axboeea3e51c2010-05-17 19:51:45 +02001248 file first (blkparse <device> -o /dev/null -d file_for_fio.bin).
Bruce Cran66c098b2012-11-27 12:16:07 +00001249
David Nellans64bbb862010-08-24 22:13:30 +02001250replay_no_stall=int When replaying I/O with read_iolog the default behavior
Jens Axboe62776222010-09-02 15:30:16 +02001251 is to attempt to respect the time stamps within the log and
1252 replay them with the appropriate delay between IOPS. By
1253 setting this variable fio will not respect the timestamps and
1254 attempt to replay them as fast as possible while still
1255 respecting ordering. The result is the same I/O pattern to a
1256 given device, but different timings.
Jens Axboe71bfa162006-10-25 11:08:19 +02001257
David Nellansd1c46c02010-08-31 21:20:47 +02001258replay_redirect=str While replaying I/O patterns using read_iolog the
1259 default behavior is to replay the IOPS onto the major/minor
1260 device that each IOP was recorded from. This is sometimes
Anatol Pomozovde8f6de2013-09-26 16:31:34 -07001261 undesirable because on a different machine those major/minor
David Nellansd1c46c02010-08-31 21:20:47 +02001262 numbers can map to a different device. Changing hardware on
1263 the same system can also result in a different major/minor
1264 mapping. Replay_redirect causes all IOPS to be replayed onto
1265 the single specified device regardless of the device it was
1266 recorded from. i.e. replay_redirect=/dev/sdc would cause all
1267 IO in the blktrace to be replayed onto /dev/sdc. This means
1268 multiple devices will be replayed onto a single, if the trace
1269 contains multiple devices. If you want multiple devices to be
1270 replayed concurrently to multiple redirected devices you must
1271 blkparse your trace into separate traces and replay them with
1272 independent fio invocations. Unfortuantely this also breaks
1273 the strict time ordering between multiple device accesses.
1274
Jens Axboee3cedca2008-11-19 19:57:52 +01001275write_bw_log=str If given, write a bandwidth log of the jobs in this job
Jens Axboe71bfa162006-10-25 11:08:19 +02001276 file. Can be used to store data of the bandwidth of the
Jens Axboee0da9bc2006-10-25 13:08:57 +02001277 jobs in their lifetime. The included fio_generate_plots
1278 script uses gnuplot to turn these text files into nice
Lucian Adrian Grijincuddb754d2012-04-05 18:18:35 -06001279 graphs. See write_lat_log for behaviour of given
1280 filename. For this option, the suffix is _bw.log.
Jens Axboe71bfa162006-10-25 11:08:19 +02001281
Jens Axboee3cedca2008-11-19 19:57:52 +01001282write_lat_log=str Same as write_bw_log, except that this option stores io
Jens Axboe02af0982010-06-24 09:59:34 +02001283 submission, completion, and total latencies instead. If no
1284 filename is given with this option, the default filename of
1285 "jobname_type.log" is used. Even if the filename is given,
1286 fio will still append the type of log. So if one specifies
Jens Axboee3cedca2008-11-19 19:57:52 +01001287
1288 write_lat_log=foo
1289
Jens Axboed5d94592013-05-09 21:10:58 +02001290 The actual log names will be foo_slat.log, foo_clat.log,
Jens Axboe02af0982010-06-24 09:59:34 +02001291 and foo_lat.log. This helps fio_generate_plot fine the logs
1292 automatically.
Jens Axboe71bfa162006-10-25 11:08:19 +02001293
Jens Axboeb8bc8cb2011-12-01 09:04:31 +01001294write_iops_log=str Same as write_bw_log, but writes IOPS. If no filename is
1295 given with this option, the default filename of
1296 "jobname_type.log" is used. Even if the filename is given,
1297 fio will still append the type of log.
1298
1299log_avg_msec=int By default, fio will log an entry in the iops, latency,
1300 or bw log for every IO that completes. When writing to the
1301 disk log, that can quickly grow to a very large size. Setting
1302 this option makes fio average the each log entry over the
1303 specified period of time, reducing the resolution of the log.
1304 Defaults to 0.
1305
Jens Axboef7fa2652009-03-09 14:20:20 +01001306lockmem=int Pin down the specified amount of memory with mlock(2). Can
Jens Axboe71bfa162006-10-25 11:08:19 +02001307 potentially be used instead of removing memory or booting
1308 with less memory to simulate a smaller amount of memory.
Jens Axboe81c6b6c2013-04-10 19:30:50 +02001309 The amount specified is per worker.
Jens Axboe71bfa162006-10-25 11:08:19 +02001310
1311exec_prerun=str Before running this job, issue the command specified
Jens Axboe74c8c482013-07-17 22:15:09 -06001312 through system(3). Output is redirected in a file called
1313 jobname.prerun.txt.
Jens Axboe71bfa162006-10-25 11:08:19 +02001314
1315exec_postrun=str After the job completes, issue the command specified
Jens Axboe74c8c482013-07-17 22:15:09 -06001316 though system(3). Output is redirected in a file called
1317 jobname.postrun.txt.
Jens Axboe71bfa162006-10-25 11:08:19 +02001318
1319ioscheduler=str Attempt to switch the device hosting the file to the specified
1320 io scheduler before running.
1321
Jens Axboe0a839f32007-04-26 09:02:34 +02001322disk_util=bool Generate disk utilization statistics, if the platform
1323 supports it. Defaults to on.
1324
Jens Axboe02af0982010-06-24 09:59:34 +02001325disable_lat=bool Disable measurements of total latency numbers. Useful
Jens Axboe9520ebb2008-10-16 21:03:27 +02001326 only for cutting back the number of calls to gettimeofday,
1327 as that does impact performance at really high IOPS rates.
1328 Note that to really get rid of a large amount of these
1329 calls, this option must be used with disable_slat and
1330 disable_bw as well.
1331
Jens Axboe02af0982010-06-24 09:59:34 +02001332disable_clat=bool Disable measurements of completion latency numbers. See
1333 disable_lat.
1334
Jens Axboe9520ebb2008-10-16 21:03:27 +02001335disable_slat=bool Disable measurements of submission latency numbers. See
Jens Axboe02af0982010-06-24 09:59:34 +02001336 disable_slat.
Jens Axboe9520ebb2008-10-16 21:03:27 +02001337
1338disable_bw=bool Disable measurements of throughput/bandwidth numbers. See
Jens Axboe02af0982010-06-24 09:59:34 +02001339 disable_lat.
Jens Axboe9520ebb2008-10-16 21:03:27 +02001340
Yu-ju Hong83349192011-08-13 00:53:44 +02001341clat_percentiles=bool Enable the reporting of percentiles of
1342 completion latencies.
1343
1344percentile_list=float_list Overwrite the default list of percentiles
1345 for completion latencies. Each number is a floating
1346 number in the range (0,100], and the maximum length of
1347 the list is 20. Use ':' to separate the numbers, and
1348 list the numbers in ascending order. For example,
1349 --percentile_list=99.5:99.9 will cause fio to report
1350 the values of completion latency below which 99.5% and
1351 99.9% of the observed latencies fell, respectively.
1352
Jens Axboe23893642012-12-17 14:44:08 +01001353clocksource=str Use the given clocksource as the base of timing. The
1354 supported options are:
1355
1356 gettimeofday gettimeofday(2)
1357
1358 clock_gettime clock_gettime(2)
1359
1360 cpu Internal CPU clock source
1361
1362 cpu is the preferred clocksource if it is reliable, as it
1363 is very fast (and fio is heavy on time calls). Fio will
1364 automatically use this clocksource if it's supported and
1365 considered reliable on the system it is running on, unless
1366 another clocksource is specifically set. For x86/x86-64 CPUs,
1367 this means supporting TSC Invariant.
1368
Jens Axboe993bf482008-11-14 13:04:53 +01001369gtod_reduce=bool Enable all of the gettimeofday() reducing options
1370 (disable_clat, disable_slat, disable_bw) plus reduce
1371 precision of the timeout somewhat to really shrink
1372 the gettimeofday() call count. With this option enabled,
1373 we only do about 0.4% of the gtod() calls we would have
1374 done if all time keeping was enabled.
1375
Jens Axboebe4ecfd2008-12-08 14:10:52 +01001376gtod_cpu=int Sometimes it's cheaper to dedicate a single thread of
1377 execution to just getting the current time. Fio (and
1378 databases, for instance) are very intensive on gettimeofday()
1379 calls. With this option, you can set one CPU aside for
1380 doing nothing but logging current time to a shared memory
1381 location. Then the other threads/processes that run IO
1382 workloads need only copy that segment, instead of entering
1383 the kernel with a gettimeofday() call. The CPU set aside
1384 for doing these time calls will be excluded from other
1385 uses. Fio will manually clear it from the CPU mask of other
1386 jobs.
Jens Axboea696fa22009-12-04 10:05:02 +01001387
Steven Lang06842022011-11-17 09:45:17 +01001388continue_on_error=str Normally fio will exit the job on the first observed
Radha Ramachandranf2bba182009-06-15 08:40:16 +02001389 failure. If this option is set, fio will continue the job when
1390 there is a 'non-fatal error' (EIO or EILSEQ) until the runtime
1391 is exceeded or the I/O size specified is completed. If this
1392 option is used, there are two more stats that are appended,
1393 the total error count and the first error. The error field
1394 given in the stats is the first error that was hit during the
1395 run.
Jens Axboebe4ecfd2008-12-08 14:10:52 +01001396
Steven Lang06842022011-11-17 09:45:17 +01001397 The allowed values are:
1398
1399 none Exit on any IO or verify errors.
1400
1401 read Continue on read errors, exit on all others.
1402
1403 write Continue on write errors, exit on all others.
1404
1405 io Continue on any IO error, exit on all others.
1406
1407 verify Continue on verify errors, exit on all others.
1408
1409 all Continue on all errors.
1410
1411 0 Backward-compatible alias for 'none'.
1412
1413 1 Backward-compatible alias for 'all'.
1414
Dmitry Monakhov8b28bd42012-09-23 15:46:09 +04001415ignore_error=str Sometimes you want to ignore some errors during test
1416 in that case you can specify error list for each error type.
1417 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1418 errors for given error type is separated with ':'. Error
1419 may be symbol ('ENOSPC', 'ENOMEM') or integer.
1420 Example:
1421 ignore_error=EAGAIN,ENOSPC:122
Bruce Cran66c098b2012-11-27 12:16:07 +00001422 This option will ignore EAGAIN from READ, and ENOSPC and
1423 122(EDQUOT) from WRITE.
Dmitry Monakhov8b28bd42012-09-23 15:46:09 +04001424
1425error_dump=bool If set dump every error even if it is non fatal, true
1426 by default. If disabled only fatal error will be dumped
Bruce Cran66c098b2012-11-27 12:16:07 +00001427
Jens Axboe6adb38a2009-12-07 08:01:26 +01001428cgroup=str Add job to this control group. If it doesn't exist, it will
1429 be created. The system must have a mounted cgroup blkio
1430 mount point for this to work. If your system doesn't have it
1431 mounted, you can do so with:
Jens Axboea696fa22009-12-04 10:05:02 +01001432
1433 # mount -t cgroup -o blkio none /cgroup
1434
Jens Axboea696fa22009-12-04 10:05:02 +01001435cgroup_weight=int Set the weight of the cgroup to this value. See
1436 the documentation that comes with the kernel, allowed values
1437 are in the range of 100..1000.
Jens Axboe71bfa162006-10-25 11:08:19 +02001438
Vivek Goyal7de87092010-03-31 22:55:15 +02001439cgroup_nodelete=bool Normally fio will delete the cgroups it has created after
1440 the job completion. To override this behavior and to leave
1441 cgroups around after the job completion, set cgroup_nodelete=1.
1442 This can be useful if one wants to inspect various cgroup
1443 files after job completion. Default: false
1444
Jens Axboee0b0d892009-12-08 10:10:14 +01001445uid=int Instead of running as the invoking user, set the user ID to
1446 this value before the thread/process does any work.
1447
1448gid=int Set group ID, see uid.
1449
Dan Ehrenberg9e684a42012-02-20 11:05:14 +01001450flow_id=int The ID of the flow. If not specified, it defaults to being a
1451 global flow. See flow.
1452
1453flow=int Weight in token-based flow control. If this value is used, then
1454 there is a 'flow counter' which is used to regulate the
1455 proportion of activity between two or more jobs. fio attempts
1456 to keep this flow counter near zero. The 'flow' parameter
1457 stands for how much should be added or subtracted to the flow
1458 counter on each iteration of the main I/O loop. That is, if
1459 one job has flow=8 and another job has flow=-1, then there
1460 will be a roughly 1:8 ratio in how much one runs vs the other.
1461
1462flow_watermark=int The maximum value that the absolute value of the flow
1463 counter is allowed to reach before the job must wait for a
1464 lower value of the counter.
1465
1466flow_sleep=int The period of time, in microseconds, to wait after the flow
1467 watermark has been exceeded before retrying operations
1468
Steven Langde890a12011-11-09 14:03:34 +01001469In addition, there are some parameters which are only valid when a specific
1470ioengine is in use. These are used identically to normal parameters, with the
1471caveat that when used on the command line, they must come after the ioengine
1472that defines them is selected.
1473
1474[libaio] userspace_reap Normally, with the libaio engine in use, fio will use
1475 the io_getevents system call to reap newly returned events.
1476 With this flag turned on, the AIO ring will be read directly
1477 from user-space to reap events. The reaping mode is only
1478 enabled when polling for a minimum of 0 events (eg when
1479 iodepth_batch_complete=0).
1480
Jens Axboe03530502012-03-19 21:45:12 +01001481[cpu] cpuload=int Attempt to use the specified percentage of CPU cycles.
1482
1483[cpu] cpuchunks=int Split the load into cycles of the given time. In
1484 microseconds.
1485
Steven Langde890a12011-11-09 14:03:34 +01001486[netsplice] hostname=str
1487[net] hostname=str The host name or IP address to use for TCP or UDP based IO.
1488 If the job is a TCP listener or UDP reader, the hostname is not
Shawn Bohrerb511c9a2013-07-19 13:24:06 -05001489 used and must be omitted unless it is a valid UDP multicast
1490 address.
Steven Langde890a12011-11-09 14:03:34 +01001491
1492[netsplice] port=int
1493[net] port=int The TCP or UDP port to bind to or connect to.
1494
Shawn Bohrerb93b6a22013-07-19 13:24:07 -05001495[netsplice] interface=str
1496[net] interface=str The IP address of the network interface used to send or
1497 receive UDP multicast
1498
Shawn Bohrerd3a623d2013-07-19 13:24:08 -05001499[netsplice] ttl=int
1500[net] ttl=int Time-to-live value for outgoing UDP multicast packets.
1501 Default: 1
1502
Jens Axboe1d360ff2013-01-31 13:33:45 +01001503[netsplice] nodelay=bool
1504[net] nodelay=bool Set TCP_NODELAY on TCP connections.
1505
Steven Langde890a12011-11-09 14:03:34 +01001506[netsplice] protocol=str
1507[netsplice] proto=str
1508[net] protocol=str
1509[net] proto=str The network protocol to use. Accepted values are:
1510
1511 tcp Transmission control protocol
Jens Axboe49ccb8c2014-01-23 16:49:37 -08001512 tcpv6 Transmission control protocol V6
Bruce Cranf5cc3d02012-10-10 08:17:44 -06001513 udp User datagram protocol
Jens Axboe49ccb8c2014-01-23 16:49:37 -08001514 udpv6 User datagram protocol V6
Steven Langde890a12011-11-09 14:03:34 +01001515 unix UNIX domain socket
1516
1517 When the protocol is TCP or UDP, the port must also be given,
1518 as well as the hostname if the job is a TCP listener or UDP
1519 reader. For unix sockets, the normal filename option should be
1520 used and the port is invalid.
1521
1522[net] listen For TCP network connections, tell fio to listen for incoming
1523 connections rather than initiating an outgoing connection. The
1524 hostname must be omitted if this option is used.
Shawn Bohrerb511c9a2013-07-19 13:24:06 -05001525[net] pingpong Normaly a network writer will just continue writing data, and
Jens Axboe7aeb1e92012-12-06 20:53:57 +01001526 a network reader will just consume packages. If pingpong=1
1527 is set, a writer will send its normal payload to the reader,
1528 then wait for the reader to send the same payload back. This
1529 allows fio to measure network latencies. The submission
1530 and completion latencies then measure local time spent
1531 sending or receiving, and the completion latency measures
1532 how long it took for the other end to receive and send back.
Shawn Bohrerb511c9a2013-07-19 13:24:06 -05001533 For UDP multicast traffic pingpong=1 should only be set for a
1534 single reader when multiple readers are listening to the same
1535 address.
Jens Axboe7aeb1e92012-12-06 20:53:57 +01001536
Dmitry Monakhovd54fce82012-09-20 15:37:17 +04001537[e4defrag] donorname=str
1538 File will be used as a block donor(swap extents between files)
1539[e4defrag] inplace=int
Bruce Cran66c098b2012-11-27 12:16:07 +00001540 Configure donor file blocks allocation strategy
Dmitry Monakhovd54fce82012-09-20 15:37:17 +04001541 0(default): Preallocate donor's file on init
1542 1 : allocate space immidietly inside defragment event,
1543 and free right after event
1544
Steven Langde890a12011-11-09 14:03:34 +01001545
1546
Jens Axboe71bfa162006-10-25 11:08:19 +020015476.0 Interpreting the output
1548---------------------------
1549
1550fio spits out a lot of output. While running, fio will display the
1551status of the jobs created. An example of that would be:
1552
Jens Axboe73c8b082007-01-11 19:25:52 +01001553Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
Jens Axboe71bfa162006-10-25 11:08:19 +02001554
1555The characters inside the square brackets denote the current status of
1556each thread. The possible values (in typical life cycle order) are:
1557
1558Idle Run
1559---- ---
1560P Thread setup, but not started.
1561C Thread created.
Jens Axboe9c6f6312012-11-07 09:15:45 +01001562I Thread initialized, waiting or generating necessary data.
Jens Axboeb0f65862009-05-20 11:52:15 +02001563 p Thread running pre-reading file(s).
Jens Axboe71bfa162006-10-25 11:08:19 +02001564 R Running, doing sequential reads.
1565 r Running, doing random reads.
1566 W Running, doing sequential writes.
1567 w Running, doing random writes.
1568 M Running, doing mixed sequential reads/writes.
1569 m Running, doing mixed random reads/writes.
1570 F Running, currently waiting for fsync()
Jens Axboefc6bd432009-04-29 09:52:10 +02001571 V Running, doing verification of written data.
Jens Axboe71bfa162006-10-25 11:08:19 +02001572E Thread exited, not reaped by main thread yet.
Jens Axboe4f7e57a2012-03-30 21:21:20 +02001573_ Thread reaped, or
1574X Thread reaped, exited with an error.
Jens Axboea5e371a2012-04-02 09:47:09 -07001575K Thread reaped, exited due to signal.
Jens Axboe71bfa162006-10-25 11:08:19 +02001576
1577The other values are fairly self explanatory - number of threads
Jens Axboec9f60302007-07-20 12:43:05 +02001578currently running and doing io, rate of io since last check (read speed
1579listed first, then write speed), and the estimated completion percentage
1580and time for the running group. It's impossible to estimate runtime of
Jens Axboe4f7e57a2012-03-30 21:21:20 +02001581the following groups (if any). Note that the string is displayed in order,
1582so it's possible to tell which of the jobs are currently doing what. The
1583first character is the first job defined in the job file, and so forth.
Jens Axboe71bfa162006-10-25 11:08:19 +02001584
1585When fio is done (or interrupted by ctrl-c), it will show the data for
1586each thread, group of threads, and disks in that order. For each data
1587direction, the output looks like:
1588
1589Client1 (g=0): err= 0:
Paul Dubs35649e52011-07-21 16:04:52 +02001590 write: io= 32MB, bw= 666KB/s, iops=89 , runt= 50320msec
Jens Axboe6104ddb2007-01-11 14:24:29 +01001591 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
1592 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
Jens Axboeb22989b2009-07-17 22:29:23 +02001593 bw (KB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
Jens Axboee7823a92007-09-07 20:33:33 +02001594 cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17
Jens Axboe71619dc2007-01-13 23:56:33 +01001595 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
Jens Axboe838bc702008-05-22 13:08:23 +02001596 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
1597 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
Jens Axboe30061b92007-04-17 13:31:34 +02001598 issued r/w: total=0/32768, short=0/0
Jens Axboe8abdce62007-02-21 10:22:55 +01001599 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
1600 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
Jens Axboe71bfa162006-10-25 11:08:19 +02001601
1602The client number is printed, along with the group id and error of that
1603thread. Below is the io statistics, here for writes. In the order listed,
1604they denote:
1605
1606io= Number of megabytes io performed
1607bw= Average bandwidth rate
Paul Dubs35649e52011-07-21 16:04:52 +02001608iops= Average IOs performed per second
Jens Axboe71bfa162006-10-25 11:08:19 +02001609runt= The runtime of that thread
Jens Axboe72fbda22007-03-20 10:02:06 +01001610 slat= Submission latency (avg being the average, stdev being the
Jens Axboe71bfa162006-10-25 11:08:19 +02001611 standard deviation). This is the time it took to submit
1612 the io. For sync io, the slat is really the completion
Jens Axboe8a35c712007-06-19 09:53:31 +02001613 latency, since queue/complete is one operation there. This
Jens Axboebf9a3ed2008-06-05 11:53:08 +02001614 value can be in milliseconds or microseconds, fio will choose
Jens Axboe8a35c712007-06-19 09:53:31 +02001615 the most appropriate base and print that. In the example
Lucian Adrian Grijincu0d237712012-04-03 14:42:48 -06001616 above, milliseconds is the best scale. Note: in --minimal mode
1617 latencies are always expressed in microseconds.
Jens Axboe71bfa162006-10-25 11:08:19 +02001618 clat= Completion latency. Same names as slat, this denotes the
1619 time from submission to completion of the io pieces. For
1620 sync io, clat will usually be equal (or very close) to 0,
1621 as the time from submit to complete is basically just
1622 CPU time (io has already been done, see slat explanation).
1623 bw= Bandwidth. Same names as the xlat stats, but also includes
1624 an approximate percentage of total aggregate bandwidth
1625 this thread received in this group. This last value is
1626 only really useful if the threads in this group are on the
1627 same disk, since they are then competing for disk access.
1628cpu= CPU usage. User and system time, along with the number
Jens Axboee7823a92007-09-07 20:33:33 +02001629 of context switches this thread went through, usage of
1630 system and user time, and finally the number of major
1631 and minor page faults.
Jens Axboe71619dc2007-01-13 23:56:33 +01001632IO depths= The distribution of io depths over the job life time. The
1633 numbers are divided into powers of 2, so for example the
1634 16= entries includes depths up to that value but higher
1635 than the previous entry. In other words, it covers the
1636 range from 16 to 31.
Jens Axboe838bc702008-05-22 13:08:23 +02001637IO submit= How many pieces of IO were submitting in a single submit
1638 call. Each entry denotes that amount and below, until
1639 the previous entry - eg, 8=100% mean that we submitted
1640 anywhere in between 5-8 ios per submit call.
1641IO complete= Like the above submit number, but for completions instead.
Jens Axboe30061b92007-04-17 13:31:34 +02001642IO issued= The number of read/write requests issued, and how many
1643 of them were short.
Jens Axboeec118302007-02-17 04:38:20 +01001644IO latencies= The distribution of IO completion latencies. This is the
1645 time from when IO leaves fio and when it gets completed.
1646 The numbers follow the same pattern as the IO depths,
1647 meaning that 2=1.6% means that 1.6% of the IO completed
Jens Axboe8abdce62007-02-21 10:22:55 +01001648 within 2 msecs, 20=12.8% means that 12.8% of the IO
1649 took more than 10 msecs, but less than (or equal to) 20 msecs.
Jens Axboe71bfa162006-10-25 11:08:19 +02001650
1651After each client has been listed, the group statistics are printed. They
1652will look like this:
1653
1654Run status group 0 (all jobs):
Jens Axboeb22989b2009-07-17 22:29:23 +02001655 READ: io=64MB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
1656 WRITE: io=64MB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
Jens Axboe71bfa162006-10-25 11:08:19 +02001657
1658For each data direction, it prints:
1659
1660io= Number of megabytes io performed.
1661aggrb= Aggregate bandwidth of threads in this group.
1662minb= The minimum average bandwidth a thread saw.
1663maxb= The maximum average bandwidth a thread saw.
1664mint= The smallest runtime of the threads in that group.
1665maxt= The longest runtime of the threads in that group.
1666
1667And finally, the disk statistics are printed. They will look like this:
1668
1669Disk stats (read/write):
1670 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
1671
1672Each value is printed for both reads and writes, with reads first. The
1673numbers denote:
1674
1675ios= Number of ios performed by all groups.
1676merge= Number of merges io the io scheduler.
1677ticks= Number of ticks we kept the disk busy.
1678io_queue= Total time spent in the disk queue.
1679util= The disk utilization. A value of 100% means we kept the disk
1680 busy constantly, 50% would be a disk idling half of the time.
1681
Jens Axboe8423bd12012-04-12 09:18:38 +02001682It is also possible to get fio to dump the current output while it is
1683running, without terminating the job. To do that, send fio the USR1 signal.
Jens Axboe06464902013-04-24 20:38:54 -06001684You can also get regularly timed dumps by using the --status-interval
1685parameter, or by creating a file in /tmp named fio-dump-status. If fio
1686sees this file, it will unlink it and dump the current output status.
Jens Axboe8423bd12012-04-12 09:18:38 +02001687
Jens Axboe71bfa162006-10-25 11:08:19 +02001688
16897.0 Terse output
1690----------------
1691
1692For scripted usage where you typically want to generate tables or graphs
Jens Axboe6af019c2007-03-06 19:50:58 +01001693of the results, fio can output the results in a semicolon separated format.
Jens Axboe71bfa162006-10-25 11:08:19 +02001694The format is one long line of values, such as:
1695
David Nellans562c2d22010-09-23 08:38:17 +020016962;card0;0;0;7139336;121836;60004;1;10109;27.932460;116.933948;220;126861;3495.446807;1085.368601;226;126864;3523.635629;1089.012448;24063;99944;50.275485%;59818.274627;5540.657370;7155060;122104;60004;1;8338;29.086342;117.839068;388;128077;5032.488518;1234.785715;391;128085;5061.839412;1236.909129;23436;100928;50.287926%;59964.832030;5644.844189;14.595833%;19.394167%;123706;0;7313;0.1%;0.1%;0.1%;0.1%;0.1%;0.1%;100.0%;0.00%;0.00%;0.00%;0.00%;0.00%;0.00%;0.01%;0.02%;0.05%;0.16%;6.04%;40.40%;52.68%;0.64%;0.01%;0.00%;0.01%;0.00%;0.00%;0.00%;0.00%;0.00%
1697A description of this job goes here.
1698
1699The job description (if provided) follows on a second line.
Jens Axboe71bfa162006-10-25 11:08:19 +02001700
Jens Axboe525c2bf2010-06-30 15:22:21 +02001701To enable terse output, use the --minimal command line option. The first
1702value is the version of the terse output format. If the output has to
1703be changed for some reason, this number will be incremented by 1 to
1704signify that change.
Jens Axboe6820cb32008-09-27 12:33:53 +02001705
Jens Axboe71bfa162006-10-25 11:08:19 +02001706Split up, the format is as follows:
1707
Jens Axboe5e726d02011-10-14 08:08:10 +02001708 terse version, fio version, jobname, groupid, error
Jens Axboe71bfa162006-10-25 11:08:19 +02001709 READ status:
Jens Axboe312b4af2011-10-13 13:11:42 +02001710 Total IO (KB), bandwidth (KB/sec), IOPS, runtime (msec)
Jens Axboede196b82012-04-02 07:03:26 -07001711 Submission latency: min, max, mean, deviation (usec)
1712 Completion latency: min, max, mean, deviation (usec)
Jens Axboe1db92cb2011-10-13 13:43:36 +02001713 Completion latency percentiles: 20 fields (see below)
Jens Axboede196b82012-04-02 07:03:26 -07001714 Total latency: min, max, mean, deviation (usec)
Lucian Adrian Grijincu0d237712012-04-03 14:42:48 -06001715 Bw (KB/s): min, max, aggregate percentage of total, mean, deviation
Jens Axboe71bfa162006-10-25 11:08:19 +02001716 WRITE status:
Jens Axboe312b4af2011-10-13 13:11:42 +02001717 Total IO (KB), bandwidth (KB/sec), IOPS, runtime (msec)
Jens Axboede196b82012-04-02 07:03:26 -07001718 Submission latency: min, max, mean, deviation (usec)
1719 Completion latency: min, max, mean, deviation (usec)
Jens Axboe1db92cb2011-10-13 13:43:36 +02001720 Completion latency percentiles: 20 fields (see below)
Jens Axboede196b82012-04-02 07:03:26 -07001721 Total latency: min, max, mean, deviation (usec)
Lucian Adrian Grijincu0d237712012-04-03 14:42:48 -06001722 Bw (KB/s): min, max, aggregate percentage of total, mean, deviation
Shawn Lewis046ee302007-11-21 09:38:34 +01001723 CPU usage: user, system, context switches, major faults, minor faults
Jens Axboe22708902007-03-06 17:05:32 +01001724 IO depths: <=1, 2, 4, 8, 16, 32, >=64
David Nellans562c2d22010-09-23 08:38:17 +02001725 IO latencies microseconds: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1726 IO latencies milliseconds: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
Jens Axboef2f788d2011-10-13 14:03:52 +02001727 Disk utilization: Disk name, Read ios, write ios,
1728 Read merges, write merges,
1729 Read ticks, write ticks,
Jens Axboe3d7cd9b2011-10-18 08:31:01 +02001730 Time spent in queue, disk utilization percentage
Anatol Pomozovde8f6de2013-09-26 16:31:34 -07001731 Additional Info (dependent on continue_on_error, default off): total # errors, first error code
Bruce Cran66c098b2012-11-27 12:16:07 +00001732
Anatol Pomozovde8f6de2013-09-26 16:31:34 -07001733 Additional Info (dependent on description being set): Text description
Paul Dubs25c8b9d2011-07-21 17:26:02 +02001734
Jens Axboe1db92cb2011-10-13 13:43:36 +02001735Completion latency percentiles can be a grouping of up to 20 sets, so
1736for the terse output fio writes all of them. Each field will look like this:
1737
1738 1.00%=6112
1739
1740which is the Xth percentile, and the usec latency associated with it.
1741
Jens Axboef2f788d2011-10-13 14:03:52 +02001742For disk utilization, all disks used by fio are shown. So for each disk
1743there will be a disk utilization section.
1744
Paul Dubs25c8b9d2011-07-21 17:26:02 +02001745
17468.0 Trace file format
1747---------------------
Bruce Cran66c098b2012-11-27 12:16:07 +00001748There are two trace file format that you can encounter. The older (v1) format
Paul Dubs25c8b9d2011-07-21 17:26:02 +02001749is unsupported since version 1.20-rc3 (March 2008). It will still be described
1750below in case that you get an old trace and want to understand it.
1751
1752In any case the trace is a simple text file with a single action per line.
1753
1754
17558.1 Trace file format v1
1756------------------------
1757Each line represents a single io action in the following format:
1758
1759rw, offset, length
1760
1761where rw=0/1 for read/write, and the offset and length entries being in bytes.
1762
1763This format is not supported in Fio versions => 1.20-rc3.
1764
1765
17668.2 Trace file format v2
1767------------------------
1768The second version of the trace file format was added in Fio version 1.17.
1769It allows to access more then one file per trace and has a bigger set of
1770possible file actions.
1771
1772The first line of the trace file has to be:
1773
1774fio version 2 iolog
1775
1776Following this can be lines in two different formats, which are described below.
1777
1778The file management format:
1779
1780filename action
1781
1782The filename is given as an absolute path. The action can be one of these:
1783
1784add Add the given filename to the trace
Bruce Cran66c098b2012-11-27 12:16:07 +00001785open Open the file with the given filename. The filename has to have
Paul Dubs25c8b9d2011-07-21 17:26:02 +02001786 been added with the add action before.
1787close Close the file with the given filename. The file has to have been
1788 opened before.
1789
1790
1791The file io action format:
1792
1793filename action offset length
1794
1795The filename is given as an absolute path, and has to have been added and opened
Bruce Cran66c098b2012-11-27 12:16:07 +00001796before it can be used with this format. The offset and length are given in
Paul Dubs25c8b9d2011-07-21 17:26:02 +02001797bytes. The action can be one of these:
1798
1799wait Wait for 'offset' microseconds. Everything below 100 is discarded.
1800read Read 'length' bytes beginning from 'offset'
1801write Write 'length' bytes beginning from 'offset'
1802sync fsync() the file
1803datasync fdatasync() the file
1804trim trim the given file from the given 'offset' for 'length' bytes
Huadong Liuf2a2ce02013-01-30 13:22:24 +01001805
1806
18079.0 CPU idleness profiling
Jens Axboe06464902013-04-24 20:38:54 -06001808--------------------------
Huadong Liuf2a2ce02013-01-30 13:22:24 +01001809In some cases, we want to understand CPU overhead in a test. For example,
1810we test patches for the specific goodness of whether they reduce CPU usage.
1811fio implements a balloon approach to create a thread per CPU that runs at
1812idle priority, meaning that it only runs when nobody else needs the cpu.
1813By measuring the amount of work completed by the thread, idleness of each
1814CPU can be derived accordingly.
1815
1816An unit work is defined as touching a full page of unsigned characters. Mean
1817and standard deviation of time to complete an unit work is reported in "unit
1818work" section. Options can be chosen to report detailed percpu idleness or
1819overall system idleness by aggregating percpu stats.