zbd: ensure that global max open zones limit is respected
[fio.git] / HOWTO
<
CommitLineData
f80dba8d
MT
1How fio works
2-------------
3
4The first step in getting fio to simulate a desired I/O workload, is writing a
5job file describing that specific setup. A job file may contain any number of
6threads and/or files -- the typical contents of the job file is a *global*
7section defining shared parameters, and one or more job sections describing the
8jobs involved. When run, fio parses this file and sets everything up as
9described. If we break down a job from top to bottom, it contains the following
10basic parameters:
11
12`I/O type`_
13
14 Defines the I/O pattern issued to the file(s). We may only be reading
15 sequentially from this file(s), or we may be writing randomly. Or even
16 mixing reads and writes, sequentially or randomly.
17 Should we be doing buffered I/O, or direct/raw I/O?
18
19`Block size`_
20
21 In how large chunks are we issuing I/O? This may be a single value,
22 or it may describe a range of block sizes.
23
24`I/O size`_
25
26 How much data are we going to be reading/writing.
27
28`I/O engine`_
29
30 How do we issue I/O? We could be memory mapping the file, we could be
31 using regular read/write, we could be using splice, async I/O, or even
32 SG (SCSI generic sg).
33
34`I/O depth`_
35
36 If the I/O engine is async, how large a queuing depth do we want to
37 maintain?
38
39
40`Target file/device`_
41
42 How many files are we spreading the workload over.
43
44`Threads, processes and job synchronization`_
45
46 How many threads or processes should we spread this workload over.
47
48The above are the basic parameters defined for a workload, in addition there's a
49multitude of parameters that modify other aspects of how this job behaves.
50
51
52Command line options
53--------------------
54
55.. option:: --debug=type
56
f50fbdda 57 Enable verbose tracing `type` of various fio actions. May be ``all`` for all types
b034c0dd
TK
58 or individual types separated by a comma (e.g. ``--debug=file,mem`` will
59 enable file and memory debugging). Currently, additional logging is
60 available for:
f80dba8d 61
b034c0dd 62 *process*
f80dba8d 63 Dump info related to processes.
b034c0dd 64 *file*
f80dba8d 65 Dump info related to file actions.
b034c0dd 66 *io*
f80dba8d 67 Dump info related to I/O queuing.
b034c0dd 68 *mem*
f80dba8d 69 Dump info related to memory allocations.
b034c0dd 70 *blktrace*
f80dba8d 71 Dump info related to blktrace setup.
b034c0dd 72 *verify*
f80dba8d 73 Dump info related to I/O verification.
b034c0dd 74 *all*
f80dba8d 75 Enable all debug options.
b034c0dd 76 *random*
f80dba8d 77 Dump info related to random offset generation.
b034c0dd 78 *parse*
f80dba8d 79 Dump info related to option matching and parsing.
b034c0dd 80 *diskutil*
f80dba8d 81 Dump info related to disk utilization updates.
b034c0dd 82 *job:x*
f80dba8d 83 Dump info only related to job number x.
b034c0dd 84 *mutex*
f80dba8d 85 Dump info only related to mutex up/down ops.
b034c0dd 86 *profile*
f80dba8d 87 Dump info related to profile extensions.
b034c0dd 88 *time*
f80dba8d 89 Dump info related to internal time keeping.
b034c0dd 90 *net*
f80dba8d 91 Dump info related to networking connections.
b034c0dd 92 *rate*
f80dba8d 93 Dump info related to I/O rate switching.
b034c0dd 94 *compress*
f80dba8d 95 Dump info related to log compress/decompress.
a02ec45a
VF
96 *steadystate*
97 Dump info related to steadystate detection.
98 *helperthread*
99 Dump info related to the helper thread.
100 *zbd*
101 Dump info related to support for zoned block devices.
b034c0dd 102 *?* or *help*
f80dba8d
MT
103 Show available debug options.
104
105.. option:: --parse-only
106
25cd4b95 107 Parse options only, don't start any I/O.
f80dba8d 108
b9921d1a
DZ
109.. option:: --merge-blktrace-only
110
111 Merge blktraces only, don't start any I/O.
112
f80dba8d
MT
113.. option:: --output=filename
114
115 Write output to file `filename`.
116
f50fbdda 117.. option:: --output-format=format
b8f7e412 118
f50fbdda 119 Set the reporting `format` to `normal`, `terse`, `json`, or `json+`. Multiple
b8f7e412
TK
120 formats can be selected, separated by a comma. `terse` is a CSV based
121 format. `json+` is like `json`, except it adds a full dump of the latency
122 buckets.
123
f80dba8d
MT
124.. option:: --bandwidth-log
125
126 Generate aggregate bandwidth logs.
127
128.. option:: --minimal
129
130 Print statistics in a terse, semicolon-delimited format.
131
132.. option:: --append-terse
133
b034c0dd
TK
134 Print statistics in selected mode AND terse, semicolon-delimited format.
135 **Deprecated**, use :option:`--output-format` instead to select multiple
136 formats.
f80dba8d 137
f50fbdda 138.. option:: --terse-version=version
f80dba8d 139
f50fbdda 140 Set terse `version` output format (default 3, or 2 or 4 or 5).
f80dba8d
MT
141
142.. option:: --version
143
b8f7e412 144 Print version information and exit.
f80dba8d
MT
145
146.. option:: --help
147
113f0e7c 148 Print a summary of the command line options and exit.
f80dba8d
MT
149
150.. option:: --cpuclock-test
151
152 Perform test and validation of internal CPU clock.
153
113f0e7c 154.. option:: --crctest=[test]
f80dba8d 155
b034c0dd
TK
156 Test the speed of the built-in checksumming functions. If no argument is
157 given, all of them are tested. Alternatively, a comma separated list can
158 be passed, in which case the given ones are tested.
f80dba8d
MT
159
160.. option:: --cmdhelp=command
161
162 Print help information for `command`. May be ``all`` for all commands.
163
164.. option:: --enghelp=[ioengine[,command]]
165
f50fbdda
TK
166 List all commands defined by `ioengine`, or print help for `command`
167 defined by `ioengine`. If no `ioengine` is given, list all
b034c0dd 168 available ioengines.
f80dba8d
MT
169
170.. option:: --showcmd=jobfile
171
b8f7e412 172 Convert `jobfile` to a set of command-line options.
f80dba8d
MT
173
174.. option:: --readonly
175
4027b2a1
VF
176 Turn on safety read-only checks, preventing writes and trims. The
177 ``--readonly`` option is an extra safety guard to prevent users from
178 accidentally starting a write or trim workload when that is not desired.
179 Fio will only modify the device under test if
180 `rw=write/randwrite/rw/randrw/trim/randtrim/trimwrite` is given. This
181 safety net can be used as an extra precaution.
f80dba8d
MT
182
183.. option:: --eta=when
184
b8f7e412 185 Specifies when real-time ETA estimate should be printed. `when` may be
db37d890
JA
186 `always`, `never` or `auto`. `auto` is the default, it prints ETA
187 when requested if the output is a TTY. `always` disregards the output
188 type, and prints ETA when requested. `never` never prints ETA.
189
190.. option:: --eta-interval=time
191
192 By default, fio requests client ETA status roughly every second. With
193 this option, the interval is configurable. Fio imposes a minimum
194 allowed time to avoid flooding the console, less than 250 msec is
195 not supported.
f80dba8d
MT
196
197.. option:: --eta-newline=time
198
947e0fe0
SW
199 Force a new line for every `time` period passed. When the unit is omitted,
200 the value is interpreted in seconds.
f80dba8d
MT
201
202.. option:: --status-interval=time
203
aa6cb459
VF
204 Force a full status dump of cumulative (from job start) values at `time`
205 intervals. This option does *not* provide per-period measurements. So
206 values such as bandwidth are running averages. When the time unit is omitted,
c1f4de8a
JA
207 `time` is interpreted in seconds. Note that using this option with
208 ``--output-format=json`` will yield output that technically isn't valid
209 json, since the output will be collated sets of valid json. It will need
210 to be split into valid sets of json after the run.
f80dba8d
MT
211
212.. option:: --section=name
213
b034c0dd
TK
214 Only run specified section `name` in job file. Multiple sections can be specified.
215 The ``--section`` option allows one to combine related jobs into one file.
216 E.g. one job file could define light, moderate, and heavy sections. Tell
217 fio to run only the "heavy" section by giving ``--section=heavy``
218 command line option. One can also specify the "write" operations in one
219 section and "verify" operation in another section. The ``--section`` option
220 only applies to job sections. The reserved *global* section is always
221 parsed and used.
f80dba8d
MT
222
223.. option:: --alloc-size=kb
224
4a419903
VF
225 Allocate additional internal smalloc pools of size `kb` in KiB. The
226 ``--alloc-size`` option increases shared memory set aside for use by fio.
b034c0dd
TK
227 If running large jobs with randommap enabled, fio can run out of memory.
228 Smalloc is an internal allocator for shared structures from a fixed size
229 memory pool and can grow to 16 pools. The pool size defaults to 16MiB.
f80dba8d 230
b034c0dd
TK
231 NOTE: While running :file:`.fio_smalloc.*` backing store files are visible
232 in :file:`/tmp`.
f80dba8d
MT
233
234.. option:: --warnings-fatal
235
b034c0dd
TK
236 All fio parser warnings are fatal, causing fio to exit with an
237 error.
f80dba8d
MT
238
239.. option:: --max-jobs=nr
240
f50fbdda 241 Set the maximum number of threads/processes to support to `nr`.
818322cc 242 NOTE: On Linux, it may be necessary to increase the shared-memory
71aa48eb 243 limit (:file:`/proc/sys/kernel/shmmax`) if fio runs into errors while
818322cc 244 creating jobs.
f80dba8d
MT
245
246.. option:: --server=args
247
b034c0dd
TK
248 Start a backend server, with `args` specifying what to listen to.
249 See `Client/Server`_ section.
f80dba8d
MT
250
251.. option:: --daemonize=pidfile
252
b034c0dd 253 Background a fio server, writing the pid to the given `pidfile` file.
f80dba8d
MT
254
255.. option:: --client=hostname
256
f50fbdda 257 Instead of running the jobs locally, send and run them on the given `hostname`
71aa48eb 258 or set of `hostname`\s. See `Client/Server`_ section.
f80dba8d
MT
259
260.. option:: --remote-config=file
261
f50fbdda 262 Tell fio server to load this local `file`.
f80dba8d
MT
263
264.. option:: --idle-prof=option
265
b8f7e412 266 Report CPU idleness. `option` is one of the following:
113f0e7c
SW
267
268 **calibrate**
269 Run unit work calibration only and exit.
270
271 **system**
272 Show aggregate system idleness and unit work.
273
274 **percpu**
275 As **system** but also show per CPU idleness.
f80dba8d
MT
276
277.. option:: --inflate-log=log
278
f50fbdda 279 Inflate and output compressed `log`.
f80dba8d
MT
280
281.. option:: --trigger-file=file
282
f50fbdda 283 Execute trigger command when `file` exists.
f80dba8d 284
f50fbdda 285.. option:: --trigger-timeout=time
f80dba8d 286
f50fbdda 287 Execute trigger at this `time`.
f80dba8d 288
f50fbdda 289.. option:: --trigger=command
f80dba8d 290
f50fbdda 291 Set this `command` as local trigger.
f80dba8d 292
f50fbdda 293.. option:: --trigger-remote=command
f80dba8d 294
f50fbdda 295 Set this `command` as remote trigger.
f80dba8d
MT
296
297.. option:: --aux-path=path
298
f4401bf8
SW
299 Use the directory specified by `path` for generated state files instead
300 of the current working directory.
f80dba8d
MT
301
302Any parameters following the options will be assumed to be job files, unless
303they match a job file parameter. Multiple job files can be listed and each job
304file will be regarded as a separate group. Fio will :option:`stonewall`
305execution between each group.
306
307
308Job file format
309---------------
310
311As previously described, fio accepts one or more job files describing what it is
312supposed to do. The job file format is the classic ini file, where the names
c60ebc45 313enclosed in [] brackets define the job name. You are free to use any ASCII name
f80dba8d
MT
314you want, except *global* which has special meaning. Following the job name is
315a sequence of zero or more parameters, one per line, that define the behavior of
316the job. If the first character in a line is a ';' or a '#', the entire line is
317discarded as a comment.
318
319A *global* section sets defaults for the jobs described in that file. A job may
320override a *global* section parameter, and a job file may even have several
321*global* sections if so desired. A job is only affected by a *global* section
322residing above it.
323
f50fbdda
TK
324The :option:`--cmdhelp` option also lists all options. If used with a `command`
325argument, :option:`--cmdhelp` will detail the given `command`.
f80dba8d
MT
326
327See the `examples/` directory for inspiration on how to write job files. Note
328the copyright and license requirements currently apply to `examples/` files.
329
330So let's look at a really simple job file that defines two processes, each
331randomly reading from a 128MiB file:
332
333.. code-block:: ini
334
335 ; -- start job file --
336 [global]
337 rw=randread
338 size=128m
339
340 [job1]
341
342 [job2]
343
344 ; -- end job file --
345
346As you can see, the job file sections themselves are empty as all the described
347parameters are shared. As no :option:`filename` option is given, fio makes up a
348`filename` for each of the jobs as it sees fit. On the command line, this job
349would look as follows::
350
351$ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
352
353
354Let's look at an example that has a number of processes writing randomly to
355files:
356
357.. code-block:: ini
358
359 ; -- start job file --
360 [random-writers]
361 ioengine=libaio
362 iodepth=4
363 rw=randwrite
364 bs=32k
365 direct=0
366 size=64m
367 numjobs=4
368 ; -- end job file --
369
370Here we have no *global* section, as we only have one job defined anyway. We
371want to use async I/O here, with a depth of 4 for each file. We also increased
372the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical
373jobs. The result is 4 processes each randomly writing to their own 64MiB
374file. Instead of using the above job file, you could have given the parameters
375on the command line. For this case, you would specify::
376
377$ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
378
379When fio is utilized as a basis of any reasonably large test suite, it might be
380desirable to share a set of standardized settings across multiple job files.
381Instead of copy/pasting such settings, any section may pull in an external
382:file:`filename.fio` file with *include filename* directive, as in the following
383example::
384
385 ; -- start job file including.fio --
386 [global]
387 filename=/tmp/test
388 filesize=1m
389 include glob-include.fio
390
391 [test]
392 rw=randread
393 bs=4k
394 time_based=1
395 runtime=10
396 include test-include.fio
397 ; -- end job file including.fio --
398
399.. code-block:: ini
400
401 ; -- start job file glob-include.fio --
402 thread=1
403 group_reporting=1
404 ; -- end job file glob-include.fio --
405
406.. code-block:: ini
407
408 ; -- start job file test-include.fio --
409 ioengine=libaio
410 iodepth=4
411 ; -- end job file test-include.fio --
412
413Settings pulled into a section apply to that section only (except *global*
414section). Include directives may be nested in that any included file may contain
415further include directive(s). Include files may not contain [] sections.
416
417
418Environment variables
419~~~~~~~~~~~~~~~~~~~~~
420
421Fio also supports environment variable expansion in job files. Any sub-string of
422the form ``${VARNAME}`` as part of an option value (in other words, on the right
423of the '='), will be expanded to the value of the environment variable called
424`VARNAME`. If no such environment variable is defined, or `VARNAME` is the
425empty string, the empty string will be substituted.
426
427As an example, let's look at a sample fio invocation and job file::
428
429$ SIZE=64m NUMJOBS=4 fio jobfile.fio
430
431.. code-block:: ini
432
433 ; -- start job file --
434 [random-writers]
435 rw=randwrite
436 size=${SIZE}
437 numjobs=${NUMJOBS}
438 ; -- end job file --
439
440This will expand to the following equivalent job file at runtime:
441
442.. code-block:: ini
443
444 ; -- start job file --
445 [random-writers]
446 rw=randwrite
447 size=64m
448 numjobs=4
449 ; -- end job file --
450
451Fio ships with a few example job files, you can also look there for inspiration.
452
453Reserved keywords
454~~~~~~~~~~~~~~~~~
455
456Additionally, fio has a set of reserved keywords that will be replaced
457internally with the appropriate value. Those keywords are:
458
459**$pagesize**
460
461 The architecture page size of the running system.
462
463**$mb_memory**
464
465 Megabytes of total memory in the system.
466
467**$ncpus**
468
469 Number of online available CPUs.
470
471These can be used on the command line or in the job file, and will be
472automatically substituted with the current system values when the job is
473run. Simple math is also supported on these keywords, so you can perform actions
474like::
475
b034c0dd 476 size=8*$mb_memory
f80dba8d
MT
477
478and get that properly expanded to 8 times the size of memory in the machine.
479
480
481Job file parameters
482-------------------
483
484This section describes in details each parameter associated with a job. Some
485parameters take an option of a given type, such as an integer or a
486string. Anywhere a numeric value is required, an arithmetic expression may be
487used, provided it is surrounded by parentheses. Supported operators are:
488
489 - addition (+)
490 - subtraction (-)
491 - multiplication (*)
492 - division (/)
493 - modulus (%)
494 - exponentiation (^)
495
496For time values in expressions, units are microseconds by default. This is
497different than for time values not in expressions (not enclosed in
498parentheses). The following types are used:
499
500
501Parameter types
502~~~~~~~~~~~~~~~
503
504**str**
b034c0dd 505 String: A sequence of alphanumeric characters.
f80dba8d
MT
506
507**time**
008d0feb
SW
508 Integer with possible time suffix. Without a unit value is interpreted as
509 seconds unless otherwise specified. Accepts a suffix of 'd' for days, 'h' for
510 hours, 'm' for minutes, 's' for seconds, 'ms' (or 'msec') for milliseconds and
511 'us' (or 'usec') for microseconds. For example, use 10m for 10 minutes.
f80dba8d
MT
512
513.. _int:
514
515**int**
516 Integer. A whole number value, which may contain an integer prefix
517 and an integer suffix:
518
b034c0dd 519 [*integer prefix*] **number** [*integer suffix*]
f80dba8d
MT
520
521 The optional *integer prefix* specifies the number's base. The default
522 is decimal. *0x* specifies hexadecimal.
523
524 The optional *integer suffix* specifies the number's units, and includes an
525 optional unit prefix and an optional unit. For quantities of data, the
947e0fe0
SW
526 default unit is bytes. For quantities of time, the default unit is seconds
527 unless otherwise specified.
f80dba8d 528
9207a0cb 529 With :option:`kb_base`\=1000, fio follows international standards for unit
f80dba8d
MT
530 prefixes. To specify power-of-10 decimal values defined in the
531 International System of Units (SI):
532
eccce61a
TK
533 * *K* -- means kilo (K) or 1000
534 * *M* -- means mega (M) or 1000**2
535 * *G* -- means giga (G) or 1000**3
536 * *T* -- means tera (T) or 1000**4
537 * *P* -- means peta (P) or 1000**5
f80dba8d
MT
538
539 To specify power-of-2 binary values defined in IEC 80000-13:
540
eccce61a
TK
541 * *Ki* -- means kibi (Ki) or 1024
542 * *Mi* -- means mebi (Mi) or 1024**2
543 * *Gi* -- means gibi (Gi) or 1024**3
544 * *Ti* -- means tebi (Ti) or 1024**4
545 * *Pi* -- means pebi (Pi) or 1024**5
f80dba8d 546
193aaf6a
G
547 For Zone Block Device Mode:
548 * *z* -- means Zone
549
9207a0cb 550 With :option:`kb_base`\=1024 (the default), the unit prefixes are opposite
f80dba8d
MT
551 from those specified in the SI and IEC 80000-13 standards to provide
552 compatibility with old scripts. For example, 4k means 4096.
553
554 For quantities of data, an optional unit of 'B' may be included
b8f7e412 555 (e.g., 'kB' is the same as 'k').
f80dba8d
MT
556
557 The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega,
558 not milli). 'b' and 'B' both mean byte, not bit.
559
9207a0cb 560 Examples with :option:`kb_base`\=1000:
f80dba8d
MT
561
562 * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB
563 * *1 MiB*: 1048576, 1mi, 1024ki
564 * *1 MB*: 1000000, 1m, 1000k
565 * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi
566 * *1 TB*: 1000000000, 1t, 1000m, 1000000k
567
9207a0cb 568 Examples with :option:`kb_base`\=1024 (default):
f80dba8d
MT
569
570 * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
571 * *1 MiB*: 1048576, 1m, 1024k
572 * *1 MB*: 1000000, 1mi, 1000ki
573 * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m
574 * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki
575
576 To specify times (units are not case sensitive):
577
578 * *D* -- means days
579 * *H* -- means hours
4502cb42 580 * *M* -- means minutes
f80dba8d
MT
581 * *s* -- or sec means seconds (default)
582 * *ms* -- or *msec* means milliseconds
583 * *us* -- or *usec* means microseconds
584
585 If the option accepts an upper and lower range, use a colon ':' or
586 minus '-' to separate such values. See :ref:`irange <irange>`.
4502cb42
SW
587 If the lower value specified happens to be larger than the upper value
588 the two values are swapped.
f80dba8d
MT
589
590.. _bool:
591
592**bool**
593 Boolean. Usually parsed as an integer, however only defined for
594 true and false (1 and 0).
595
596.. _irange:
597
598**irange**
599 Integer range with suffix. Allows value range to be given, such as
c60ebc45 600 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the
f80dba8d
MT
601 option allows two sets of ranges, they can be specified with a ',' or '/'
602 delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`.
603
604**float_list**
605 A list of floating point numbers, separated by a ':' character.
606
f5c3bcf2
TK
607With the above in mind, here follows the complete list of fio job parameters.
608
f80dba8d
MT
609
610Units
611~~~~~
612
613.. option:: kb_base=int
614
615 Select the interpretation of unit prefixes in input parameters.
616
617 **1000**
618 Inputs comply with IEC 80000-13 and the International
619 System of Units (SI). Use:
620
621 - power-of-2 values with IEC prefixes (e.g., KiB)
622 - power-of-10 values with SI prefixes (e.g., kB)
623
624 **1024**
625 Compatibility mode (default). To avoid breaking old scripts:
626
627 - power-of-2 values with SI prefixes
628 - power-of-10 values with IEC prefixes
629
630 See :option:`bs` for more details on input parameters.
631
632 Outputs always use correct prefixes. Most outputs include both
633 side-by-side, like::
634
635 bw=2383.3kB/s (2327.4KiB/s)
636
637 If only one value is reported, then kb_base selects the one to use:
638
639 **1000** -- SI prefixes
640
641 **1024** -- IEC prefixes
642
643.. option:: unit_base=int
644
645 Base unit for reporting. Allowed values are:
646
647 **0**
648 Use auto-detection (default).
649 **8**
650 Byte based.
651 **1**
652 Bit based.
653
654
f80dba8d
MT
655Job description
656~~~~~~~~~~~~~~~
657
658.. option:: name=str
659
660 ASCII name of the job. This may be used to override the name printed by fio
661 for this job. Otherwise the job name is used. On the command line this
662 parameter has the special purpose of also signaling the start of a new job.
663
664.. option:: description=str
665
666 Text description of the job. Doesn't do anything except dump this text
667 description when this job is run. It's not parsed.
668
669.. option:: loops=int
670
671 Run the specified number of iterations of this job. Used to repeat the same
672 workload a given number of times. Defaults to 1.
673
674.. option:: numjobs=int
675
79591fa9
TK
676 Create the specified number of clones of this job. Each clone of job
677 is spawned as an independent thread or process. May be used to setup a
f80dba8d
MT
678 larger number of threads/processes doing the same thing. Each thread is
679 reported separately; to see statistics for all clones as a whole, use
680 :option:`group_reporting` in conjunction with :option:`new_group`.
a47b697c 681 See :option:`--max-jobs`. Default: 1.
f80dba8d
MT
682
683
684Time related parameters
685~~~~~~~~~~~~~~~~~~~~~~~
686
687.. option:: runtime=time
688
f75ede1d 689 Tell fio to terminate processing after the specified period of time. It
f80dba8d 690 can be quite hard to determine for how long a specified job will run, so
f75ede1d 691 this parameter is handy to cap the total runtime to a given time. When
804c0839 692 the unit is omitted, the value is interpreted in seconds.
f80dba8d
MT
693
694.. option:: time_based
695
696 If set, fio will run for the duration of the :option:`runtime` specified
697 even if the file(s) are completely read or written. It will simply loop over
698 the same workload as many times as the :option:`runtime` allows.
699
a881438b 700.. option:: startdelay=irange(time)
f80dba8d 701
947e0fe0
SW
702 Delay the start of job for the specified amount of time. Can be a single
703 value or a range. When given as a range, each thread will choose a value
704 randomly from within the range. Value is in seconds if a unit is omitted.
f80dba8d
MT
705
706.. option:: ramp_time=time
707
708 If set, fio will run the specified workload for this amount of time before
709 logging any performance numbers. Useful for letting performance settle
710 before logging results, thus minimizing the runtime required for stable
711 results. Note that the ``ramp_time`` is considered lead in time for a job,
712 thus it will increase the total runtime if a special timeout or
f75ede1d
SW
713 :option:`runtime` is specified. When the unit is omitted, the value is
714 given in seconds.
f80dba8d
MT
715
716.. option:: clocksource=str
717
718 Use the given clocksource as the base of timing. The supported options are:
719
720 **gettimeofday**
721 :manpage:`gettimeofday(2)`
722
723 **clock_gettime**
724 :manpage:`clock_gettime(2)`
725
726 **cpu**
727 Internal CPU clock source
728
729 cpu is the preferred clocksource if it is reliable, as it is very fast (and
730 fio is heavy on time calls). Fio will automatically use this clocksource if
731 it's supported and considered reliable on the system it is running on,
732 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
733 means supporting TSC Invariant.
734
735.. option:: gtod_reduce=bool
736
737 Enable all of the :manpage:`gettimeofday(2)` reducing options
f75ede1d 738 (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus
f80dba8d
MT
739 reduce precision of the timeout somewhat to really shrink the
740 :manpage:`gettimeofday(2)` call count. With this option enabled, we only do
741 about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all
742 time keeping was enabled.
743
744.. option:: gtod_cpu=int
745
746 Sometimes it's cheaper to dedicate a single thread of execution to just
747 getting the current time. Fio (and databases, for instance) are very
748 intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set
749 one CPU aside for doing nothing but logging current time to a shared memory
750 location. Then the other threads/processes that run I/O workloads need only
751 copy that segment, instead of entering the kernel with a
752 :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time
753 calls will be excluded from other uses. Fio will manually clear it from the
754 CPU mask of other jobs.
755
756
757Target file/device
758~~~~~~~~~~~~~~~~~~
759
760.. option:: directory=str
761
762 Prefix filenames with this directory. Used to place files in a different
763 location than :file:`./`. You can specify a number of directories by
764 separating the names with a ':' character. These directories will be
02dd2689 765 assigned equally distributed to job clones created by :option:`numjobs` as
f80dba8d
MT
766 long as they are using generated filenames. If specific `filename(s)` are
767 set fio will use the first listed directory, and thereby matching the
f4401bf8
SW
768 `filename` semantic (which generates a file for each clone if not
769 specified, but lets all clones use the same file if set).
f80dba8d 770
3b803fe1
SW
771 See the :option:`filename` option for information on how to escape "``:``"
772 characters within the directory path itself.
f80dba8d 773
f4401bf8
SW
774 Note: To control the directory fio will use for internal state files
775 use :option:`--aux-path`.
776
f80dba8d
MT
777.. option:: filename=str
778
779 Fio normally makes up a `filename` based on the job name, thread number, and
02dd2689
SW
780 file number (see :option:`filename_format`). If you want to share files
781 between threads in a job or several
79591fa9
TK
782 jobs with fixed file paths, specify a `filename` for each of them to override
783 the default. If the ioengine is file based, you can specify a number of files
784 by separating the names with a ':' colon. So if you wanted a job to open
785 :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use
786 ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is
787 specified, :option:`nrfiles` is ignored. The size of regular files specified
02dd2689 788 by this option will be :option:`size` divided by number of files unless an
79591fa9
TK
789 explicit size is specified by :option:`filesize`.
790
3b803fe1 791 Each colon in the wanted path must be escaped with a ``\``
02dd2689
SW
792 character. For instance, if the path is :file:`/dev/dsk/foo@3,0:c` then you
793 would use ``filename=/dev/dsk/foo@3,0\:c`` and if the path is
3b803fe1 794 :file:`F:\\filename` then you would use ``filename=F\:\filename``.
02dd2689 795
f80dba8d
MT
796 On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for
797 the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc.
798 Note: Windows and FreeBSD prevent write access to areas
02dd2689
SW
799 of the disk containing in-use data (e.g. filesystems).
800
801 The filename "`-`" is a reserved name, meaning *stdin* or *stdout*. Which
802 of the two depends on the read/write direction set.
f80dba8d
MT
803
804.. option:: filename_format=str
805
806 If sharing multiple files between jobs, it is usually necessary to have fio
807 generate the exact names that you want. By default, fio will name a file
808 based on the default file format specification of
809 :file:`jobname.jobnumber.filenumber`. With this option, that can be
810 customized. Fio will recognize and replace the following keywords in this
811 string:
812
813 **$jobname**
814 The name of the worker thread or process.
8d53c5f8
TG
815 **$clientuid**
816 IP of the fio process when using client/server mode.
f80dba8d
MT
817 **$jobnum**
818 The incremental number of the worker thread or process.
819 **$filenum**
820 The incremental number of the file for that worker thread or
821 process.
822
823 To have dependent jobs share a set of files, this option can be set to have
824 fio generate filenames that are shared between the two. For instance, if
825 :file:`testfiles.$filenum` is specified, file number 4 for any job will be
826 named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum`
827 will be used if no other format specifier is given.
828
645943c0
JB
829 If you specify a path then the directories will be created up to the
830 main directory for the file. So for example if you specify
831 ``filename_format=a/b/c/$jobnum`` then the directories a/b/c will be
832 created before the file setup part of the job. If you specify
833 :option:`directory` then the path will be relative that directory,
834 otherwise it is treated as the absolute path.
835
f80dba8d
MT
836.. option:: unique_filename=bool
837
838 To avoid collisions between networked clients, fio defaults to prefixing any
839 generated filenames (with a directory specified) with the source of the
840 client connecting. To disable this behavior, set this option to 0.
841
842.. option:: opendir=str
843
844 Recursively open any files below directory `str`.
845
846.. option:: lockfile=str
847
848 Fio defaults to not locking any files before it does I/O to them. If a file
849 or file descriptor is shared, fio can serialize I/O to that file to make the
850 end result consistent. This is usual for emulating real workloads that share
851 files. The lock modes are:
852
853 **none**
854 No locking. The default.
855 **exclusive**
856 Only one thread or process may do I/O at a time, excluding all
857 others.
858 **readwrite**
859 Read-write locking on the file. Many readers may
860 access the file at the same time, but writes get exclusive access.
861
862.. option:: nrfiles=int
863
79591fa9
TK
864 Number of files to use for this job. Defaults to 1. The size of files
865 will be :option:`size` divided by this unless explicit size is specified by
866 :option:`filesize`. Files are created for each thread separately, and each
867 file will have a file number within its name by default, as explained in
868 :option:`filename` section.
869
f80dba8d
MT
870
871.. option:: openfiles=int
872
873 Number of files to keep open at the same time. Defaults to the same as
874 :option:`nrfiles`, can be set smaller to limit the number simultaneous
875 opens.
876
877.. option:: file_service_type=str
878
879 Defines how fio decides which file from a job to service next. The following
880 types are defined:
881
882 **random**
883 Choose a file at random.
884
885 **roundrobin**
886 Round robin over opened files. This is the default.
887
888 **sequential**
889 Finish one file before moving on to the next. Multiple files can
f50fbdda 890 still be open depending on :option:`openfiles`.
f80dba8d
MT
891
892 **zipf**
c60ebc45 893 Use a *Zipf* distribution to decide what file to access.
f80dba8d
MT
894
895 **pareto**
c60ebc45 896 Use a *Pareto* distribution to decide what file to access.
f80dba8d 897
dd3503d3 898 **normal**
c60ebc45 899 Use a *Gaussian* (normal) distribution to decide what file to
f80dba8d
MT
900 access.
901
dd3503d3
SW
902 **gauss**
903 Alias for normal.
904
f80dba8d
MT
905 For *random*, *roundrobin*, and *sequential*, a postfix can be appended to
906 tell fio how many I/Os to issue before switching to a new file. For example,
907 specifying ``file_service_type=random:8`` would cause fio to issue
908 8 I/Os before selecting a new file at random. For the non-uniform
909 distributions, a floating point postfix can be given to influence how the
910 distribution is skewed. See :option:`random_distribution` for a description
911 of how that would work.
912
913.. option:: ioscheduler=str
914
915 Attempt to switch the device hosting the file to the specified I/O scheduler
916 before running.
917
918.. option:: create_serialize=bool
919
920 If true, serialize the file creation for the jobs. This may be handy to
921 avoid interleaving of data files, which may greatly depend on the filesystem
a47b697c 922 used and even the number of processors in the system. Default: true.
f80dba8d
MT
923
924.. option:: create_fsync=bool
925
22413915 926 :manpage:`fsync(2)` the data file after creation. This is the default.
f80dba8d
MT
927
928.. option:: create_on_open=bool
929
730bd7d9
SW
930 If true, don't pre-create files but allow the job's open() to create a file
931 when it's time to do I/O. Default: false -- pre-create all necessary files
932 when the job starts.
f80dba8d
MT
933
934.. option:: create_only=bool
935
936 If true, fio will only run the setup phase of the job. If files need to be
4502cb42 937 laid out or updated on disk, only that will be done -- the actual job contents
a47b697c 938 are not executed. Default: false.
f80dba8d
MT
939
940.. option:: allow_file_create=bool
941
730bd7d9
SW
942 If true, fio is permitted to create files as part of its workload. If this
943 option is false, then fio will error out if
f80dba8d
MT
944 the files it needs to use don't already exist. Default: true.
945
946.. option:: allow_mounted_write=bool
947
c60ebc45 948 If this isn't set, fio will abort jobs that are destructive (e.g. that write)
f80dba8d
MT
949 to what appears to be a mounted device or partition. This should help catch
950 creating inadvertently destructive tests, not realizing that the test will
b1db0375
TK
951 destroy data on the mounted file system. Note that some platforms don't allow
952 writing against a mounted device regardless of this option. Default: false.
f80dba8d
MT
953
954.. option:: pre_read=bool
955
956 If this is given, files will be pre-read into memory before starting the
957 given I/O operation. This will also clear the :option:`invalidate` flag,
958 since it is pointless to pre-read and then drop the cache. This will only
959 work for I/O engines that are seek-able, since they allow you to read the
a47b697c
SW
960 same data multiple times. Thus it will not work on non-seekable I/O engines
961 (e.g. network, splice). Default: false.
f80dba8d
MT
962
963.. option:: unlink=bool
964
965 Unlink the job files when done. Not the default, as repeated runs of that
a47b697c
SW
966 job would then waste time recreating the file set again and again. Default:
967 false.
f80dba8d
MT
968
969.. option:: unlink_each_loop=bool
970
a47b697c 971 Unlink job files after each iteration or loop. Default: false.
f80dba8d 972
7b865a2f
BVA
973.. option:: zonemode=str
974
975 Accepted values are:
976
977 **none**
b8dd9750
HH
978 The :option:`zonerange`, :option:`zonesize`,
979 :option `zonecapacity` and option:`zoneskip`
980 parameters are ignored.
7b865a2f
BVA
981 **strided**
982 I/O happens in a single zone until
983 :option:`zonesize` bytes have been transferred.
984 After that number of bytes has been
985 transferred processing of the next zone
b8dd9750 986 starts. :option `zonecapacity` is ignored.
7b865a2f
BVA
987 **zbd**
988 Zoned block device mode. I/O happens
989 sequentially in each zone, even if random I/O
990 has been selected. Random I/O happens across
991 all zones instead of being restricted to a
992 single zone. The :option:`zoneskip` parameter
993 is ignored. :option:`zonerange` and
994 :option:`zonesize` must be identical.
995
5faddc64 996.. option:: zonerange=int
f80dba8d 997
7b865a2f
BVA
998 Size of a single zone. See also :option:`zonesize` and
999 :option:`zoneskip`.
f80dba8d 1000
5faddc64 1001.. option:: zonesize=int
f80dba8d 1002
7b865a2f
BVA
1003 For :option:`zonemode` =strided, this is the number of bytes to
1004 transfer before skipping :option:`zoneskip` bytes. If this parameter
1005 is smaller than :option:`zonerange` then only a fraction of each zone
1006 with :option:`zonerange` bytes will be accessed. If this parameter is
1007 larger than :option:`zonerange` then each zone will be accessed
1008 multiple times before skipping to the next zone.
1009
1010 For :option:`zonemode` =zbd, this is the size of a single zone. The
1011 :option:`zonerange` parameter is ignored in this mode.
f80dba8d 1012
b8dd9750
HH
1013
1014.. option:: zonecapacity=int
1015
1016 For :option:`zonemode` =zbd, this defines the capacity of a single zone,
1017 which is the accessible area starting from the zone start address.
1018 This parameter only applies when using :option:`zonemode` =zbd in
1019 combination with regular block devices. If not specified it defaults to
1020 the zone size. If the target device is a zoned block device, the zone
1021 capacity is obtained from the device information and this option is
1022 ignored.
1023
f80dba8d
MT
1024.. option:: zoneskip=int
1025
7b865a2f
BVA
1026 For :option:`zonemode` =strided, the number of bytes to skip after
1027 :option:`zonesize` bytes of data have been transferred. This parameter
1028 must be zero for :option:`zonemode` =zbd.
f80dba8d 1029
bfbdd35b
BVA
1030.. option:: read_beyond_wp=bool
1031
1032 This parameter applies to :option:`zonemode` =zbd only.
1033
1034 Zoned block devices are block devices that consist of multiple zones.
1035 Each zone has a type, e.g. conventional or sequential. A conventional
1036 zone can be written at any offset that is a multiple of the block
1037 size. Sequential zones must be written sequentially. The position at
1038 which a write must occur is called the write pointer. A zoned block
1039 device can be either drive managed, host managed or host aware. For
1040 host managed devices the host must ensure that writes happen
1041 sequentially. Fio recognizes host managed devices and serializes
1042 writes to sequential zones for these devices.
1043
1044 If a read occurs in a sequential zone beyond the write pointer then
1045 the zoned block device will complete the read without reading any data
1046 from the storage medium. Since such reads lead to unrealistically high
1047 bandwidth and IOPS numbers fio only reads beyond the write pointer if
1048 explicitly told to do so. Default: false.
1049
59b07544
BVA
1050.. option:: max_open_zones=int
1051
1052 When running a random write test across an entire drive many more
1053 zones will be open than in a typical application workload. Hence this
1054 command line option that allows to limit the number of open zones. The
1055 number of open zones is defined as the number of zones to which write
1056 commands are issued.
1057
a7c2b6fc
BVA
1058.. option:: zone_reset_threshold=float
1059
1060 A number between zero and one that indicates the ratio of logical
1061 blocks with data to the total number of logical blocks in the test
1062 above which zones should be reset periodically.
1063
1064.. option:: zone_reset_frequency=float
1065
1066 A number between zero and one that indicates how often a zone reset
1067 should be issued if the zone reset threshold has been exceeded. A zone
1068 reset is submitted after each (1 / zone_reset_frequency) write
1069 requests. This and the previous parameter can be used to simulate
1070 garbage collection activity.
1071
f80dba8d
MT
1072
1073I/O type
1074~~~~~~~~
1075
1076.. option:: direct=bool
1077
1078 If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that
8e889110 1079 OpenBSD and ZFS on Solaris don't support direct I/O. On Windows the synchronous
f80dba8d
MT
1080 ioengines don't support direct I/O. Default: false.
1081
1082.. option:: atomic=bool
1083
1084 If value is true, attempt to use atomic direct I/O. Atomic writes are
1085 guaranteed to be stable once acknowledged by the operating system. Only
1086 Linux supports O_ATOMIC right now.
1087
1088.. option:: buffered=bool
1089
1090 If value is true, use buffered I/O. This is the opposite of the
1091 :option:`direct` option. Defaults to true.
1092
1093.. option:: readwrite=str, rw=str
1094
1095 Type of I/O pattern. Accepted values are:
1096
1097 **read**
1098 Sequential reads.
1099 **write**
1100 Sequential writes.
1101 **trim**
3740cfc8
VF
1102 Sequential trims (Linux block devices and SCSI
1103 character devices only).
f80dba8d
MT
1104 **randread**
1105 Random reads.
2831be97
SW
1106 **randwrite**
1107 Random writes.
f80dba8d 1108 **randtrim**
3740cfc8
VF
1109 Random trims (Linux block devices and SCSI
1110 character devices only).
f80dba8d
MT
1111 **rw,readwrite**
1112 Sequential mixed reads and writes.
1113 **randrw**
1114 Random mixed reads and writes.
1115 **trimwrite**
1116 Sequential trim+write sequences. Blocks will be trimmed first,
1117 then the same blocks will be written to.
1118
1119 Fio defaults to read if the option is not specified. For the mixed I/O
1120 types, the default is to split them 50/50. For certain types of I/O the
730bd7d9
SW
1121 result may still be skewed a bit, since the speed may be different.
1122
1123 It is possible to specify the number of I/Os to do before getting a new
1124 offset by appending ``:<nr>`` to the end of the string given. For a
f80dba8d
MT
1125 random read, it would look like ``rw=randread:8`` for passing in an offset
1126 modifier with a value of 8. If the suffix is used with a sequential I/O
730bd7d9
SW
1127 pattern, then the *<nr>* value specified will be **added** to the generated
1128 offset for each I/O turning sequential I/O into sequential I/O with holes.
1129 For instance, using ``rw=write:4k`` will skip 4k for every write. Also see
1130 the :option:`rw_sequencer` option.
f80dba8d
MT
1131
1132.. option:: rw_sequencer=str
1133
1134 If an offset modifier is given by appending a number to the ``rw=<str>``
1135 line, then this option controls how that number modifies the I/O offset
1136 being generated. Accepted values are:
1137
1138 **sequential**
1139 Generate sequential offset.
1140 **identical**
1141 Generate the same offset.
1142
1143 ``sequential`` is only useful for random I/O, where fio would normally
c60ebc45 1144 generate a new random offset for every I/O. If you append e.g. 8 to randread,
007c7be9
SW
1145 you would get a new random offset for every 8 I/Os. The result would be a
1146 seek for only every 8 I/Os, instead of for every I/O. Use ``rw=randread:8``
f80dba8d
MT
1147 to specify that. As sequential I/O is already sequential, setting
1148 ``sequential`` for that would not result in any differences. ``identical``
1149 behaves in a similar fashion, except it sends the same offset 8 number of
1150 times before generating a new offset.
1151
5cb8a8cd 1152.. option:: unified_rw_reporting=str
f80dba8d
MT
1153
1154 Fio normally reports statistics on a per data direction basis, meaning that
5cb8a8cd
BP
1155 reads, writes, and trims are accounted and reported separately. This option
1156 determines whether fio reports the results normally, summed together, or as
1157 both options.
1158 Accepted values are:
1159
1160 **none**
1161 Normal statistics reporting.
1162
1163 **mixed**
1164 Statistics are summed per data direction and reported together.
1165
1166 **both**
1167 Statistics are reported normally, followed by the mixed statistics.
1168
1169 **0**
1170 Backward-compatible alias for **none**.
1171
1172 **1**
1173 Backward-compatible alias for **mixed**.
9326926b 1174
5cb8a8cd
BP
1175 **2**
1176 Alias for **both**.
f80dba8d
MT
1177
1178.. option:: randrepeat=bool
1179
1180 Seed the random number generator used for random I/O patterns in a
1181 predictable way so the pattern is repeatable across runs. Default: true.
1182
1183.. option:: allrandrepeat=bool
1184
1185 Seed all random number generators in a predictable way so results are
1186 repeatable across runs. Default: false.
1187
1188.. option:: randseed=int
1189
1190 Seed the random number generators based on this seed value, to be able to
1191 control what sequence of output is being generated. If not set, the random
1192 sequence depends on the :option:`randrepeat` setting.
1193
1194.. option:: fallocate=str
1195
1196 Whether pre-allocation is performed when laying down files.
1197 Accepted values are:
1198
1199 **none**
1200 Do not pre-allocate space.
1201
2c3e17be
SW
1202 **native**
1203 Use a platform's native pre-allocation call but fall back to
1204 **none** behavior if it fails/is not implemented.
1205
f80dba8d
MT
1206 **posix**
1207 Pre-allocate via :manpage:`posix_fallocate(3)`.
1208
1209 **keep**
1210 Pre-allocate via :manpage:`fallocate(2)` with
1211 FALLOC_FL_KEEP_SIZE set.
1212
38ca5f03
TV
1213 **truncate**
1214 Extend file to final size via :manpage:`ftruncate(2)`
1215 instead of allocating.
1216
f80dba8d
MT
1217 **0**
1218 Backward-compatible alias for **none**.
1219
1220 **1**
1221 Backward-compatible alias for **posix**.
1222
1223 May not be available on all supported platforms. **keep** is only available
2c3e17be
SW
1224 on Linux. If using ZFS on Solaris this cannot be set to **posix**
1225 because ZFS doesn't support pre-allocation. Default: **native** if any
38ca5f03
TV
1226 pre-allocation methods except **truncate** are available, **none** if not.
1227
1228 Note that using **truncate** on Windows will interact surprisingly
1229 with non-sequential write patterns. When writing to a file that has
1230 been extended by setting the end-of-file information, Windows will
1231 backfill the unwritten portion of the file up to that offset with
1232 zeroes before issuing the new write. This means that a single small
1233 write to the end of an extended file will stall until the entire
1234 file has been filled with zeroes.
f80dba8d
MT
1235
1236.. option:: fadvise_hint=str
1237
c712c97a
JA
1238 Use :manpage:`posix_fadvise(2)` or :manpage:`posix_fadvise(2)` to
1239 advise the kernel on what I/O patterns are likely to be issued.
1240 Accepted values are:
f80dba8d
MT
1241
1242 **0**
1243 Backwards-compatible hint for "no hint".
1244
1245 **1**
1246 Backwards compatible hint for "advise with fio workload type". This
1247 uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL**
1248 for a sequential workload.
1249
1250 **sequential**
1251 Advise using **FADV_SEQUENTIAL**.
1252
1253 **random**
1254 Advise using **FADV_RANDOM**.
1255
8f4b9f24 1256.. option:: write_hint=str
f80dba8d 1257
8f4b9f24
JA
1258 Use :manpage:`fcntl(2)` to advise the kernel what life time to expect
1259 from a write. Only supported on Linux, as of version 4.13. Accepted
1260 values are:
1261
1262 **none**
1263 No particular life time associated with this file.
1264
1265 **short**
1266 Data written to this file has a short life time.
1267
1268 **medium**
1269 Data written to this file has a medium life time.
1270
1271 **long**
1272 Data written to this file has a long life time.
1273
1274 **extreme**
1275 Data written to this file has a very long life time.
1276
1277 The values are all relative to each other, and no absolute meaning
1278 should be associated with them.
f80dba8d
MT
1279
1280.. option:: offset=int
1281
82dbb8cb 1282 Start I/O at the provided offset in the file, given as either a fixed size in
193aaf6a 1283 bytes, zones or a percentage. If a percentage is given, the generated offset will be
83c8b093
JF
1284 aligned to the minimum ``blocksize`` or to the value of ``offset_align`` if
1285 provided. Data before the given offset will not be touched. This
89978a6b
BW
1286 effectively caps the file size at `real_size - offset`. Can be combined with
1287 :option:`size` to constrain the start and end range of the I/O workload.
44bb1142 1288 A percentage can be specified by a number between 1 and 100 followed by '%',
193aaf6a
G
1289 for example, ``offset=20%`` to specify 20%. In ZBD mode, value can be set as
1290 number of zones using 'z'.
f80dba8d 1291
83c8b093
JF
1292.. option:: offset_align=int
1293
1294 If set to non-zero value, the byte offset generated by a percentage ``offset``
1295 is aligned upwards to this value. Defaults to 0 meaning that a percentage
1296 offset is aligned to the minimum block size.
1297
f80dba8d
MT
1298.. option:: offset_increment=int
1299
1300 If this is provided, then the real offset becomes `offset + offset_increment
1301 * thread_number`, where the thread number is a counter that starts at 0 and
1302 is incremented for each sub-job (i.e. when :option:`numjobs` option is
1303 specified). This option is useful if there are several jobs which are
1304 intended to operate on a file in parallel disjoint segments, with even
0b288ba1
VF
1305 spacing between the starting points. Percentages can be used for this option.
1306 If a percentage is given, the generated offset will be aligned to the minimum
193aaf6a
G
1307 ``blocksize`` or to the value of ``offset_align`` if provided. In ZBD mode, value can
1308 also be set as number of zones using 'z'.
f80dba8d
MT
1309
1310.. option:: number_ios=int
1311
c60ebc45 1312 Fio will normally perform I/Os until it has exhausted the size of the region
f80dba8d
MT
1313 set by :option:`size`, or if it exhaust the allocated time (or hits an error
1314 condition). With this setting, the range/size can be set independently of
c60ebc45 1315 the number of I/Os to perform. When fio reaches this number, it will exit
f80dba8d
MT
1316 normally and report status. Note that this does not extend the amount of I/O
1317 that will be done, it will only stop fio if this condition is met before
1318 other end-of-job criteria.
1319
1320.. option:: fsync=int
1321
730bd7d9
SW
1322 If writing to a file, issue an :manpage:`fsync(2)` (or its equivalent) of
1323 the dirty data for every number of blocks given. For example, if you give 32
1324 as a parameter, fio will sync the file after every 32 writes issued. If fio is
1325 using non-buffered I/O, we may not sync the file. The exception is the sg
1326 I/O engine, which synchronizes the disk cache anyway. Defaults to 0, which
1327 means fio does not periodically issue and wait for a sync to complete. Also
1328 see :option:`end_fsync` and :option:`fsync_on_close`.
f80dba8d
MT
1329
1330.. option:: fdatasync=int
1331
1332 Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and
44f668d7 1333 not metadata blocks. In Windows, FreeBSD, DragonFlyBSD or OSX there is no
730bd7d9
SW
1334 :manpage:`fdatasync(2)` so this falls back to using :manpage:`fsync(2)`.
1335 Defaults to 0, which means fio does not periodically issue and wait for a
1336 data-only sync to complete.
f80dba8d
MT
1337
1338.. option:: write_barrier=int
1339
2831be97 1340 Make every `N-th` write a barrier write.
f80dba8d 1341
f50fbdda 1342.. option:: sync_file_range=str:int
f80dba8d 1343
f50fbdda 1344 Use :manpage:`sync_file_range(2)` for every `int` number of write
f80dba8d
MT
1345 operations. Fio will track range of writes that have happened since the last
1346 :manpage:`sync_file_range(2)` call. `str` can currently be one or more of:
1347
1348 **wait_before**
1349 SYNC_FILE_RANGE_WAIT_BEFORE
1350 **write**
1351 SYNC_FILE_RANGE_WRITE
1352 **wait_after**
1353 SYNC_FILE_RANGE_WAIT_AFTER
1354
1355 So if you do ``sync_file_range=wait_before,write:8``, fio would use
1356 ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8
1357 writes. Also see the :manpage:`sync_file_range(2)` man page. This option is
1358 Linux specific.
1359
1360.. option:: overwrite=bool
1361
1362 If true, writes to a file will always overwrite existing data. If the file
1363 doesn't already exist, it will be created before the write phase begins. If
1364 the file exists and is large enough for the specified write phase, nothing
a47b697c 1365 will be done. Default: false.
f80dba8d
MT
1366
1367.. option:: end_fsync=bool
1368
a47b697c
SW
1369 If true, :manpage:`fsync(2)` file contents when a write stage has completed.
1370 Default: false.
f80dba8d
MT
1371
1372.. option:: fsync_on_close=bool
1373
1374 If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs
a47b697c
SW
1375 from :option:`end_fsync` in that it will happen on every file close, not
1376 just at the end of the job. Default: false.
f80dba8d
MT
1377
1378.. option:: rwmixread=int
1379
1380 Percentage of a mixed workload that should be reads. Default: 50.
1381
1382.. option:: rwmixwrite=int
1383
1384 Percentage of a mixed workload that should be writes. If both
1385 :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not
1386 add up to 100%, the latter of the two will be used to override the
1387 first. This may interfere with a given rate setting, if fio is asked to
1388 limit reads or writes to a certain rate. If that is the case, then the
1389 distribution may be skewed. Default: 50.
1390
a87c90fd 1391.. option:: random_distribution=str:float[:float][,str:float][,str:float]
f80dba8d
MT
1392
1393 By default, fio will use a completely uniform random distribution when asked
1394 to perform random I/O. Sometimes it is useful to skew the distribution in
1395 specific ways, ensuring that some parts of the data is more hot than others.
1396 fio includes the following distribution models:
1397
1398 **random**
1399 Uniform random distribution
1400
1401 **zipf**
1402 Zipf distribution
1403
1404 **pareto**
1405 Pareto distribution
1406
b2f4b559 1407 **normal**
c60ebc45 1408 Normal (Gaussian) distribution
f80dba8d
MT
1409
1410 **zoned**
1411 Zoned random distribution
1412
59466396
JA
1413 **zoned_abs**
1414 Zone absolute random distribution
1415
f80dba8d 1416 When using a **zipf** or **pareto** distribution, an input value is also
f50fbdda 1417 needed to define the access pattern. For **zipf**, this is the `Zipf
c60ebc45 1418 theta`. For **pareto**, it's the `Pareto power`. Fio includes a test
f50fbdda 1419 program, :command:`fio-genzipf`, that can be used visualize what the given input
f80dba8d
MT
1420 values will yield in terms of hit rates. If you wanted to use **zipf** with
1421 a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the
1422 option. If a non-uniform model is used, fio will disable use of the random
b2f4b559
SW
1423 map. For the **normal** distribution, a normal (Gaussian) deviation is
1424 supplied as a value between 0 and 100.
f80dba8d 1425
a87c90fd
AK
1426 The second, optional float is allowed for **pareto**, **zipf** and **normal** distributions.
1427 It allows to set base of distribution in non-default place, giving more control
1428 over most probable outcome. This value is in range [0-1] which maps linearly to
1429 range of possible random values.
1430 Defaults are: random for **pareto** and **zipf**, and 0.5 for **normal**.
1431 If you wanted to use **zipf** with a `theta` of 1.2 centered on 1/4 of allowed value range,
1432 you would use ``random_distibution=zipf:1.2:0.25``.
1433
f80dba8d
MT
1434 For a **zoned** distribution, fio supports specifying percentages of I/O
1435 access that should fall within what range of the file or device. For
1436 example, given a criteria of:
1437
f50fbdda
TK
1438 * 60% of accesses should be to the first 10%
1439 * 30% of accesses should be to the next 20%
1440 * 8% of accesses should be to the next 30%
1441 * 2% of accesses should be to the next 40%
f80dba8d
MT
1442
1443 we can define that through zoning of the random accesses. For the above
1444 example, the user would do::
1445
1446 random_distribution=zoned:60/10:30/20:8/30:2/40
1447
59466396
JA
1448 A **zoned_abs** distribution works exactly like the **zoned**, except
1449 that it takes absolute sizes. For example, let's say you wanted to
1450 define access according to the following criteria:
1451
1452 * 60% of accesses should be to the first 20G
1453 * 30% of accesses should be to the next 100G
1454 * 10% of accesses should be to the next 500G
1455
1456 we can define an absolute zoning distribution with:
1457
1458 random_distribution=zoned_abs=60/20G:30/100G:10/500g
1459
6a16ece8
JA
1460 For both **zoned** and **zoned_abs**, fio supports defining up to
1461 256 separate zones.
1462
59466396
JA
1463 Similarly to how :option:`bssplit` works for setting ranges and
1464 percentages of block sizes. Like :option:`bssplit`, it's possible to
1465 specify separate zones for reads, writes, and trims. If just one set
1466 is given, it'll apply to all of them. This goes for both **zoned**
1467 **zoned_abs** distributions.
f80dba8d
MT
1468
1469.. option:: percentage_random=int[,int][,int]
1470
1471 For a random workload, set how big a percentage should be random. This
1472 defaults to 100%, in which case the workload is fully random. It can be set
1473 from anywhere from 0 to 100. Setting it to 0 would make the workload fully
1474 sequential. Any setting in between will result in a random mix of sequential
1475 and random I/O, at the given percentages. Comma-separated values may be
1476 specified for reads, writes, and trims as described in :option:`blocksize`.
1477
1478.. option:: norandommap
1479
1480 Normally fio will cover every block of the file when doing random I/O. If
1481 this option is given, fio will just get a new random offset without looking
1482 at past I/O history. This means that some blocks may not be read or written,
1483 and that some blocks may be read/written more than once. If this option is
1484 used with :option:`verify` and multiple blocksizes (via :option:`bsrange`),
1485 only intact blocks are verified, i.e., partially-overwritten blocks are
47e6a6e5
SW
1486 ignored. With an async I/O engine and an I/O depth > 1, it is possible for
1487 the same block to be overwritten, which can cause verification errors. Either
1488 do not use norandommap in this case, or also use the lfsr random generator.
f80dba8d
MT
1489
1490.. option:: softrandommap=bool
1491
1492 See :option:`norandommap`. If fio runs with the random block map enabled and
1493 it fails to allocate the map, if this option is set it will continue without
1494 a random block map. As coverage will not be as complete as with random maps,
1495 this option is disabled by default.
1496
1497.. option:: random_generator=str
1498
f50fbdda 1499 Fio supports the following engines for generating I/O offsets for random I/O:
f80dba8d
MT
1500
1501 **tausworthe**
f50fbdda 1502 Strong 2^88 cycle random number generator.
f80dba8d 1503 **lfsr**
f50fbdda 1504 Linear feedback shift register generator.
f80dba8d 1505 **tausworthe64**
f50fbdda 1506 Strong 64-bit 2^258 cycle random number generator.
f80dba8d
MT
1507
1508 **tausworthe** is a strong random number generator, but it requires tracking
1509 on the side if we want to ensure that blocks are only read or written
f50fbdda 1510 once. **lfsr** guarantees that we never generate the same offset twice, and
f80dba8d 1511 it's also less computationally expensive. It's not a true random generator,
f50fbdda 1512 however, though for I/O purposes it's typically good enough. **lfsr** only
f80dba8d
MT
1513 works with single block sizes, not with workloads that use multiple block
1514 sizes. If used with such a workload, fio may read or write some blocks
1515 multiple times. The default value is **tausworthe**, unless the required
1516 space exceeds 2^32 blocks. If it does, then **tausworthe64** is
1517 selected automatically.
1518
1519
1520Block size
1521~~~~~~~~~~
1522
1523.. option:: blocksize=int[,int][,int], bs=int[,int][,int]
1524
1525 The block size in bytes used for I/O units. Default: 4096. A single value
1526 applies to reads, writes, and trims. Comma-separated values may be
1527 specified for reads, writes, and trims. A value not terminated in a comma
1528 applies to subsequent types.
1529
1530 Examples:
1531
1532 **bs=256k**
1533 means 256k for reads, writes and trims.
1534
1535 **bs=8k,32k**
1536 means 8k for reads, 32k for writes and trims.
1537
1538 **bs=8k,32k,**
1539 means 8k for reads, 32k for writes, and default for trims.
1540
1541 **bs=,8k**
1542 means default for reads, 8k for writes and trims.
1543
1544 **bs=,8k,**
b443ae44 1545 means default for reads, 8k for writes, and default for trims.
f80dba8d
MT
1546
1547.. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange]
1548
1549 A range of block sizes in bytes for I/O units. The issued I/O unit will
1550 always be a multiple of the minimum size, unless
1551 :option:`blocksize_unaligned` is set.
1552
1553 Comma-separated ranges may be specified for reads, writes, and trims as
1554 described in :option:`blocksize`.
1555
1556 Example: ``bsrange=1k-4k,2k-8k``.
1557
1558.. option:: bssplit=str[,str][,str]
1559
6a16ece8
JA
1560 Sometimes you want even finer grained control of the block sizes
1561 issued, not just an even split between them. This option allows you to
1562 weight various block sizes, so that you are able to define a specific
1563 amount of block sizes issued. The format for this option is::
f80dba8d
MT
1564
1565 bssplit=blocksize/percentage:blocksize/percentage
1566
6a16ece8
JA
1567 for as many block sizes as needed. So if you want to define a workload
1568 that has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would
1569 write::
f80dba8d
MT
1570
1571 bssplit=4k/10:64k/50:32k/40
1572
6a16ece8
JA
1573 Ordering does not matter. If the percentage is left blank, fio will
1574 fill in the remaining values evenly. So a bssplit option like this one::
f80dba8d
MT
1575
1576 bssplit=4k/50:1k/:32k/
1577
6a16ece8
JA
1578 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always
1579 add up to 100, if bssplit is given a range that adds up to more, it
1580 will error out.
f80dba8d
MT
1581
1582 Comma-separated values may be specified for reads, writes, and trims as
1583 described in :option:`blocksize`.
1584
6a16ece8
JA
1585 If you want a workload that has 50% 2k reads and 50% 4k reads, while
1586 having 90% 4k writes and 10% 8k writes, you would specify::
f80dba8d 1587
cf04b906 1588 bssplit=2k/50:4k/50,4k/90:8k/10
f80dba8d 1589
6a16ece8
JA
1590 Fio supports defining up to 64 different weights for each data
1591 direction.
1592
f80dba8d
MT
1593.. option:: blocksize_unaligned, bs_unaligned
1594
1595 If set, fio will issue I/O units with any size within
1596 :option:`blocksize_range`, not just multiples of the minimum size. This
1597 typically won't work with direct I/O, as that normally requires sector
1598 alignment.
1599
589e88b7 1600.. option:: bs_is_seq_rand=bool
f80dba8d
MT
1601
1602 If this option is set, fio will use the normal read,write blocksize settings
1603 as sequential,random blocksize settings instead. Any random read or write
1604 will use the WRITE blocksize settings, and any sequential read or write will
1605 use the READ blocksize settings.
1606
1607.. option:: blockalign=int[,int][,int], ba=int[,int][,int]
1608
1609 Boundary to which fio will align random I/O units. Default:
1610 :option:`blocksize`. Minimum alignment is typically 512b for using direct
1611 I/O, though it usually depends on the hardware block size. This option is
1612 mutually exclusive with using a random map for files, so it will turn off
1613 that option. Comma-separated values may be specified for reads, writes, and
1614 trims as described in :option:`blocksize`.
1615
1616
1617Buffers and memory
1618~~~~~~~~~~~~~~~~~~
1619
1620.. option:: zero_buffers
1621
1622 Initialize buffers with all zeros. Default: fill buffers with random data.
1623
1624.. option:: refill_buffers
1625
1626 If this option is given, fio will refill the I/O buffers on every
72592780
SW
1627 submit. Only makes sense if :option:`zero_buffers` isn't specified,
1628 naturally. Defaults to being unset i.e., the buffer is only filled at
1629 init time and the data in it is reused when possible but if any of
1630 :option:`verify`, :option:`buffer_compress_percentage` or
1631 :option:`dedupe_percentage` are enabled then `refill_buffers` is also
1632 automatically enabled.
f80dba8d
MT
1633
1634.. option:: scramble_buffers=bool
1635
1636 If :option:`refill_buffers` is too costly and the target is using data
1637 deduplication, then setting this option will slightly modify the I/O buffer
1638 contents to defeat normal de-dupe attempts. This is not enough to defeat
1639 more clever block compression attempts, but it will stop naive dedupe of
1640 blocks. Default: true.
1641
1642.. option:: buffer_compress_percentage=int
1643
72592780
SW
1644 If this is set, then fio will attempt to provide I/O buffer content
1645 (on WRITEs) that compresses to the specified level. Fio does this by
1646 providing a mix of random data followed by fixed pattern data. The
1647 fixed pattern is either zeros, or the pattern specified by
1648 :option:`buffer_pattern`. If the `buffer_pattern` option is used, it
1649 might skew the compression ratio slightly. Setting
1650 `buffer_compress_percentage` to a value other than 100 will also
1651 enable :option:`refill_buffers` in order to reduce the likelihood that
1652 adjacent blocks are so similar that they over compress when seen
1653 together. See :option:`buffer_compress_chunk` for how to set a finer or
1654 coarser granularity for the random/fixed data region. Defaults to unset
1655 i.e., buffer data will not adhere to any compression level.
f80dba8d
MT
1656
1657.. option:: buffer_compress_chunk=int
1658
72592780
SW
1659 This setting allows fio to manage how big the random/fixed data region
1660 is when using :option:`buffer_compress_percentage`. When
1661 `buffer_compress_chunk` is set to some non-zero value smaller than the
1662 block size, fio can repeat the random/fixed region throughout the I/O
1663 buffer at the specified interval (which particularly useful when
1664 bigger block sizes are used for a job). When set to 0, fio will use a
1665 chunk size that matches the block size resulting in a single
1666 random/fixed region within the I/O buffer. Defaults to 512. When the
1667 unit is omitted, the value is interpreted in bytes.
f80dba8d
MT
1668
1669.. option:: buffer_pattern=str
1670
a1554f65
SB
1671 If set, fio will fill the I/O buffers with this pattern or with the contents
1672 of a file. If not set, the contents of I/O buffers are defined by the other
1673 options related to buffer contents. The setting can be any pattern of bytes,
1674 and can be prefixed with 0x for hex values. It may also be a string, where
1675 the string must then be wrapped with ``""``. Or it may also be a filename,
1676 where the filename must be wrapped with ``''`` in which case the file is
1677 opened and read. Note that not all the file contents will be read if that
1678 would cause the buffers to overflow. So, for example::
1679
1680 buffer_pattern='filename'
1681
1682 or::
f80dba8d
MT
1683
1684 buffer_pattern="abcd"
1685
1686 or::
1687
1688 buffer_pattern=-12
1689
1690 or::
1691
1692 buffer_pattern=0xdeadface
1693
1694 Also you can combine everything together in any order::
1695
a1554f65 1696 buffer_pattern=0xdeadface"abcd"-12'filename'
f80dba8d
MT
1697
1698.. option:: dedupe_percentage=int
1699
1700 If set, fio will generate this percentage of identical buffers when
1701 writing. These buffers will be naturally dedupable. The contents of the
1702 buffers depend on what other buffer compression settings have been set. It's
1703 possible to have the individual buffers either fully compressible, or not at
72592780
SW
1704 all -- this option only controls the distribution of unique buffers. Setting
1705 this option will also enable :option:`refill_buffers` to prevent every buffer
1706 being identical.
f80dba8d
MT
1707
1708.. option:: invalidate=bool
1709
730bd7d9
SW
1710 Invalidate the buffer/page cache parts of the files to be used prior to
1711 starting I/O if the platform and file type support it. Defaults to true.
21c1b29e
TK
1712 This will be ignored if :option:`pre_read` is also specified for the
1713 same job.
f80dba8d 1714
eb9f8d7f
AF
1715.. option:: sync=str
1716
1717 Whether, and what type, of synchronous I/O to use for writes. The allowed
1718 values are:
1719
1720 **none**
1721 Do not use synchronous IO, the default.
1722
1723 **0**
1724 Same as **none**.
1725
1726 **sync**
1727 Use synchronous file IO. For the majority of I/O engines,
1728 this means using O_SYNC.
1729
1730 **1**
1731 Same as **sync**.
1732
1733 **dsync**
1734 Use synchronous data IO. For the majority of I/O engines,
1735 this means using O_DSYNC.
f80dba8d 1736
f80dba8d
MT
1737
1738.. option:: iomem=str, mem=str
1739
1740 Fio can use various types of memory as the I/O unit buffer. The allowed
1741 values are:
1742
1743 **malloc**
1744 Use memory from :manpage:`malloc(3)` as the buffers. Default memory
1745 type.
1746
1747 **shm**
1748 Use shared memory as the buffers. Allocated through
1749 :manpage:`shmget(2)`.
1750
1751 **shmhuge**
1752 Same as shm, but use huge pages as backing.
1753
1754 **mmap**
22413915 1755 Use :manpage:`mmap(2)` to allocate buffers. May either be anonymous memory, or can
f80dba8d
MT
1756 be file backed if a filename is given after the option. The format
1757 is `mem=mmap:/path/to/file`.
1758
1759 **mmaphuge**
1760 Use a memory mapped huge file as the buffer backing. Append filename
1761 after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`.
1762
1763 **mmapshared**
1764 Same as mmap, but use a MMAP_SHARED mapping.
1765
03553853
YR
1766 **cudamalloc**
1767 Use GPU memory as the buffers for GPUDirect RDMA benchmark.
f50fbdda 1768 The :option:`ioengine` must be `rdma`.
03553853 1769
f80dba8d
MT
1770 The area allocated is a function of the maximum allowed bs size for the job,
1771 multiplied by the I/O depth given. Note that for **shmhuge** and
1772 **mmaphuge** to work, the system must have free huge pages allocated. This
1773 can normally be checked and set by reading/writing
1774 :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page
1775 is 4MiB in size. So to calculate the number of huge pages you need for a
1776 given job file, add up the I/O depth of all jobs (normally one unless
1777 :option:`iodepth` is used) and multiply by the maximum bs set. Then divide
1778 that number by the huge page size. You can see the size of the huge pages in
1779 :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero
1780 number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also
1781 see :option:`hugepage-size`.
1782
1783 **mmaphuge** also needs to have hugetlbfs mounted and the file location
1784 should point there. So if it's mounted in :file:`/huge`, you would use
1785 `mem=mmaphuge:/huge/somefile`.
1786
f50fbdda 1787.. option:: iomem_align=int, mem_align=int
f80dba8d
MT
1788
1789 This indicates the memory alignment of the I/O memory buffers. Note that
1790 the given alignment is applied to the first I/O unit buffer, if using
1791 :option:`iodepth` the alignment of the following buffers are given by the
1792 :option:`bs` used. In other words, if using a :option:`bs` that is a
1793 multiple of the page sized in the system, all buffers will be aligned to
1794 this value. If using a :option:`bs` that is not page aligned, the alignment
1795 of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and
1796 :option:`bs` used.
1797
1798.. option:: hugepage-size=int
1799
1800 Defines the size of a huge page. Must at least be equal to the system
1801 setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably
1802 always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the
1803 preferred way to set this to avoid setting a non-pow-2 bad value.
1804
1805.. option:: lockmem=int
1806
1807 Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to
1808 simulate a smaller amount of memory. The amount specified is per worker.
1809
1810
1811I/O size
1812~~~~~~~~
1813
1814.. option:: size=int
1815
79591fa9
TK
1816 The total size of file I/O for each thread of this job. Fio will run until
1817 this many bytes has been transferred, unless runtime is limited by other options
1818 (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`).
1819 Fio will divide this size between the available files determined by options
1820 such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is
1821 specified by the job. If the result of division happens to be 0, the size is
c4aa2d08 1822 set to the physical size of the given files or devices if they exist.
79591fa9 1823 If this option is not specified, fio will use the full size of the given
f80dba8d
MT
1824 files or devices. If the files do not exist, size must be given. It is also
1825 possible to give size as a percentage between 1 and 100. If ``size=20%`` is
193aaf6a
G
1826 given, fio will use 20% of the full size of the given files or devices.
1827 In ZBD mode, value can also be set as number of zones using 'z'.
9d25d068
SW
1828 Can be combined with :option:`offset` to constrain the start and end range
1829 that I/O will be done within.
f80dba8d
MT
1830
1831.. option:: io_size=int, io_limit=int
1832
1833 Normally fio operates within the region set by :option:`size`, which means
1834 that the :option:`size` option sets both the region and size of I/O to be
1835 performed. Sometimes that is not what you want. With this option, it is
1836 possible to define just the amount of I/O that fio should do. For instance,
1837 if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio
1838 will perform I/O within the first 20GiB but exit when 5GiB have been
1839 done. The opposite is also possible -- if :option:`size` is set to 20GiB,
1840 and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within
1841 the 0..20GiB region.
1842
7fdd97ca 1843.. option:: filesize=irange(int)
f80dba8d
MT
1844
1845 Individual file sizes. May be a range, in which case fio will select sizes
1846 for files at random within the given range and limited to :option:`size` in
1847 total (if that is given). If not given, each created file is the same size.
79591fa9
TK
1848 This option overrides :option:`size` in terms of file size, which means
1849 this value is used as a fixed size or possible range of each file.
f80dba8d
MT
1850
1851.. option:: file_append=bool
1852
1853 Perform I/O after the end of the file. Normally fio will operate within the
1854 size of a file. If this option is set, then fio will append to the file
1855 instead. This has identical behavior to setting :option:`offset` to the size
1856 of a file. This option is ignored on non-regular files.
1857
1858.. option:: fill_device=bool, fill_fs=bool
1859
1860 Sets size to something really large and waits for ENOSPC (no space left on
418f5399
MB
1861 device) or EDQUOT (disk quota exceeded)
1862 as the terminating condition. Only makes sense with sequential
f80dba8d
MT
1863 write. For a read workload, the mount point will be filled first then I/O
1864 started on the result. This option doesn't make sense if operating on a raw
1865 device node, since the size of that is already known by the file system.
1866 Additionally, writing beyond end-of-device will not return ENOSPC there.
1867
1868
1869I/O engine
1870~~~~~~~~~~
1871
1872.. option:: ioengine=str
1873
1874 Defines how the job issues I/O to the file. The following types are defined:
1875
1876 **sync**
1877 Basic :manpage:`read(2)` or :manpage:`write(2)`
1878 I/O. :manpage:`lseek(2)` is used to position the I/O location.
54227e6b 1879 See :option:`fsync` and :option:`fdatasync` for syncing write I/Os.
f80dba8d
MT
1880
1881 **psync**
1882 Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on
1883 all supported operating systems except for Windows.
1884
1885 **vsync**
1886 Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate
c60ebc45 1887 queuing by coalescing adjacent I/Os into a single submission.
f80dba8d
MT
1888
1889 **pvsync**
1890 Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O.
1891
1892 **pvsync2**
1893 Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O.
1894
029b42ac
JA
1895 **io_uring**
1896 Fast Linux native asynchronous I/O. Supports async IO
1897 for both direct and buffered IO.
1898 This engine defines engine specific options.
1899
f80dba8d
MT
1900 **libaio**
1901 Linux native asynchronous I/O. Note that Linux may only support
22413915 1902 queued behavior with non-buffered I/O (set ``direct=1`` or
f80dba8d
MT
1903 ``buffered=0``).
1904 This engine defines engine specific options.
1905
1906 **posixaio**
1907 POSIX asynchronous I/O using :manpage:`aio_read(3)` and
1908 :manpage:`aio_write(3)`.
1909
1910 **solarisaio**
1911 Solaris native asynchronous I/O.
1912
1913 **windowsaio**
1914 Windows native asynchronous I/O. Default on Windows.
1915
1916 **mmap**
1917 File is memory mapped with :manpage:`mmap(2)` and data copied
1918 to/from using :manpage:`memcpy(3)`.
1919
1920 **splice**
1921 :manpage:`splice(2)` is used to transfer the data and
1922 :manpage:`vmsplice(2)` to transfer data from user space to the
1923 kernel.
1924
1925 **sg**
1926 SCSI generic sg v3 I/O. May either be synchronous using the SG_IO
1927 ioctl, or if the target is an sg character device we use
1928 :manpage:`read(2)` and :manpage:`write(2)` for asynchronous
f50fbdda 1929 I/O. Requires :option:`filename` option to specify either block or
3740cfc8 1930 character devices. This engine supports trim operations.
52b81b7c 1931 The sg engine includes engine specific options.
f80dba8d
MT
1932
1933 **null**
1934 Doesn't transfer any data, just pretends to. This is mainly used to
1935 exercise fio itself and for debugging/testing purposes.
1936
1937 **net**
1938 Transfer over the network to given ``host:port``. Depending on the
1939 :option:`protocol` used, the :option:`hostname`, :option:`port`,
1940 :option:`listen` and :option:`filename` options are used to specify
1941 what sort of connection to make, while the :option:`protocol` option
1942 determines which protocol will be used. This engine defines engine
1943 specific options.
1944
1945 **netsplice**
1946 Like **net**, but uses :manpage:`splice(2)` and
1947 :manpage:`vmsplice(2)` to map data and send/receive.
1948 This engine defines engine specific options.
1949
1950 **cpuio**
1951 Doesn't transfer any data, but burns CPU cycles according to the
9de473a8
EV
1952 :option:`cpuload`, :option:`cpuchunks` and :option:`cpumode` options.
1953 Setting :option:`cpuload`\=85 will cause that job to do nothing but burn 85%
71aa48eb 1954 of the CPU. In case of SMP machines, use :option:`numjobs`\=<nr_of_cpu>
f50fbdda 1955 to get desired CPU usage, as the cpuload only loads a
f80dba8d
MT
1956 single CPU at the desired rate. A job never finishes unless there is
1957 at least one non-cpuio job.
9de473a8
EV
1958 Setting :option:`cpumode`\=qsort replace the default noop instructions loop
1959 by a qsort algorithm to consume more energy.
f80dba8d 1960
f80dba8d
MT
1961 **rdma**
1962 The RDMA I/O engine supports both RDMA memory semantics
1963 (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the
609ac152
SB
1964 InfiniBand, RoCE and iWARP protocols. This engine defines engine
1965 specific options.
f80dba8d
MT
1966
1967 **falloc**
1968 I/O engine that does regular fallocate to simulate data transfer as
1969 fio ioengine.
1970
1971 DDIR_READ
1972 does fallocate(,mode = FALLOC_FL_KEEP_SIZE,).
1973
1974 DDIR_WRITE
1975 does fallocate(,mode = 0).
1976
1977 DDIR_TRIM
1978 does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE).
1979
761cd093
SW
1980 **ftruncate**
1981 I/O engine that sends :manpage:`ftruncate(2)` operations in response
1982 to write (DDIR_WRITE) events. Each ftruncate issued sets the file's
f50fbdda 1983 size to the current block offset. :option:`blocksize` is ignored.
761cd093 1984
f80dba8d
MT
1985 **e4defrag**
1986 I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate
1987 defragment activity in request to DDIR_WRITE event.
1988
f3f96717
IF
1989 **rados**
1990 I/O engine supporting direct access to Ceph Reliable Autonomic
1991 Distributed Object Store (RADOS) via librados. This ioengine
1992 defines engine specific options.
1993
f80dba8d
MT
1994 **rbd**
1995 I/O engine supporting direct access to Ceph Rados Block Devices
1996 (RBD) via librbd without the need to use the kernel rbd driver. This
1997 ioengine defines engine specific options.
1998
c2f6a13d
LMB
1999 **http**
2000 I/O engine supporting GET/PUT requests over HTTP(S) with libcurl to
2001 a WebDAV or S3 endpoint. This ioengine defines engine specific options.
2002
2003 This engine only supports direct IO of iodepth=1; you need to scale this
2004 via numjobs. blocksize defines the size of the objects to be created.
2005
2006 TRIM is translated to object deletion.
2007
f80dba8d 2008 **gfapi**
ac8ca2af
SW
2009 Using GlusterFS libgfapi sync interface to direct access to
2010 GlusterFS volumes without having to go through FUSE. This ioengine
f80dba8d
MT
2011 defines engine specific options.
2012
2013 **gfapi_async**
ac8ca2af
SW
2014 Using GlusterFS libgfapi async interface to direct access to
2015 GlusterFS volumes without having to go through FUSE. This ioengine
f80dba8d
MT
2016 defines engine specific options.
2017
2018 **libhdfs**
f50fbdda 2019 Read and write through Hadoop (HDFS). The :option:`filename` option
f80dba8d
MT
2020 is used to specify host,port of the hdfs name-node to connect. This
2021 engine interprets offsets a little differently. In HDFS, files once
e25c0c91
SW
2022 created cannot be modified so random writes are not possible. To
2023 imitate this the libhdfs engine expects a bunch of small files to be
2024 created over HDFS and will randomly pick a file from them
2025 based on the offset generated by fio backend (see the example
f80dba8d 2026 job file to create such files, use ``rw=write`` option). Please
e25c0c91
SW
2027 note, it may be necessary to set environment variables to work
2028 with HDFS/libhdfs properly. Each job uses its own connection to
f80dba8d
MT
2029 HDFS.
2030
2031 **mtd**
2032 Read, write and erase an MTD character device (e.g.,
2033 :file:`/dev/mtd0`). Discards are treated as erases. Depending on the
2034 underlying device type, the I/O may have to go in a certain pattern,
2035 e.g., on NAND, writing sequentially to erase blocks and discarding
c298ee71 2036 before overwriting. The `trimwrite` mode works well for this
f80dba8d
MT
2037 constraint.
2038
2039 **pmemblk**
2040 Read and write using filesystem DAX to a file on a filesystem
363a5f65 2041 mounted with DAX on a persistent memory device through the PMDK
f80dba8d
MT
2042 libpmemblk library.
2043
2044 **dev-dax**
2045 Read and write using device DAX to a persistent memory device (e.g.,
363a5f65 2046 /dev/dax0.0) through the PMDK libpmem library.
f80dba8d
MT
2047
2048 **external**
2049 Prefix to specify loading an external I/O engine object file. Append
c60ebc45 2050 the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load
d243fd6d
TK
2051 ioengine :file:`foo.o` in :file:`/tmp`. The path can be either
2052 absolute or relative. See :file:`engines/skeleton_external.c` for
2053 details of writing an external I/O engine.
f80dba8d 2054
1216cc5a 2055 **filecreate**
b71968b1 2056 Simply create the files and do no I/O to them. You still need to
1216cc5a 2057 set `filesize` so that all the accounting still occurs, but no
b71968b1 2058 actual I/O will be done other than creating the file.
f80dba8d 2059
73ccd14e
SF
2060 **filestat**
2061 Simply do stat() and do no I/O to the file. You need to set 'filesize'
2062 and 'nrfiles', so that files will be created.
2063 This engine is to measure file lookup and meta data access.
2064
5561e9dd
FS
2065 **filedelete**
2066 Simply delete the files by unlink() and do no I/O to them. You need to set 'filesize'
2067 and 'nrfiles', so that the files will be created.
2068 This engine is to measure file delete.
2069
ae0db592
TI
2070 **libpmem**
2071 Read and write using mmap I/O to a file on a filesystem
363a5f65 2072 mounted with DAX on a persistent memory device through the PMDK
ae0db592
TI
2073 libpmem library.
2074
a40e7a59
GB
2075 **ime_psync**
2076 Synchronous read and write using DDN's Infinite Memory Engine (IME).
2077 This engine is very basic and issues calls to IME whenever an IO is
2078 queued.
2079
2080 **ime_psyncv**
2081 Synchronous read and write using DDN's Infinite Memory Engine (IME).
2082 This engine uses iovecs and will try to stack as much IOs as possible
2083 (if the IOs are "contiguous" and the IO depth is not exceeded)
2084 before issuing a call to IME.
2085
2086 **ime_aio**
2087 Asynchronous read and write using DDN's Infinite Memory Engine (IME).
2088 This engine will try to stack as much IOs as possible by creating
2089 requests for IME. FIO will then decide when to commit these requests.
247ef2aa
KZ
2090 **libiscsi**
2091 Read and write iscsi lun with libiscsi.
d643a1e2 2092 **nbd**
f2d6de5d 2093 Read and write a Network Block Device (NBD).
a40e7a59 2094
10756b2c
BS
2095 **libcufile**
2096 I/O engine supporting libcufile synchronous access to nvidia-fs and a
2097 GPUDirect Storage-supported filesystem. This engine performs
2098 I/O without transferring buffers between user-space and the kernel,
2099 unless :option:`verify` is set or :option:`cuda_io` is `posix`.
2100 :option:`iomem` must not be `cudamalloc`. This ioengine defines
2101 engine specific options.
c363fdd7
JL
2102 **dfs**
2103 I/O engine supporting asynchronous read and write operations to the
2104 DAOS File System (DFS) via libdfs.
10756b2c 2105
9326926b
TG
2106 **nfs**
2107 I/O engine supporting asynchronous read and write operations to
2108 NFS filesystems from userspace via libnfs. This is useful for
2109 achieving higher concurrency and thus throughput than is possible
2110 via kernel NFS.
2111
f80dba8d
MT
2112I/O engine specific parameters
2113~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2114
2115In addition, there are some parameters which are only valid when a specific
f50fbdda
TK
2116:option:`ioengine` is in use. These are used identically to normal parameters,
2117with the caveat that when used on the command line, they must come after the
f80dba8d
MT
2118:option:`ioengine` that defines them is selected.
2119
b2a432bf 2120.. option:: cmdprio_percentage=int : [io_uring] [libaio]
029b42ac 2121
b2a432bf
PC
2122 Set the percentage of I/O that will be issued with higher priority by setting
2123 the priority bit. Non-read I/O is likely unaffected by ``cmdprio_percentage``.
2124 This option cannot be used with the `prio` or `prioclass` options. For this
2125 option to set the priority bit properly, NCQ priority must be supported and
7896180a
VF
2126 enabled and :option:`direct`\=1 option must be used. fio must also be run as
2127 the root user.
029b42ac
JA
2128
2129.. option:: fixedbufs : [io_uring]
2130
b2a432bf
PC
2131 If fio is asked to do direct IO, then Linux will map pages for each
2132 IO call, and release them when IO is done. If this option is set, the
2133 pages are pre-mapped before IO is started. This eliminates the need to
2134 map and release for each IO. This is more efficient, and reduces the
2135 IO latency as well.
2136
2137.. option:: hipri : [io_uring]
2138
2139 If this option is set, fio will attempt to use polled IO completions.
2140 Normal IO completions generate interrupts to signal the completion of
2141 IO, polled completions do not. Hence they are require active reaping
2142 by the application. The benefits are more efficient IO for high IOPS
2143 scenarios, and lower latencies for low queue depth IO.
029b42ac 2144
5ffd5626 2145.. option:: registerfiles : [io_uring]
2c870598 2146
5ffd5626
JA
2147 With this option, fio registers the set of files being used with the
2148 kernel. This avoids the overhead of managing file counts in the kernel,
2149 making the submission and completion part more lightweight. Required
2150 for the below :option:`sqthread_poll` option.
2151
029b42ac
JA
2152.. option:: sqthread_poll : [io_uring]
2153
2154 Normally fio will submit IO by issuing a system call to notify the
2155 kernel of available items in the SQ ring. If this option is set, the
2156 act of submitting IO will be done by a polling thread in the kernel.
2157 This frees up cycles for fio, at the cost of using more CPU in the
2158 system.
2159
2160.. option:: sqthread_poll_cpu : [io_uring]
2161
2162 When :option:`sqthread_poll` is set, this option provides a way to
2163 define which CPU should be used for the polling thread.
2164
f80dba8d
MT
2165.. option:: userspace_reap : [libaio]
2166
2167 Normally, with the libaio engine in use, fio will use the
2168 :manpage:`io_getevents(2)` system call to reap newly returned events. With
2169 this flag turned on, the AIO ring will be read directly from user-space to
2170 reap events. The reaping mode is only enabled when polling for a minimum of
c60ebc45 2171 0 events (e.g. when :option:`iodepth_batch_complete` `=0`).
f80dba8d 2172
9d25d068 2173.. option:: hipri : [pvsync2]
f80dba8d
MT
2174
2175 Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority
2176 than normal.
2177
a0679ce5
SB
2178.. option:: hipri_percentage : [pvsync2]
2179
f50fbdda 2180 When hipri is set this determines the probability of a pvsync2 I/O being high
a0679ce5
SB
2181 priority. The default is 100%.
2182
7d42e66e
KK
2183.. option:: nowait : [pvsync2] [libaio] [io_uring]
2184
2185 By default if a request cannot be executed immediately (e.g. resource starvation,
2186 waiting on locks) it is queued and the initiating process will be blocked until
2187 the required resource becomes free.
2188
2189 This option sets the RWF_NOWAIT flag (supported from the 4.14 Linux kernel) and
2190 the call will return instantly with EAGAIN or a partial result rather than waiting.
2191
2192 It is useful to also use ignore_error=EAGAIN when using this option.
2193
2194 Note: glibc 2.27, 2.28 have a bug in syscall wrappers preadv2, pwritev2.
2195 They return EOPNOTSUP instead of EAGAIN.
2196
2197 For cached I/O, using this option usually means a request operates only with
2198 cached data. Currently the RWF_NOWAIT flag does not supported for cached write.
2199
2200 For direct I/O, requests will only succeed if cache invalidation isn't required,
2201 file blocks are fully allocated and the disk request could be issued immediately.
2202
f80dba8d
MT
2203.. option:: cpuload=int : [cpuio]
2204
da19cdb4
TK
2205 Attempt to use the specified percentage of CPU cycles. This is a mandatory
2206 option when using cpuio I/O engine.
f80dba8d
MT
2207
2208.. option:: cpuchunks=int : [cpuio]
2209
2210 Split the load into cycles of the given time. In microseconds.
2211
2212.. option:: exit_on_io_done=bool : [cpuio]
2213
2214 Detect when I/O threads are done, then exit.
2215
f80dba8d
MT
2216.. option:: namenode=str : [libhdfs]
2217
22413915 2218 The hostname or IP address of a HDFS cluster namenode to contact.
f80dba8d
MT
2219
2220.. option:: port=int
2221
f50fbdda
TK
2222 [libhdfs]
2223
2224 The listening port of the HFDS cluster namenode.
2225
f80dba8d
MT
2226 [netsplice], [net]
2227
2228 The TCP or UDP port to bind to or connect to. If this is used with
2229 :option:`numjobs` to spawn multiple instances of the same job type, then
2230 this will be the starting port number since fio will use a range of
2231 ports.
2232
e4c4625f 2233 [rdma], [librpma_*]
609ac152
SB
2234
2235 The port to use for RDMA-CM communication. This should be the same value
2236 on the client and the server side.
2237
2238.. option:: hostname=str : [netsplice] [net] [rdma]
f80dba8d 2239
609ac152
SB
2240 The hostname or IP address to use for TCP, UDP or RDMA-CM based I/O. If the job
2241 is a TCP listener or UDP reader, the hostname is not used and must be omitted
f50fbdda 2242 unless it is a valid UDP multicast address.
f80dba8d 2243
e4c4625f
JM
2244.. option:: serverip=str : [librpma_*]
2245
2246 The IP address to be used for RDMA-CM based I/O.
2247
2248.. option:: direct_write_to_pmem=bool : [librpma_*]
2249
2250 Set to 1 only when Direct Write to PMem from the remote host is possible.
2251 Otherwise, set to 0.
2252
6a229978
OS
2253.. option:: busy_wait_polling=bool : [librpma_*_server]
2254
2255 Set to 0 to wait for completion instead of busy-wait polling completion.
2256 Default: 1.
2257
f80dba8d
MT
2258.. option:: interface=str : [netsplice] [net]
2259
2260 The IP address of the network interface used to send or receive UDP
2261 multicast.
2262
2263.. option:: ttl=int : [netsplice] [net]
2264
2265 Time-to-live value for outgoing UDP multicast packets. Default: 1.
2266
2267.. option:: nodelay=bool : [netsplice] [net]
2268
2269 Set TCP_NODELAY on TCP connections.
2270
f50fbdda 2271.. option:: protocol=str, proto=str : [netsplice] [net]
f80dba8d
MT
2272
2273 The network protocol to use. Accepted values are:
2274
2275 **tcp**
2276 Transmission control protocol.
2277 **tcpv6**
2278 Transmission control protocol V6.
2279 **udp**
2280 User datagram protocol.
2281 **udpv6**
2282 User datagram protocol V6.
2283 **unix**
2284 UNIX domain socket.
2285
2286 When the protocol is TCP or UDP, the port must also be given, as well as the
2287 hostname if the job is a TCP listener or UDP reader. For unix sockets, the
f50fbdda 2288 normal :option:`filename` option should be used and the port is invalid.
f80dba8d 2289
e9184ec1 2290.. option:: listen : [netsplice] [net]
f80dba8d
MT
2291
2292 For TCP network connections, tell fio to listen for incoming connections
2293 rather than initiating an outgoing connection. The :option:`hostname` must
2294 be omitted if this option is used.
2295
e9184ec1 2296.. option:: pingpong : [netsplice] [net]
f80dba8d
MT
2297
2298 Normally a network writer will just continue writing data, and a network
2299 reader will just consume packages. If ``pingpong=1`` is set, a writer will
2300 send its normal payload to the reader, then wait for the reader to send the
2301 same payload back. This allows fio to measure network latencies. The
2302 submission and completion latencies then measure local time spent sending or
2303 receiving, and the completion latency measures how long it took for the
2304 other end to receive and send back. For UDP multicast traffic
2305 ``pingpong=1`` should only be set for a single reader when multiple readers
2306 are listening to the same address.
2307
e9184ec1 2308.. option:: window_size : [netsplice] [net]
f80dba8d
MT
2309
2310 Set the desired socket buffer size for the connection.
2311
e9184ec1 2312.. option:: mss : [netsplice] [net]
f80dba8d
MT
2313
2314 Set the TCP maximum segment size (TCP_MAXSEG).
2315
2316.. option:: donorname=str : [e4defrag]
2317
730bd7d9 2318 File will be used as a block donor (swap extents between files).
f80dba8d
MT
2319
2320.. option:: inplace=int : [e4defrag]
2321
2322 Configure donor file blocks allocation strategy:
2323
2324 **0**
2325 Default. Preallocate donor's file on init.
2326 **1**
2b455dbf 2327 Allocate space immediately inside defragment event, and free right
f80dba8d
MT
2328 after event.
2329
f3f96717 2330.. option:: clustername=str : [rbd,rados]
f80dba8d
MT
2331
2332 Specifies the name of the Ceph cluster.
2333
2334.. option:: rbdname=str : [rbd]
2335
2336 Specifies the name of the RBD.
2337
f3f96717 2338.. option:: pool=str : [rbd,rados]
f80dba8d 2339
f3f96717 2340 Specifies the name of the Ceph pool containing RBD or RADOS data.
f80dba8d 2341
f3f96717 2342.. option:: clientname=str : [rbd,rados]
f80dba8d
MT
2343
2344 Specifies the username (without the 'client.' prefix) used to access the
2345 Ceph cluster. If the *clustername* is specified, the *clientname* shall be
2346 the full *type.id* string. If no type. prefix is given, fio will add
2347 'client.' by default.
2348
f3f96717
IF
2349.. option:: busy_poll=bool : [rbd,rados]
2350
2351 Poll store instead of waiting for completion. Usually this provides better
2352 throughput at cost of higher(up to 100%) CPU utilization.
2353
2b728756
AK
2354.. option:: touch_objects=bool : [rados]
2355
2356 During initialization, touch (create if do not exist) all objects (files).
2357 Touching all objects affects ceph caches and likely impacts test results.
2358 Enabled by default.
2359
f80dba8d
MT
2360.. option:: skip_bad=bool : [mtd]
2361
2362 Skip operations against known bad blocks.
2363
2364.. option:: hdfsdirectory : [libhdfs]
2365
2366 libhdfs will create chunk in this HDFS directory.
2367
2368.. option:: chunk_size : [libhdfs]
2369
2b455dbf 2370 The size of the chunk to use for each file.
f80dba8d 2371
609ac152
SB
2372.. option:: verb=str : [rdma]
2373
2374 The RDMA verb to use on this side of the RDMA ioengine connection. Valid
2375 values are write, read, send and recv. These correspond to the equivalent
2376 RDMA verbs (e.g. write = rdma_write etc.). Note that this only needs to be
2377 specified on the client side of the connection. See the examples folder.
2378
2379.. option:: bindname=str : [rdma]
2380
2381 The name to use to bind the local RDMA-CM connection to a local RDMA device.
2382 This could be a hostname or an IPv4 or IPv6 address. On the server side this
2383 will be passed into the rdma_bind_addr() function and on the client site it
2384 will be used in the rdma_resolve_add() function. This can be useful when
2385 multiple paths exist between the client and the server or in certain loopback
2386 configurations.
f80dba8d 2387
93a13ba5 2388.. option:: stat_type=str : [filestat]
c446eff0 2389
93a13ba5
TK
2390 Specify stat system call type to measure lookup/getattr performance.
2391 Default is **stat** for :manpage:`stat(2)`.
c446eff0 2392
52b81b7c
KD
2393.. option:: readfua=bool : [sg]
2394
2395 With readfua option set to 1, read operations include
2396 the force unit access (fua) flag. Default is 0.
2397
2398.. option:: writefua=bool : [sg]
2399
2400 With writefua option set to 1, write operations include
2401 the force unit access (fua) flag. Default is 0.
2402
2c3a9150 2403.. option:: sg_write_mode=str : [sg]
3740cfc8 2404
2c3a9150
VF
2405 Specify the type of write commands to issue. This option can take three values:
2406
2407 **write**
2408 This is the default where write opcodes are issued as usual.
2409 **verify**
2410 Issue WRITE AND VERIFY commands. The BYTCHK bit is set to 0. This
2411 directs the device to carry out a medium verification with no data
2412 comparison. The writefua option is ignored with this selection.
2413 **same**
2414 Issue WRITE SAME commands. This transfers a single block to the device
2415 and writes this same block of data to a contiguous sequence of LBAs
2416 beginning at the specified offset. fio's block size parameter specifies
2417 the amount of data written with each command. However, the amount of data
2418 actually transferred to the device is equal to the device's block
2419 (sector) size. For a device with 512 byte sectors, blocksize=8k will
2420 write 16 sectors with each command. fio will still generate 8k of data
2421 for each command but only the first 512 bytes will be used and
2422 transferred to the device. The writefua option is ignored with this
2423 selection.
52b81b7c 2424
e493ceae
JA
2425.. option:: hipri : [sg]
2426
2427 If this option is set, fio will attempt to use polled IO completions.
2428 This will have a similar effect as (io_uring)hipri. Only SCSI READ and
2429 WRITE commands will have the SGV4_FLAG_HIPRI set (not UNMAP (trim) nor
2430 VERIFY). Older versions of the Linux sg driver that do not support
2431 hipri will simply ignore this flag and do normal IO. The Linux SCSI
2432 Low Level Driver (LLD) that "owns" the device also needs to support
2433 hipri (also known as iopoll and mq_poll). The MegaRAID driver is an
2434 example of a SCSI LLD. Default: clear (0) which does normal
2435 (interrupted based) IO.
2436
c2f6a13d
LMB
2437.. option:: http_host=str : [http]
2438
2439 Hostname to connect to. For S3, this could be the bucket hostname.
2440 Default is **localhost**
2441
2442.. option:: http_user=str : [http]
2443
2444 Username for HTTP authentication.
2445
2446.. option:: http_pass=str : [http]
2447
2448 Password for HTTP authentication.
2449
09fd2966 2450.. option:: https=str : [http]
c2f6a13d 2451
09fd2966
LMB
2452 Enable HTTPS instead of http. *on* enables HTTPS; *insecure*
2453 will enable HTTPS, but disable SSL peer verification (use with
2454 caution!). Default is **off**
c2f6a13d 2455
09fd2966 2456.. option:: http_mode=str : [http]
c2f6a13d 2457
09fd2966
LMB
2458 Which HTTP access mode to use: *webdav*, *swift*, or *s3*.
2459 Default is **webdav**
c2f6a13d
LMB
2460
2461.. option:: http_s3_region=str : [http]
2462
2463 The S3 region/zone string.
2464 Default is **us-east-1**
2465
2466.. option:: http_s3_key=str : [http]
2467
2468 The S3 secret key.
2469
2470.. option:: http_s3_keyid=str : [http]
2471
2472 The S3 key/access id.
2473
09fd2966
LMB
2474.. option:: http_swift_auth_token=str : [http]
2475
2476 The Swift auth token. See the example configuration file on how
2477 to retrieve this.
2478
c2f6a13d
LMB
2479.. option:: http_verbose=int : [http]
2480
2481 Enable verbose requests from libcurl. Useful for debugging. 1
2482 turns on verbose logging from libcurl, 2 additionally enables
2483 HTTP IO tracing. Default is **0**
2484
f2d6de5d
RJ
2485.. option:: uri=str : [nbd]
2486
2487 Specify the NBD URI of the server to test. The string
2488 is a standard NBD URI
2489 (see https://github.com/NetworkBlockDevice/nbd/tree/master/doc).
2490 Example URIs: nbd://localhost:10809
2491 nbd+unix:///?socket=/tmp/socket
2492 nbds://tlshost/exportname
2493
10756b2c
BS
2494.. option:: gpu_dev_ids=str : [libcufile]
2495
2496 Specify the GPU IDs to use with CUDA. This is a colon-separated list of
2497 int. GPUs are assigned to workers roundrobin. Default is 0.
2498
2499.. option:: cuda_io=str : [libcufile]
2500
2501 Specify the type of I/O to use with CUDA. Default is **cufile**.
2502
2503 **cufile**
2504 Use libcufile and nvidia-fs. This option performs I/O directly
2505 between a GPUDirect Storage filesystem and GPU buffers,
2506 avoiding use of a bounce buffer. If :option:`verify` is set,
2507 cudaMemcpy is used to copy verificaton data between RAM and GPU.
2508 Verification data is copied from RAM to GPU before a write
2509 and from GPU to RAM after a read. :option:`direct` must be 1.
2510 **posix**
2511 Use POSIX to perform I/O with a RAM buffer, and use cudaMemcpy
2512 to transfer data between RAM and the GPUs. Data is copied from
2513 GPU to RAM before a write and copied from RAM to GPU after a
2514 read. :option:`verify` does not affect use of cudaMemcpy.
2515
c363fdd7
JL
2516.. option:: pool=str : [dfs]
2517
2518 Specify the UUID of the DAOS pool to connect to.
2519
2520.. option:: cont=str : [dfs]
2521
2522 Specify the UUID of the DAOS container to open.
2523
2524.. option:: chunk_size=int : [dfs]
2525
2526 Specificy a different chunk size (in bytes) for the dfs file.
2527 Use DAOS container's chunk size by default.
2528
2529.. option:: object_class=str : [dfs]
2530
2531 Specificy a different object class for the dfs file.
2532 Use DAOS container's object class by default.
2533
9326926b
TG
2534.. option:: nfs_url=str : [nfs]
2535
2536 URL in libnfs format, eg nfs://<server|ipv4|ipv6>/path[?arg=val[&arg=val]*]
2537 Refer to the libnfs README for more details.
2538
f80dba8d
MT
2539I/O depth
2540~~~~~~~~~
2541
2542.. option:: iodepth=int
2543
2544 Number of I/O units to keep in flight against the file. Note that
2545 increasing *iodepth* beyond 1 will not affect synchronous ioengines (except
c60ebc45 2546 for small degrees when :option:`verify_async` is in use). Even async
f80dba8d
MT
2547 engines may impose OS restrictions causing the desired depth not to be
2548 achieved. This may happen on Linux when using libaio and not setting
9207a0cb 2549 :option:`direct`\=1, since buffered I/O is not async on that OS. Keep an
f80dba8d
MT
2550 eye on the I/O depth distribution in the fio output to verify that the
2551 achieved depth is as expected. Default: 1.
2552
2553.. option:: iodepth_batch_submit=int, iodepth_batch=int
2554
2555 This defines how many pieces of I/O to submit at once. It defaults to 1
2556 which means that we submit each I/O as soon as it is available, but can be
2557 raised to submit bigger batches of I/O at the time. If it is set to 0 the
2558 :option:`iodepth` value will be used.
2559
2560.. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int
2561
2562 This defines how many pieces of I/O to retrieve at once. It defaults to 1
2563 which means that we'll ask for a minimum of 1 I/O in the retrieval process
2564 from the kernel. The I/O retrieval will go on until we hit the limit set by
2565 :option:`iodepth_low`. If this variable is set to 0, then fio will always
2566 check for completed events before queuing more I/O. This helps reduce I/O
2567 latency, at the cost of more retrieval system calls.
2568
2569.. option:: iodepth_batch_complete_max=int
2570
2571 This defines maximum pieces of I/O to retrieve at once. This variable should
9207a0cb 2572 be used along with :option:`iodepth_batch_complete_min`\=int variable,
f80dba8d 2573 specifying the range of min and max amount of I/O which should be
730bd7d9 2574 retrieved. By default it is equal to the :option:`iodepth_batch_complete_min`
f80dba8d
MT
2575 value.
2576
2577 Example #1::
2578
2579 iodepth_batch_complete_min=1
2580 iodepth_batch_complete_max=<iodepth>
2581
2582 which means that we will retrieve at least 1 I/O and up to the whole
2583 submitted queue depth. If none of I/O has been completed yet, we will wait.
2584
2585 Example #2::
2586
2587 iodepth_batch_complete_min=0
2588 iodepth_batch_complete_max=<iodepth>
2589
2590 which means that we can retrieve up to the whole submitted queue depth, but
2591 if none of I/O has been completed yet, we will NOT wait and immediately exit
2592 the system call. In this example we simply do polling.
2593
2594.. option:: iodepth_low=int
2595
2596 The low water mark indicating when to start filling the queue
2597 again. Defaults to the same as :option:`iodepth`, meaning that fio will
2598 attempt to keep the queue full at all times. If :option:`iodepth` is set to
c60ebc45 2599 e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of
f80dba8d
MT
2600 16 requests, it will let the depth drain down to 4 before starting to fill
2601 it again.
2602
997b5680
SW
2603.. option:: serialize_overlap=bool
2604
2605 Serialize in-flight I/Os that might otherwise cause or suffer from data races.
2606 When two or more I/Os are submitted simultaneously, there is no guarantee that
2607 the I/Os will be processed or completed in the submitted order. Further, if
2608 two or more of those I/Os are writes, any overlapping region between them can
2609 become indeterminate/undefined on certain storage. These issues can cause
2610 verification to fail erratically when at least one of the racing I/Os is
2611 changing data and the overlapping region has a non-zero size. Setting
2612 ``serialize_overlap`` tells fio to avoid provoking this behavior by explicitly
2613 serializing in-flight I/Os that have a non-zero overlap. Note that setting
ee21ebee 2614 this option can reduce both performance and the :option:`iodepth` achieved.
3d6a6f04
VF
2615
2616 This option only applies to I/Os issued for a single job except when it is
a02ec45a 2617 enabled along with :option:`io_submit_mode`\=offload. In offload mode, fio
3d6a6f04 2618 will check for overlap among all I/Os submitted by offload jobs with :option:`serialize_overlap`
307f2246 2619 enabled.
3d6a6f04
VF
2620
2621 Default: false.
997b5680 2622
f80dba8d
MT
2623.. option:: io_submit_mode=str
2624
2625 This option controls how fio submits the I/O to the I/O engine. The default
2626 is `inline`, which means that the fio job threads submit and reap I/O
2627 directly. If set to `offload`, the job threads will offload I/O submission
2628 to a dedicated pool of I/O threads. This requires some coordination and thus
2629 has a bit of extra overhead, especially for lower queue depth I/O where it
2630 can increase latencies. The benefit is that fio can manage submission rates
2631 independently of the device completion rates. This avoids skewed latency
730bd7d9 2632 reporting if I/O gets backed up on the device side (the coordinated omission
abfd235a
JA
2633 problem). Note that this option cannot reliably be used with async IO
2634 engines.
f80dba8d
MT
2635
2636
2637I/O rate
2638~~~~~~~~
2639
a881438b 2640.. option:: thinktime=time
f80dba8d 2641
f75ede1d
SW
2642 Stall the job for the specified period of time after an I/O has completed before issuing the
2643 next. May be used to simulate processing being done by an application.
947e0fe0 2644 When the unit is omitted, the value is interpreted in microseconds. See
f80dba8d
MT
2645 :option:`thinktime_blocks` and :option:`thinktime_spin`.
2646
a881438b 2647.. option:: thinktime_spin=time
f80dba8d
MT
2648
2649 Only valid if :option:`thinktime` is set - pretend to spend CPU time doing
2650 something with the data received, before falling back to sleeping for the
f75ede1d 2651 rest of the period specified by :option:`thinktime`. When the unit is
947e0fe0 2652 omitted, the value is interpreted in microseconds.
f80dba8d
MT
2653
2654.. option:: thinktime_blocks=int
2655
2656 Only valid if :option:`thinktime` is set - control how many blocks to issue,
f50fbdda
TK
2657 before waiting :option:`thinktime` usecs. If not set, defaults to 1 which will make
2658 fio wait :option:`thinktime` usecs after every block. This effectively makes any
f80dba8d 2659 queue depth setting redundant, since no more than 1 I/O will be queued
f50fbdda 2660 before we have to complete it and do our :option:`thinktime`. In other words, this
f80dba8d 2661 setting effectively caps the queue depth if the latter is larger.
71bfa161 2662
33f42c20
HQ
2663.. option:: thinktime_blocks_type=str
2664
2665 Only valid if :option:`thinktime` is set - control how :option:`thinktime_blocks`
2666 triggers. The default is `complete`, which triggers thinktime when fio completes
2667 :option:`thinktime_blocks` blocks. If this is set to `issue`, then the trigger happens
2668 at the issue side.
2669
f80dba8d 2670.. option:: rate=int[,int][,int]
71bfa161 2671
f80dba8d
MT
2672 Cap the bandwidth used by this job. The number is in bytes/sec, the normal
2673 suffix rules apply. Comma-separated values may be specified for reads,
2674 writes, and trims as described in :option:`blocksize`.
71bfa161 2675
b25b3464
SW
2676 For example, using `rate=1m,500k` would limit reads to 1MiB/sec and writes to
2677 500KiB/sec. Capping only reads or writes can be done with `rate=,500k` or
2678 `rate=500k,` where the former will only limit writes (to 500KiB/sec) and the
2679 latter will only limit reads.
2680
f80dba8d 2681.. option:: rate_min=int[,int][,int]
71bfa161 2682
f80dba8d
MT
2683 Tell fio to do whatever it can to maintain at least this bandwidth. Failing
2684 to meet this requirement will cause the job to exit. Comma-separated values
2685 may be specified for reads, writes, and trims as described in
2686 :option:`blocksize`.
71bfa161 2687
f80dba8d 2688.. option:: rate_iops=int[,int][,int]
71bfa161 2689
f80dba8d
MT
2690 Cap the bandwidth to this number of IOPS. Basically the same as
2691 :option:`rate`, just specified independently of bandwidth. If the job is
2692 given a block size range instead of a fixed value, the smallest block size
2693 is used as the metric. Comma-separated values may be specified for reads,
2694 writes, and trims as described in :option:`blocksize`.
71bfa161 2695
f80dba8d 2696.. option:: rate_iops_min=int[,int][,int]
71bfa161 2697
f80dba8d
MT
2698 If fio doesn't meet this rate of I/O, it will cause the job to exit.
2699 Comma-separated values may be specified for reads, writes, and trims as
2700 described in :option:`blocksize`.
71bfa161 2701
f80dba8d 2702.. option:: rate_process=str
66c098b8 2703
f80dba8d
MT
2704 This option controls how fio manages rated I/O submissions. The default is
2705 `linear`, which submits I/O in a linear fashion with fixed delays between
c60ebc45 2706 I/Os that gets adjusted based on I/O completion rates. If this is set to
f80dba8d
MT
2707 `poisson`, fio will submit I/O based on a more real world random request
2708 flow, known as the Poisson process
2709 (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be
2710 10^6 / IOPS for the given workload.
71bfa161 2711
1a9bf814
JA
2712.. option:: rate_ignore_thinktime=bool
2713
2714 By default, fio will attempt to catch up to the specified rate setting,
2715 if any kind of thinktime setting was used. If this option is set, then
2716 fio will ignore the thinktime and continue doing IO at the specified
2717 rate, instead of entering a catch-up mode after thinktime is done.
2718
71bfa161 2719
f80dba8d
MT
2720I/O latency
2721~~~~~~~~~~~
71bfa161 2722
a881438b 2723.. option:: latency_target=time
71bfa161 2724
f80dba8d 2725 If set, fio will attempt to find the max performance point that the given
f75ede1d 2726 workload will run at while maintaining a latency below this target. When
947e0fe0 2727 the unit is omitted, the value is interpreted in microseconds. See
f75ede1d 2728 :option:`latency_window` and :option:`latency_percentile`.
71bfa161 2729
a881438b 2730.. option:: latency_window=time
71bfa161 2731
f80dba8d 2732 Used with :option:`latency_target` to specify the sample window that the job
f75ede1d 2733 is run at varying queue depths to test the performance. When the unit is
947e0fe0 2734 omitted, the value is interpreted in microseconds.
b4692828 2735
f80dba8d 2736.. option:: latency_percentile=float
71bfa161 2737
c60ebc45 2738 The percentage of I/Os that must fall within the criteria specified by
f80dba8d 2739 :option:`latency_target` and :option:`latency_window`. If not set, this
c60ebc45 2740 defaults to 100.0, meaning that all I/Os must be equal or below to the value
f80dba8d 2741 set by :option:`latency_target`.
71bfa161 2742
e1bcd541
SL
2743.. option:: latency_run=bool
2744
2745 Used with :option:`latency_target`. If false (default), fio will find
2746 the highest queue depth that meets :option:`latency_target` and exit. If
2747 true, fio will continue running and try to meet :option:`latency_target`
2748 by adjusting queue depth.
2749
f7cf63bf 2750.. option:: max_latency=time[,time][,time]
71bfa161 2751
f75ede1d 2752 If set, fio will exit the job with an ETIMEDOUT error if it exceeds this
947e0fe0 2753 maximum latency. When the unit is omitted, the value is interpreted in
f7cf63bf
VR
2754 microseconds. Comma-separated values may be specified for reads, writes,
2755 and trims as described in :option:`blocksize`.
71bfa161 2756
f80dba8d 2757.. option:: rate_cycle=int
71bfa161 2758
f80dba8d 2759 Average bandwidth for :option:`rate` and :option:`rate_min` over this number
a47b697c 2760 of milliseconds. Defaults to 1000.
71bfa161 2761
71bfa161 2762
f80dba8d
MT
2763I/O replay
2764~~~~~~~~~~
71bfa161 2765
f80dba8d 2766.. option:: write_iolog=str
c2b1e753 2767
f80dba8d
MT
2768 Write the issued I/O patterns to the specified file. See
2769 :option:`read_iolog`. Specify a separate file for each job, otherwise the
2770 iologs will be interspersed and the file may be corrupt.
c2b1e753 2771
f80dba8d 2772.. option:: read_iolog=str
71bfa161 2773
22413915 2774 Open an iolog with the specified filename and replay the I/O patterns it
f80dba8d
MT
2775 contains. This can be used to store a workload and replay it sometime
2776 later. The iolog given may also be a blktrace binary file, which allows fio
2777 to replay a workload captured by :command:`blktrace`. See
2778 :manpage:`blktrace(8)` for how to capture such logging data. For blktrace
2779 replay, the file needs to be turned into a blkparse binary data file first
2780 (``blkparse <device> -o /dev/null -d file_for_fio.bin``).
78439a18
JA
2781 You can specify a number of files by separating the names with a ':'
2782 character. See the :option:`filename` option for information on how to
3b803fe1 2783 escape ':' characters within the file names. These files will
78439a18 2784 be sequentially assigned to job clones created by :option:`numjobs`.
d19c04d1 2785 '-' is a reserved name, meaning read from stdin, notably if
2786 :option:`filename` is set to '-' which means stdin as well, then
2787 this flag can't be set to '-'.
71bfa161 2788
77be374d
AK
2789.. option:: read_iolog_chunked=bool
2790
2791 Determines how iolog is read. If false(default) entire :option:`read_iolog`
2792 will be read at once. If selected true, input from iolog will be read
2793 gradually. Useful when iolog is very large, or it is generated.
2794
b9921d1a
DZ
2795.. option:: merge_blktrace_file=str
2796
2797 When specified, rather than replaying the logs passed to :option:`read_iolog`,
2798 the logs go through a merge phase which aggregates them into a single
2799 blktrace. The resulting file is then passed on as the :option:`read_iolog`
2800 parameter. The intention here is to make the order of events consistent.
2801 This limits the influence of the scheduler compared to replaying multiple
2802 blktraces via concurrent jobs.
2803
87a48ada
DZ
2804.. option:: merge_blktrace_scalars=float_list
2805
2806 This is a percentage based option that is index paired with the list of
2807 files passed to :option:`read_iolog`. When merging is performed, scale
2808 the time of each event by the corresponding amount. For example,
2809 ``--merge_blktrace_scalars="50:100"`` runs the first trace in halftime
2810 and the second trace in realtime. This knob is separately tunable from
2811 :option:`replay_time_scale` which scales the trace during runtime and
2812 does not change the output of the merge unlike this option.
2813
55bfd8c8
DZ
2814.. option:: merge_blktrace_iters=float_list
2815
2816 This is a whole number option that is index paired with the list of files
2817 passed to :option:`read_iolog`. When merging is performed, run each trace
2818 for the specified number of iterations. For example,
2819 ``--merge_blktrace_iters="2:1"`` runs the first trace for two iterations
2820 and the second trace for one iteration.
2821
589e88b7 2822.. option:: replay_no_stall=bool
71bfa161 2823
f80dba8d 2824 When replaying I/O with :option:`read_iolog` the default behavior is to
22413915 2825 attempt to respect the timestamps within the log and replay them with the
f80dba8d
MT
2826 appropriate delay between IOPS. By setting this variable fio will not
2827 respect the timestamps and attempt to replay them as fast as possible while
2828 still respecting ordering. The result is the same I/O pattern to a given
2829 device, but different timings.
71bfa161 2830
6dd7fa77
JA
2831.. option:: replay_time_scale=int
2832
2833 When replaying I/O with :option:`read_iolog`, fio will honor the
2834 original timing in the trace. With this option, it's possible to scale
2835 the time. It's a percentage option, if set to 50 it means run at 50%
2836 the original IO rate in the trace. If set to 200, run at twice the
2837 original IO rate. Defaults to 100.
2838
f80dba8d 2839.. option:: replay_redirect=str
b4692828 2840
f80dba8d
MT
2841 While replaying I/O patterns using :option:`read_iolog` the default behavior
2842 is to replay the IOPS onto the major/minor device that each IOP was recorded
2843 from. This is sometimes undesirable because on a different machine those
2844 major/minor numbers can map to a different device. Changing hardware on the
2845 same system can also result in a different major/minor mapping.
730bd7d9 2846 ``replay_redirect`` causes all I/Os to be replayed onto the single specified
f80dba8d 2847 device regardless of the device it was recorded
9207a0cb 2848 from. i.e. :option:`replay_redirect`\= :file:`/dev/sdc` would cause all I/O
f80dba8d
MT
2849 in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means
2850 multiple devices will be replayed onto a single device, if the trace
2851 contains multiple devices. If you want multiple devices to be replayed
2852 concurrently to multiple redirected devices you must blkparse your trace
2853 into separate traces and replay them with independent fio invocations.
2854 Unfortunately this also breaks the strict time ordering between multiple
2855 device accesses.
71bfa161 2856
f80dba8d 2857.. option:: replay_align=int
74929ac2 2858
350a535d
DZ
2859 Force alignment of the byte offsets in a trace to this value. The value
2860 must be a power of 2.
3c54bc46 2861
f80dba8d 2862.. option:: replay_scale=int
3c54bc46 2863
350a535d
DZ
2864 Scale byte offsets down by this factor when replaying traces. Should most
2865 likely use :option:`replay_align` as well.
3c54bc46 2866
38f68906
JA
2867.. option:: replay_skip=str
2868
2869 Sometimes it's useful to skip certain IO types in a replay trace.
2870 This could be, for instance, eliminating the writes in the trace.
2871 Or not replaying the trims/discards, if you are redirecting to
2872 a device that doesn't support them. This option takes a comma
2873 separated list of read, write, trim, sync.
2874
3c54bc46 2875
f80dba8d
MT
2876Threads, processes and job synchronization
2877~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3c54bc46 2878
f80dba8d 2879.. option:: thread
3c54bc46 2880
730bd7d9
SW
2881 Fio defaults to creating jobs by using fork, however if this option is
2882 given, fio will create jobs by using POSIX Threads' function
2883 :manpage:`pthread_create(3)` to create threads instead.
71bfa161 2884
f80dba8d 2885.. option:: wait_for=str
74929ac2 2886
730bd7d9
SW
2887 If set, the current job won't be started until all workers of the specified
2888 waitee job are done.
74929ac2 2889
f80dba8d
MT
2890 ``wait_for`` operates on the job name basis, so there are a few
2891 limitations. First, the waitee must be defined prior to the waiter job
2892 (meaning no forward references). Second, if a job is being referenced as a
2893 waitee, it must have a unique name (no duplicate waitees).
74929ac2 2894
f80dba8d 2895.. option:: nice=int
892a6ffc 2896
f80dba8d 2897 Run the job with the given nice value. See man :manpage:`nice(2)`.
892a6ffc 2898
f80dba8d
MT
2899 On Windows, values less than -15 set the process class to "High"; -1 through
2900 -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle"
2901 priority class.
74929ac2 2902
f80dba8d 2903.. option:: prio=int
71bfa161 2904
f80dba8d
MT
2905 Set the I/O priority value of this job. Linux limits us to a positive value
2906 between 0 and 7, with 0 being the highest. See man
2907 :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating
b2a432bf
PC
2908 systems since meaning of priority may differ. For per-command priority
2909 setting, see I/O engine specific `cmdprio_percentage` and `hipri_percentage`
2910 options.
71bfa161 2911
f80dba8d 2912.. option:: prioclass=int
d59aa780 2913
b2a432bf
PC
2914 Set the I/O priority class. See man :manpage:`ionice(1)`. For per-command
2915 priority setting, see I/O engine specific `cmdprio_percentage` and
2916 `hipri_percentage` options.
d59aa780 2917
f80dba8d 2918.. option:: cpus_allowed=str
6d500c2e 2919
730bd7d9 2920 Controls the same options as :option:`cpumask`, but accepts a textual
b570e037
SW
2921 specification of the permitted CPUs instead and CPUs are indexed from 0. So
2922 to use CPUs 0 and 5 you would specify ``cpus_allowed=0,5``. This option also
2923 allows a range of CPUs to be specified -- say you wanted a binding to CPUs
2924 0, 5, and 8 to 15, you would set ``cpus_allowed=0,5,8-15``.
2925
2926 On Windows, when ``cpus_allowed`` is unset only CPUs from fio's current
2927 processor group will be used and affinity settings are inherited from the
2928 system. An fio build configured to target Windows 7 makes options that set
2929 CPUs processor group aware and values will set both the processor group
2930 and a CPU from within that group. For example, on a system where processor
2931 group 0 has 40 CPUs and processor group 1 has 32 CPUs, ``cpus_allowed``
2932 values between 0 and 39 will bind CPUs from processor group 0 and
2933 ``cpus_allowed`` values between 40 and 71 will bind CPUs from processor
2934 group 1. When using ``cpus_allowed_policy=shared`` all CPUs specified by a
2935 single ``cpus_allowed`` option must be from the same processor group. For
2936 Windows fio builds not built for Windows 7, CPUs will only be selected from
2937 (and be relative to) whatever processor group fio happens to be running in
2938 and CPUs from other processor groups cannot be used.
6d500c2e 2939
f80dba8d 2940.. option:: cpus_allowed_policy=str
6d500c2e 2941
f80dba8d 2942 Set the policy of how fio distributes the CPUs specified by
730bd7d9 2943 :option:`cpus_allowed` or :option:`cpumask`. Two policies are supported:
6d500c2e 2944
f80dba8d
MT
2945 **shared**
2946 All jobs will share the CPU set specified.
2947 **split**
2948 Each job will get a unique CPU from the CPU set.
6d500c2e 2949
22413915 2950 **shared** is the default behavior, if the option isn't specified. If
b21fc93f 2951 **split** is specified, then fio will assign one cpu per job. If not
f80dba8d
MT
2952 enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs
2953 in the set.
6d500c2e 2954
b570e037
SW
2955.. option:: cpumask=int
2956
2957 Set the CPU affinity of this job. The parameter given is a bit mask of
2958 allowed CPUs the job may run on. So if you want the allowed CPUs to be 1
2959 and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man
2960 :manpage:`sched_setaffinity(2)`. This may not work on all supported
2961 operating systems or kernel versions. This option doesn't work well for a
2962 higher CPU count than what you can store in an integer mask, so it can only
2963 control cpus 1-32. For boxes with larger CPU counts, use
2964 :option:`cpus_allowed`.
2965
f80dba8d 2966.. option:: numa_cpu_nodes=str
6d500c2e 2967
f80dba8d
MT
2968 Set this job running on specified NUMA nodes' CPUs. The arguments allow
2969 comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable
ac8ca2af 2970 NUMA options support, fio must be built on a system with libnuma-dev(el)
f80dba8d 2971 installed.
61b9861d 2972
f80dba8d 2973.. option:: numa_mem_policy=str
61b9861d 2974
f80dba8d
MT
2975 Set this job's memory policy and corresponding NUMA nodes. Format of the
2976 arguments::
5c94b008 2977
f80dba8d 2978 <mode>[:<nodelist>]
ce35b1ec 2979
804c0839 2980 ``mode`` is one of the following memory policies: ``default``, ``prefer``,
730bd7d9
SW
2981 ``bind``, ``interleave`` or ``local``. For ``default`` and ``local`` memory
2982 policies, no node needs to be specified. For ``prefer``, only one node is
2983 allowed. For ``bind`` and ``interleave`` the ``nodelist`` may be as
2984 follows: a comma delimited list of numbers, A-B ranges, or `all`.
71bfa161 2985
f80dba8d 2986.. option:: cgroup=str
390b1537 2987
f80dba8d
MT
2988 Add job to this control group. If it doesn't exist, it will be created. The
2989 system must have a mounted cgroup blkio mount point for this to work. If
2990 your system doesn't have it mounted, you can do so with::
5af1c6f3 2991
f80dba8d 2992 # mount -t cgroup -o blkio none /cgroup
5af1c6f3 2993
f80dba8d 2994.. option:: cgroup_weight=int
5af1c6f3 2995
f80dba8d
MT
2996 Set the weight of the cgroup to this value. See the documentation that comes
2997 with the kernel, allowed values are in the range of 100..1000.
a086c257 2998
f80dba8d 2999.. option:: cgroup_nodelete=bool
8c07860d 3000
f80dba8d
MT
3001 Normally fio will delete the cgroups it has created after the job
3002 completion. To override this behavior and to leave cgroups around after the
3003 job completion, set ``cgroup_nodelete=1``. This can be useful if one wants
3004 to inspect various cgroup files after job completion. Default: false.
8c07860d 3005
f80dba8d 3006.. option:: flow_id=int
8c07860d 3007
f80dba8d
MT
3008 The ID of the flow. If not specified, it defaults to being a global
3009 flow. See :option:`flow`.
1907dbc6 3010
f80dba8d 3011.. option:: flow=int
71bfa161 3012
f80dba8d
MT
3013 Weight in token-based flow control. If this value is used, then there is a
3014 'flow counter' which is used to regulate the proportion of activity between
3015 two or more jobs. Fio attempts to keep this flow counter near zero. The
3016 ``flow`` parameter stands for how much should be added or subtracted to the
3017 flow counter on each iteration of the main I/O loop. That is, if one job has
3018 ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8
3019 ratio in how much one runs vs the other.
71bfa161 3020
f80dba8d 3021.. option:: flow_sleep=int
82407585 3022
d4e74fda
DB
3023 The period of time, in microseconds, to wait after the flow counter
3024 has exceeded its proportion before retrying operations.
82407585 3025
f80dba8d 3026.. option:: stonewall, wait_for_previous
82407585 3027
f80dba8d
MT
3028 Wait for preceding jobs in the job file to exit, before starting this
3029 one. Can be used to insert serialization points in the job file. A stone
3030 wall also implies starting a new reporting group, see
3031 :option:`group_reporting`.
3032
3033.. option:: exitall
3034
64402a8a
HW
3035 By default, fio will continue running all other jobs when one job finishes.
3036 Sometimes this is not the desired action. Setting ``exitall`` will instead
3037 make fio terminate all jobs in the same group, as soon as one job of that
3038 group finishes.
3039
3040.. option:: exit_what
3041
3042 By default, fio will continue running all other jobs when one job finishes.
3043 Sometimes this is not the desired action. Setting ``exit_all`` will
3044 instead make fio terminate all jobs in the same group. The option
3045 ``exit_what`` allows to control which jobs get terminated when ``exitall`` is
3046 enabled. The default is ``group`` and does not change the behaviour of
3047 ``exitall``. The setting ``all`` terminates all jobs. The setting ``stonewall``
3048 terminates all currently running jobs across all groups and continues execution
3049 with the next stonewalled group.
f80dba8d
MT
3050
3051.. option:: exec_prerun=str
3052
3053 Before running this job, issue the command specified through
3054 :manpage:`system(3)`. Output is redirected in a file called
3055 :file:`jobname.prerun.txt`.
3056
3057.. option:: exec_postrun=str
3058
3059 After the job completes, issue the command specified though
3060 :manpage:`system(3)`. Output is redirected in a file called
3061 :file:`jobname.postrun.txt`.
3062
3063.. option:: uid=int
3064
3065 Instead of running as the invoking user, set the user ID to this value
3066 before the thread/process does any work.
3067
3068.. option:: gid=int
3069
3070 Set group ID, see :option:`uid`.
3071
3072
3073Verification
3074~~~~~~~~~~~~
3075
3076.. option:: verify_only
3077
3078 Do not perform specified workload, only verify data still matches previous
3079 invocation of this workload. This option allows one to check data multiple
3080 times at a later date without overwriting it. This option makes sense only
3081 for workloads that write data, and does not support workloads with the
3082 :option:`time_based` option set.
3083
3084.. option:: do_verify=bool
3085
3086 Run the verify phase after a write phase. Only valid if :option:`verify` is
3087 set. Default: true.
3088
3089.. option:: verify=str
3090
3091 If writing to a file, fio can verify the file contents after each iteration
3092 of the job. Each verification method also implies verification of special
3093 header, which is written to the beginning of each block. This header also
3094 includes meta information, like offset of the block, block number, timestamp
3095 when block was written, etc. :option:`verify` can be combined with
3096 :option:`verify_pattern` option. The allowed values are:
3097
3098 **md5**
3099 Use an md5 sum of the data area and store it in the header of
3100 each block.
3101
3102 **crc64**
3103 Use an experimental crc64 sum of the data area and store it in the
3104 header of each block.
3105
3106 **crc32c**
a5896300
SW
3107 Use a crc32c sum of the data area and store it in the header of
3108 each block. This will automatically use hardware acceleration
3109 (e.g. SSE4.2 on an x86 or CRC crypto extensions on ARM64) but will
3110 fall back to software crc32c if none is found. Generally the
804c0839 3111 fastest checksum fio supports when hardware accelerated.
f80dba8d
MT
3112
3113 **crc32c-intel**
a5896300 3114 Synonym for crc32c.
f80dba8d
MT
3115
3116 **crc32**
3117 Use a crc32 sum of the data area and store it in the header of each
3118 block.
3119
3120 **crc16**
3121 Use a crc16 sum of the data area and store it in the header of each
3122 block.
3123
3124 **crc7**
3125 Use a crc7 sum of the data area and store it in the header of each
3126 block.
3127
3128 **xxhash**
3129 Use xxhash as the checksum function. Generally the fastest software
3130 checksum that fio supports.
3131
3132 **sha512**
3133 Use sha512 as the checksum function.
3134
3135 **sha256**
3136 Use sha256 as the checksum function.
3137
3138 **sha1**
3139 Use optimized sha1 as the checksum function.
82407585 3140
ae3a5acc
JA
3141 **sha3-224**
3142 Use optimized sha3-224 as the checksum function.
3143
3144 **sha3-256**
3145 Use optimized sha3-256 as the checksum function.
3146
3147 **sha3-384**
3148 Use optimized sha3-384 as the checksum function.
3149
3150 **sha3-512**
3151 Use optimized sha3-512 as the checksum function.
3152
f80dba8d
MT
3153 **meta**
3154 This option is deprecated, since now meta information is included in
3155 generic verification header and meta verification happens by
3156 default. For detailed information see the description of the
3157 :option:`verify` setting. This option is kept because of
3158 compatibility's sake with old configurations. Do not use it.
3159
3160 **pattern**
3161 Verify a strict pattern. Normally fio includes a header with some
3162 basic information and checksumming, but if this option is set, only
3163 the specific pattern set with :option:`verify_pattern` is verified.
3164
3165 **null**
3166 Only pretend to verify. Useful for testing internals with
9207a0cb 3167 :option:`ioengine`\=null, not for much else.
f80dba8d
MT
3168
3169 This option can be used for repeated burn-in tests of a system to make sure
3170 that the written data is also correctly read back. If the data direction
3171 given is a read or random read, fio will assume that it should verify a
3172 previously written file. If the data direction includes any form of write,
3173 the verify will be of the newly written data.
3174
47e6a6e5
SW
3175 To avoid false verification errors, do not use the norandommap option when
3176 verifying data with async I/O engines and I/O depths > 1. Or use the
3177 norandommap and the lfsr random generator together to avoid writing to the
3178 same offset with muliple outstanding I/Os.
3179
f80dba8d
MT
3180.. option:: verify_offset=int
3181
3182 Swap the verification header with data somewhere else in the block before
3183 writing. It is swapped back before verifying.
3184
3185.. option:: verify_interval=int
3186
3187 Write the verification header at a finer granularity than the
3188 :option:`blocksize`. It will be written for chunks the size of
3189 ``verify_interval``. :option:`blocksize` should divide this evenly.
3190
3191.. option:: verify_pattern=str
3192
3193 If set, fio will fill the I/O buffers with this pattern. Fio defaults to
3194 filling with totally random bytes, but sometimes it's interesting to fill
3195 with a known pattern for I/O verification purposes. Depending on the width
730bd7d9 3196 of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time (it can
f80dba8d
MT
3197 be either a decimal or a hex number). The ``verify_pattern`` if larger than
3198 a 32-bit quantity has to be a hex number that starts with either "0x" or
3199 "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o
3200 format, which means that for each block offset will be written and then
3201 verified back, e.g.::
61b9861d
RP
3202
3203 verify_pattern=%o
3204
f80dba8d
MT
3205 Or use combination of everything::
3206
61b9861d 3207 verify_pattern=0xff%o"abcd"-12
e28218f3 3208
f80dba8d
MT
3209.. option:: verify_fatal=bool
3210
3211 Normally fio will keep checking the entire contents before quitting on a
3212 block verification failure. If this option is set, fio will exit the job on
3213 the first observed failure. Default: false.
3214
3215.. option:: verify_dump=bool
3216
3217 If set, dump the contents of both the original data block and the data block
3218 we read off disk to files. This allows later analysis to inspect just what
3219 kind of data corruption occurred. Off by default.
3220
3221.. option:: verify_async=int
3222
3223 Fio will normally verify I/O inline from the submitting thread. This option
3224 takes an integer describing how many async offload threads to create for I/O
3225 verification instead, causing fio to offload the duty of verifying I/O
3226 contents to one or more separate threads. If using this offload option, even
3227 sync I/O engines can benefit from using an :option:`iodepth` setting higher
3228 than 1, as it allows them to have I/O in flight while verifies are running.
d7e6ea1c 3229 Defaults to 0 async threads, i.e. verification is not asynchronous.
f80dba8d
MT
3230
3231.. option:: verify_async_cpus=str
3232
3233 Tell fio to set the given CPU affinity on the async I/O verification
3234 threads. See :option:`cpus_allowed` for the format used.
3235
3236.. option:: verify_backlog=int
3237
3238 Fio will normally verify the written contents of a job that utilizes verify
3239 once that job has completed. In other words, everything is written then
3240 everything is read back and verified. You may want to verify continually
3241 instead for a variety of reasons. Fio stores the meta data associated with
3242 an I/O block in memory, so for large verify workloads, quite a bit of memory
3243 would be used up holding this meta data. If this option is enabled, fio will
3244 write only N blocks before verifying these blocks.
3245
3246.. option:: verify_backlog_batch=int
3247
3248 Control how many blocks fio will verify if :option:`verify_backlog` is
3249 set. If not set, will default to the value of :option:`verify_backlog`
3250 (meaning the entire queue is read back and verified). If
3251 ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all
3252 blocks will be verified, if ``verify_backlog_batch`` is larger than
3253 :option:`verify_backlog`, some blocks will be verified more than once.
3254
3255.. option:: verify_state_save=bool
3256
3257 When a job exits during the write phase of a verify workload, save its
3258 current state. This allows fio to replay up until that point, if the verify
3259 state is loaded for the verify read phase. The format of the filename is,
3260 roughly::
3261
f50fbdda 3262 <type>-<jobname>-<jobindex>-verify.state.
f80dba8d
MT
3263
3264 <type> is "local" for a local run, "sock" for a client/server socket
3265 connection, and "ip" (192.168.0.1, for instance) for a networked
d7e6ea1c 3266 client/server connection. Defaults to true.
f80dba8d
MT
3267
3268.. option:: verify_state_load=bool
3269
3270 If a verify termination trigger was used, fio stores the current write state
3271 of each thread. This can be used at verification time so that fio knows how
3272 far it should verify. Without this information, fio will run a full
a47b697c
SW
3273 verification pass, according to the settings in the job file used. Default
3274 false.
f80dba8d
MT
3275
3276.. option:: trim_percentage=int
3277
3278 Number of verify blocks to discard/trim.
3279
3280.. option:: trim_verify_zero=bool
3281
22413915 3282 Verify that trim/discarded blocks are returned as zeros.
f80dba8d
MT
3283
3284.. option:: trim_backlog=int
3285
5cfd1e9a 3286 Trim after this number of blocks are written.
f80dba8d
MT
3287
3288.. option:: trim_backlog_batch=int
3289
3290 Trim this number of I/O blocks.
3291
3292.. option:: experimental_verify=bool
3293
3294 Enable experimental verification.
3295
f80dba8d
MT
3296Steady state
3297~~~~~~~~~~~~
3298
3299.. option:: steadystate=str:float, ss=str:float
3300
3301 Define the criterion and limit for assessing steady state performance. The
3302 first parameter designates the criterion whereas the second parameter sets
3303 the threshold. When the criterion falls below the threshold for the
3304 specified duration, the job will stop. For example, `iops_slope:0.1%` will
3305 direct fio to terminate the job when the least squares regression slope
3306 falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled
3307 this will apply to all jobs in the group. Below is the list of available
3308 steady state assessment criteria. All assessments are carried out using only
3309 data from the rolling collection window. Threshold limits can be expressed
3310 as a fixed value or as a percentage of the mean in the collection window.
3311
1cb049d9
VF
3312 When using this feature, most jobs should include the :option:`time_based`
3313 and :option:`runtime` options or the :option:`loops` option so that fio does not
3314 stop running after it has covered the full size of the specified file(s) or device(s).
3315
f80dba8d
MT
3316 **iops**
3317 Collect IOPS data. Stop the job if all individual IOPS measurements
3318 are within the specified limit of the mean IOPS (e.g., ``iops:2``
3319 means that all individual IOPS values must be within 2 of the mean,
3320 whereas ``iops:0.2%`` means that all individual IOPS values must be
3321 within 0.2% of the mean IOPS to terminate the job).
3322
3323 **iops_slope**
3324 Collect IOPS data and calculate the least squares regression
3325 slope. Stop the job if the slope falls below the specified limit.
3326
3327 **bw**
3328 Collect bandwidth data. Stop the job if all individual bandwidth
3329 measurements are within the specified limit of the mean bandwidth.
3330
3331 **bw_slope**
3332 Collect bandwidth data and calculate the least squares regression
3333 slope. Stop the job if the slope falls below the specified limit.
3334
3335.. option:: steadystate_duration=time, ss_dur=time
3336
3337 A rolling window of this duration will be used to judge whether steady state
3338 has been reached. Data will be collected once per second. The default is 0
f75ede1d 3339 which disables steady state detection. When the unit is omitted, the
947e0fe0 3340 value is interpreted in seconds.
f80dba8d
MT
3341
3342.. option:: steadystate_ramp_time=time, ss_ramp=time
3343
3344 Allow the job to run for the specified duration before beginning data
3345 collection for checking the steady state job termination criterion. The
947e0fe0 3346 default is 0. When the unit is omitted, the value is interpreted in seconds.
f80dba8d
MT
3347
3348
3349Measurements and reporting
3350~~~~~~~~~~~~~~~~~~~~~~~~~~
3351
3352.. option:: per_job_logs=bool
3353
3354 If set, this generates bw/clat/iops log with per file private filenames. If
3355 not set, jobs with identical names will share the log filename. Default:
3356 true.
3357
3358.. option:: group_reporting
3359
3360 It may sometimes be interesting to display statistics for groups of jobs as
3361 a whole instead of for each individual job. This is especially true if
3362 :option:`numjobs` is used; looking at individual thread/process output
3363 quickly becomes unwieldy. To see the final report per-group instead of
3364 per-job, use :option:`group_reporting`. Jobs in a file will be part of the
3365 same reporting group, unless if separated by a :option:`stonewall`, or by
3366 using :option:`new_group`.
3367
3368.. option:: new_group
3369
3370 Start a new reporting group. See: :option:`group_reporting`. If not given,
3371 all jobs in a file will be part of the same reporting group, unless
3372 separated by a :option:`stonewall`.
3373
589e88b7 3374.. option:: stats=bool
8243be59
JA
3375
3376 By default, fio collects and shows final output results for all jobs
3377 that run. If this option is set to 0, then fio will ignore it in
3378 the final stat output.
3379
f80dba8d
MT
3380.. option:: write_bw_log=str
3381
3382 If given, write a bandwidth log for this job. Can be used to store data of
074f0817 3383 the bandwidth of the jobs in their lifetime.
f80dba8d 3384
074f0817
SW
3385 If no str argument is given, the default filename of
3386 :file:`jobname_type.x.log` is used. Even when the argument is given, fio
3387 will still append the type of log. So if one specifies::
3388
3389 write_bw_log=foo
f80dba8d 3390
074f0817
SW
3391 The actual log name will be :file:`foo_bw.x.log` where `x` is the index
3392 of the job (`1..N`, where `N` is the number of jobs). If
3393 :option:`per_job_logs` is false, then the filename will not include the
3394 `.x` job index.
e3cedca7 3395
074f0817
SW
3396 The included :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these
3397 text files into nice graphs. See `Log File Formats`_ for how data is
3398 structured within the file.
3399
3400.. option:: write_lat_log=str
e3cedca7 3401
074f0817 3402 Same as :option:`write_bw_log`, except this option creates I/O
77b7e675
SW
3403 submission (e.g., :file:`name_slat.x.log`), completion (e.g.,
3404 :file:`name_clat.x.log`), and total (e.g., :file:`name_lat.x.log`)
074f0817
SW
3405 latency files instead. See :option:`write_bw_log` for details about
3406 the filename format and `Log File Formats`_ for how data is structured
3407 within the files.
be4ecfdf 3408
f80dba8d 3409.. option:: write_hist_log=str
06842027 3410
074f0817 3411 Same as :option:`write_bw_log` but writes an I/O completion latency
77b7e675 3412 histogram file (e.g., :file:`name_hist.x.log`) instead. Note that this
074f0817
SW
3413 file will be empty unless :option:`log_hist_msec` has also been set.
3414 See :option:`write_bw_log` for details about the filename format and
3415 `Log File Formats`_ for how data is structured within the file.
06842027 3416
f80dba8d 3417.. option:: write_iops_log=str
06842027 3418
074f0817 3419 Same as :option:`write_bw_log`, but writes an IOPS file (e.g.
15417073
SW
3420 :file:`name_iops.x.log`) instead. Because fio defaults to individual
3421 I/O logging, the value entry in the IOPS log will be 1 unless windowed
3422 logging (see :option:`log_avg_msec`) has been enabled. See
3423 :option:`write_bw_log` for details about the filename format and `Log
3424 File Formats`_ for how data is structured within the file.
06842027 3425
f80dba8d 3426.. option:: log_avg_msec=int
06842027 3427
f80dba8d
MT
3428 By default, fio will log an entry in the iops, latency, or bw log for every
3429 I/O that completes. When writing to the disk log, that can quickly grow to a
3430 very large size. Setting this option makes fio average the each log entry
3431 over the specified period of time, reducing the resolution of the log. See
3432 :option:`log_max_value` as well. Defaults to 0, logging all entries.
6fc82095 3433 Also see `Log File Formats`_.
06842027 3434
f80dba8d 3435.. option:: log_hist_msec=int
06842027 3436
f80dba8d
MT
3437 Same as :option:`log_avg_msec`, but logs entries for completion latency
3438 histograms. Computing latency percentiles from averages of intervals using
c60ebc45 3439 :option:`log_avg_msec` is inaccurate. Setting this option makes fio log
f80dba8d
MT
3440 histogram entries over the specified period of time, reducing log sizes for
3441 high IOPS devices while retaining percentile accuracy. See
074f0817
SW
3442 :option:`log_hist_coarseness` and :option:`write_hist_log` as well.
3443 Defaults to 0, meaning histogram logging is disabled.
06842027 3444
f80dba8d 3445.. option:: log_hist_coarseness=int
06842027 3446
f80dba8d
MT
3447 Integer ranging from 0 to 6, defining the coarseness of the resolution of
3448 the histogram logs enabled with :option:`log_hist_msec`. For each increment
3449 in coarseness, fio outputs half as many bins. Defaults to 0, for which
074f0817
SW
3450 histogram logs contain 1216 latency bins. See :option:`write_hist_log`
3451 and `Log File Formats`_.
8b28bd41 3452
f80dba8d 3453.. option:: log_max_value=bool
66c098b8 3454
f80dba8d
MT
3455 If :option:`log_avg_msec` is set, fio logs the average over that window. If
3456 you instead want to log the maximum value, set this option to 1. Defaults to
3457 0, meaning that averaged values are logged.
a696fa2a 3458
589e88b7 3459.. option:: log_offset=bool
a696fa2a 3460
f80dba8d 3461 If this is set, the iolog options will include the byte offset for the I/O
5a83478f
SW
3462 entry as well as the other data values. Defaults to 0 meaning that
3463 offsets are not present in logs. Also see `Log File Formats`_.
71bfa161 3464
f80dba8d 3465.. option:: log_compression=int
7de87099 3466
f80dba8d
MT
3467 If this is set, fio will compress the I/O logs as it goes, to keep the
3468 memory footprint lower. When a log reaches the specified size, that chunk is
3469 removed and compressed in the background. Given that I/O logs are fairly
3470 highly compressible, this yields a nice memory savings for longer runs. The
3471 downside is that the compression will consume some background CPU cycles, so
3472 it may impact the run. This, however, is also true if the logging ends up
3473 consuming most of the system memory. So pick your poison. The I/O logs are
3474 saved normally at the end of a run, by decompressing the chunks and storing
3475 them in the specified log file. This feature depends on the availability of
3476 zlib.
e0b0d892 3477
f80dba8d 3478.. option:: log_compression_cpus=str
e0b0d892 3479
f80dba8d
MT
3480 Define the set of CPUs that are allowed to handle online log compression for
3481 the I/O jobs. This can provide better isolation between performance
0cf90a62
SW
3482 sensitive jobs, and background compression work. See
3483 :option:`cpus_allowed` for the format used.
9e684a49 3484
f80dba8d 3485.. option:: log_store_compressed=bool
9e684a49 3486
f80dba8d
MT
3487 If set, fio will store the log files in a compressed format. They can be
3488 decompressed with fio, using the :option:`--inflate-log` command line
3489 parameter. The files will be stored with a :file:`.fz` suffix.
9e684a49 3490
f80dba8d 3491.. option:: log_unix_epoch=bool
9e684a49 3492
f80dba8d
MT
3493 If set, fio will log Unix timestamps to the log files produced by enabling
3494 write_type_log for each log type, instead of the default zero-based
3495 timestamps.
3496
3497.. option:: block_error_percentiles=bool
3498
3499 If set, record errors in trim block-sized units from writes and trims and
3500 output a histogram of how many trims it took to get to errors, and what kind
3501 of error was encountered.
3502
3503.. option:: bwavgtime=int
3504
3505 Average the calculated bandwidth over the given time. Value is specified in
3506 milliseconds. If the job also does bandwidth logging through
3507 :option:`write_bw_log`, then the minimum of this option and
3508 :option:`log_avg_msec` will be used. Default: 500ms.
3509
3510.. option:: iopsavgtime=int
3511
3512 Average the calculated IOPS over the given time. Value is specified in
3513 milliseconds. If the job also does IOPS logging through
3514 :option:`write_iops_log`, then the minimum of this option and
3515 :option:`log_avg_msec` will be used. Default: 500ms.
3516
3517.. option:: disk_util=bool
3518
3519 Generate disk utilization statistics, if the platform supports it.
3520 Default: true.
3521
3522.. option:: disable_lat=bool
3523
3524 Disable measurements of total latency numbers. Useful only for cutting back
3525 the number of calls to :manpage:`gettimeofday(2)`, as that does impact
3526 performance at really high IOPS rates. Note that to really get rid of a
3527 large amount of these calls, this option must be used with
f75ede1d 3528 :option:`disable_slat` and :option:`disable_bw_measurement` as well.
f80dba8d
MT
3529
3530.. option:: disable_clat=bool
3531
3532 Disable measurements of completion latency numbers. See
3533 :option:`disable_lat`.
3534
3535.. option:: disable_slat=bool
3536
3537 Disable measurements of submission latency numbers. See
f50fbdda 3538 :option:`disable_lat`.
f80dba8d 3539
f75ede1d 3540.. option:: disable_bw_measurement=bool, disable_bw=bool
f80dba8d
MT
3541
3542 Disable measurements of throughput/bandwidth numbers. See
3543 :option:`disable_lat`.
3544
dd39b9ce
VF
3545.. option:: slat_percentiles=bool
3546
3547 Report submission latency percentiles. Submission latency is not recorded
3548 for synchronous ioengines.
3549
f80dba8d
MT
3550.. option:: clat_percentiles=bool
3551
dd39b9ce 3552 Report completion latency percentiles.
b599759b
JA
3553
3554.. option:: lat_percentiles=bool
3555
dd39b9ce
VF
3556 Report total latency percentiles. Total latency is the sum of submission
3557 latency and completion latency.
f80dba8d
MT
3558
3559.. option:: percentile_list=float_list
3560
dd39b9ce
VF
3561 Overwrite the default list of percentiles for latencies and the block error
3562 histogram. Each number is a floating point number in the range (0,100], and
3563 the maximum length of the list is 20. Use ``:`` to separate the numbers. For
c32ba107 3564 example, ``--percentile_list=99.5:99.9`` will cause fio to report the
dd39b9ce
VF
3565 latency durations below which 99.5% and 99.9% of the observed latencies fell,
3566 respectively.
f80dba8d 3567
e883cb35
JF
3568.. option:: significant_figures=int
3569
c32ba107
JA
3570 If using :option:`--output-format` of `normal`, set the significant
3571 figures to this value. Higher values will yield more precise IOPS and
3572 throughput units, while lower values will round. Requires a minimum
3573 value of 1 and a maximum value of 10. Defaults to 4.
e883cb35 3574
f80dba8d
MT
3575
3576Error handling
3577~~~~~~~~~~~~~~
3578
3579.. option:: exitall_on_error
3580
3581 When one job finishes in error, terminate the rest. The default is to wait
3582 for each job to finish.
3583
3584.. option:: continue_on_error=str
3585
3586 Normally fio will exit the job on the first observed failure. If this option
3587 is set, fio will continue the job when there is a 'non-fatal error' (EIO or
3588 EILSEQ) until the runtime is exceeded or the I/O size specified is
3589 completed. If this option is used, there are two more stats that are
3590 appended, the total error count and the first error. The error field given
3591 in the stats is the first error that was hit during the run.
3592
3593 The allowed values are:
3594
3595 **none**
3596 Exit on any I/O or verify errors.
3597
3598 **read**
3599 Continue on read errors, exit on all others.
3600
3601 **write**
3602 Continue on write errors, exit on all others.
3603
3604 **io**
3605 Continue on any I/O error, exit on all others.
3606
3607 **verify**
3608 Continue on verify errors, exit on all others.
3609
3610 **all**
3611 Continue on all errors.
3612
3613 **0**
3614 Backward-compatible alias for 'none'.
3615
3616 **1**
3617 Backward-compatible alias for 'all'.
3618
3619.. option:: ignore_error=str
3620
3621 Sometimes you want to ignore some errors during test in that case you can
a35ef7cb
TK
3622 specify error list for each error type, instead of only being able to
3623 ignore the default 'non-fatal error' using :option:`continue_on_error`.
f80dba8d
MT
3624 ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for
3625 given error type is separated with ':'. Error may be symbol ('ENOSPC',
3626 'ENOMEM') or integer. Example::
3627
3628 ignore_error=EAGAIN,ENOSPC:122
3629
3630 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from
a35ef7cb
TK
3631 WRITE. This option works by overriding :option:`continue_on_error` with
3632 the list of errors for each error type if any.
f80dba8d
MT
3633
3634.. option:: error_dump=bool
3635
3636 If set dump every error even if it is non fatal, true by default. If
3637 disabled only fatal error will be dumped.
3638
f75ede1d
SW
3639Running predefined workloads
3640----------------------------
3641
3642Fio includes predefined profiles that mimic the I/O workloads generated by
3643other tools.
3644
3645.. option:: profile=str
3646
3647 The predefined workload to run. Current profiles are:
3648
3649 **tiobench**
3650 Threaded I/O bench (tiotest/tiobench) like workload.
3651
3652 **act**
3653 Aerospike Certification Tool (ACT) like workload.
3654
3655To view a profile's additional options use :option:`--cmdhelp` after specifying
3656the profile. For example::
3657
f50fbdda 3658 $ fio --profile=act --cmdhelp
f75ede1d
SW
3659
3660Act profile options
3661~~~~~~~~~~~~~~~~~~~
3662
3663.. option:: device-names=str
3664 :noindex:
3665
3666 Devices to use.
3667
3668.. option:: load=int
3669 :noindex:
3670
3671 ACT load multiplier. Default: 1.
3672
3673.. option:: test-duration=time
3674 :noindex:
3675
947e0fe0
SW
3676 How long the entire test takes to run. When the unit is omitted, the value
3677 is given in seconds. Default: 24h.
f75ede1d
SW
3678
3679.. option:: threads-per-queue=int
3680 :noindex:
3681
f50fbdda 3682 Number of read I/O threads per device. Default: 8.
f75ede1d
SW
3683
3684.. option:: read-req-num-512-blocks=int
3685 :noindex:
3686
3687 Number of 512B blocks to read at the time. Default: 3.
3688
3689.. option:: large-block-op-kbytes=int
3690 :noindex:
3691
3692 Size of large block ops in KiB (writes). Default: 131072.
3693
3694.. option:: prep
3695 :noindex:
3696
3697 Set to run ACT prep phase.
3698
3699Tiobench profile options
3700~~~~~~~~~~~~~~~~~~~~~~~~
3701
3702.. option:: size=str
3703 :noindex:
3704
f50fbdda 3705 Size in MiB.
f75ede1d
SW
3706
3707.. option:: block=int
3708 :noindex:
3709
3710 Block size in bytes. Default: 4096.
3711
3712.. option:: numruns=int
3713 :noindex:
3714
3715 Number of runs.
3716
3717.. option:: dir=str
3718 :noindex:
3719
3720 Test directory.
3721
3722.. option:: threads=int
3723 :noindex:
3724
3725 Number of threads.
f80dba8d
MT
3726
3727Interpreting the output
3728-----------------------
3729
36214730
SW
3730..
3731 Example output was based on the following:
3732 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --time_based \
3733 --rate=1256k --bs=14K --name=quick --runtime=1s --name=mixed \
3734 --runtime=2m --rw=rw
3735
f80dba8d
MT
3736Fio spits out a lot of output. While running, fio will display the status of the
3737jobs created. An example of that would be::
3738
9d25d068 3739 Jobs: 1 (f=1): [_(1),M(1)][24.8%][r=20.5MiB/s,w=23.5MiB/s][r=82,w=94 IOPS][eta 01m:31s]
f80dba8d 3740
36214730
SW
3741The characters inside the first set of square brackets denote the current status of
3742each thread. The first character is the first job defined in the job file, and so
3743forth. The possible values (in typical life cycle order) are:
f80dba8d
MT
3744
3745+------+-----+-----------------------------------------------------------+
3746| Idle | Run | |
3747+======+=====+===========================================================+
3748| P | | Thread setup, but not started. |
3749+------+-----+-----------------------------------------------------------+
3750| C | | Thread created. |
3751+------+-----+-----------------------------------------------------------+
3752| I | | Thread initialized, waiting or generating necessary data. |
3753+------+-----+-----------------------------------------------------------+
3754| | p | Thread running pre-reading file(s). |
3755+------+-----+-----------------------------------------------------------+
36214730
SW
3756| | / | Thread is in ramp period. |
3757+------+-----+-----------------------------------------------------------+
f80dba8d
MT
3758| | R | Running, doing sequential reads. |
3759+------+-----+-----------------------------------------------------------+
3760| | r | Running, doing random reads. |
3761+------+-----+-----------------------------------------------------------+
3762| | W | Running, doing sequential writes. |
3763+------+-----+-----------------------------------------------------------+
3764| | w | Running, doing random writes. |
3765+------+-----+-----------------------------------------------------------+
3766| | M | Running, doing mixed sequential reads/writes. |
3767+------+-----+-----------------------------------------------------------+
3768| | m | Running, doing mixed random reads/writes. |
3769+------+-----+-----------------------------------------------------------+
36214730
SW
3770| | D | Running, doing sequential trims. |
3771+------+-----+-----------------------------------------------------------+
3772| | d | Running, doing random trims. |
3773+------+-----+-----------------------------------------------------------+
3774| | F | Running, currently waiting for :manpage:`fsync(2)`. |
f80dba8d
MT
3775+------+-----+-----------------------------------------------------------+
3776| | V | Running, doing verification of written data. |
3777+------+-----+-----------------------------------------------------------+
36214730
SW
3778| f | | Thread finishing. |
3779+------+-----+-----------------------------------------------------------+
f80dba8d
MT
3780| E | | Thread exited, not reaped by main thread yet. |
3781+------+-----+-----------------------------------------------------------+
36214730 3782| _ | | Thread reaped. |
f80dba8d
MT
3783+------+-----+-----------------------------------------------------------+
3784| X | | Thread reaped, exited with an error. |
3785+------+-----+-----------------------------------------------------------+
3786| K | | Thread reaped, exited due to signal. |
3787+------+-----+-----------------------------------------------------------+
3788
36214730
SW
3789..
3790 Example output was based on the following:
3791 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --runtime=58m \
3792 --time_based --rate=2512k --bs=256K --numjobs=10 \
3793 --name=readers --rw=read --name=writers --rw=write
3794
f80dba8d 3795Fio will condense the thread string as not to take up more space on the command
36214730 3796line than needed. For instance, if you have 10 readers and 10 writers running,
f80dba8d
MT
3797the output would look like this::
3798
9d25d068 3799 Jobs: 20 (f=20): [R(10),W(10)][4.0%][r=20.5MiB/s,w=23.5MiB/s][r=82,w=94 IOPS][eta 57m:36s]
f80dba8d 3800
36214730
SW
3801Note that the status string is displayed in order, so it's possible to tell which of
3802the jobs are currently doing what. In the example above this means that jobs 1--10
3803are readers and 11--20 are writers.
f80dba8d
MT
3804
3805The other values are fairly self explanatory -- number of threads currently
36214730
SW
3806running and doing I/O, the number of currently open files (f=), the estimated
3807completion percentage, the rate of I/O since last check (read speed listed first,
f50fbdda
TK
3808then write speed and optionally trim speed) in terms of bandwidth and IOPS,
3809and time to completion for the current running group. It's impossible to estimate
3810runtime of the following groups (if any).
36214730
SW
3811
3812..
3813 Example output was based on the following:
3814 TZ=UTC fio --iodepth=16 --ioengine=posixaio --filename=/tmp/fiofile \
3815