1fio(1) General Commands Manual fio(1)
2
6 fio - flexible I/O tester
7
9 fio [options] [jobfile]...
10
12 fio is a tool that will spawn a number of threads or processes doing a
13 particular type of I/O action as specified by the user. The typical
14 use of fio is to write a job file matching the I/O load one wants to
15 simulate.
16
18 --debug=type
19 Enable verbose tracing type of various fio actions. May be `all'
20 for all types or individual types separated by a comma (e.g.
21 `--debug=file,mem' will enable file and memory debugging).
22 `help' will list all available tracing options.
23 --parse-only
25 Parse options only, don't start any I/O.
26 --merge-blktrace-only
28 Merge blktraces only, don't start any I/O.
29 --output=filename
31 Write output to filename.
32 --output-format=format
34 Set the reporting format to `normal', `terse', `json', or
35 `json+'. Multiple formats can be selected, separate by a comma.
36 `terse' is a CSV based format. `json+' is like `json', except it
37 adds a full dump of the latency buckets.
38 --bandwidth-log
40 Generate aggregate bandwidth logs.
41 --minimal
43 Print statistics in a terse, semicolon-delimited format.
44 --append-terse
46 Print statistics in selected mode AND terse, semicolon-delimited
47 format. Deprecated, use --output-format instead to select mul‐
48 tiple formats.
49 --terse-version=version
51 Set terse version output format (default `3', or `2', `4', `5').
52 --version
54 Print version information and exit.
55 --help Print a summary of the command line options and exit.
57 --cpuclock-test
59 Perform test and validation of internal CPU clock.
60 --crctest=[test]
62 Test the speed of the built-in checksumming functions. If no ar‐
63 gument is given, all of them are tested. Alternatively, a comma
64 separated list can be passed, in which case the given ones are
65 tested.
66 --cmdhelp=command
68 Print help information for command. May be `all' for all com‐
69 mands.
70 --enghelp=[ioengine[,command]]
72 List all commands defined by ioengine, or print help for command
73 defined by ioengine. If no ioengine is given, list all available
74 ioengines.
75 --showcmd=jobfile
77 Convert jobfile to a set of command-line options.
78 --readonly
80 Turn on safety read-only checks, preventing writes and trims.
81 The --readonly option is an extra safety guard to prevent users
82 from accidentally starting a write or trim workload when that is
83 not desired. Fio will only modify the device under test if
84 `rw=write/randwrite/rw/randrw/trim/randtrim/trimwrite' is given.
85 This safety net can be used as an extra precaution.
86 --eta=when
88 Specifies when real-time ETA estimate should be printed. when
89 may be `always', `never' or `auto'. `auto' is the default, it
90 prints ETA when requested if the output is a TTY. `always' dis‐
91 regards the output type, and prints ETA when requested. `never'
92 never prints ETA.
93 --eta-interval=time
95 By default, fio requests client ETA status roughly every second.
96 With this option, the interval is configurable. Fio imposes a
97 minimum allowed time to avoid flooding the console, less than
98 250 msec is not supported.
99 --eta-newline=time
101 Force a new line for every time period passed. When the unit is
102 omitted, the value is interpreted in seconds.
103 --status-interval=time
105 Force a full status dump of cumulative (from job start) values
106 at time intervals. This option does *not* provide per-period
107 measurements. So values such as bandwidth are running averages.
108 When the time unit is omitted, time is interpreted in seconds.
109 Note that using this option with `--output-format=json' will
110 yield output that technically isn't valid json, since the output
111 will be collated sets of valid json. It will need to be split
112 into valid sets of json after the run.
113 --section=name
115 Only run specified section name in job file. Multiple sections
116 can be specified. The --section option allows one to combine
117 related jobs into one file. E.g. one job file could define
118 light, moderate, and heavy sections. Tell fio to run only the
119 "heavy" section by giving `--section=heavy' command line option.
120 One can also specify the "write" operations in one section and
121 "verify" operation in another section. The --section option only
122 applies to job sections. The reserved *global* section is always
123 parsed and used.
124 --alloc-size=kb
126 Allocate additional internal smalloc pools of size kb in KiB.
127 The --alloc-size option increases shared memory set aside for
128 use by fio. If running large jobs with randommap enabled, fio
129 can run out of memory. Smalloc is an internal allocator for
130 shared structures from a fixed size memory pool and can grow to
131 16 pools. The pool size defaults to 16MiB. NOTE: While running
132 `.fio_smalloc.*' backing store files are visible in `/tmp'.
133 --warnings-fatal
135 All fio parser warnings are fatal, causing fio to exit with an
136 error.
137 --max-jobs=nr
139 Set the maximum number of threads/processes to support to nr.
140 NOTE: On Linux, it may be necessary to increase the shared-mem‐
141 ory limit (`/proc/sys/kernel/shmmax') if fio runs into errors
142 while creating jobs.
143 --server=args
145 Start a backend server, with args specifying what to listen to.
146 See CLIENT/SERVER section.
147 --daemonize=pidfile
149 Background a fio server, writing the pid to the given pidfile
150 file.
151 --client=hostname
153 Instead of running the jobs locally, send and run them on the
154 given hostname or set of hostnames. See CLIENT/SERVER section.
155 --remote-config=file
157 Tell fio server to load this local file.
158 --idle-prof=option
160 Report CPU idleness. option is one of the following:
161 calibrate
163 Run unit work calibration only and exit.
164 system Show aggregate system idleness and unit work.
166 percpu As system but also show per CPU idleness.
168 --inflate-log=log
170 Inflate and output compressed log.
171 --trigger-file=file
173 Execute trigger command when file exists.
174 --trigger-timeout=time
176 Execute trigger at this time.
177 --trigger=command
179 Set this command as local trigger.
180 --trigger-remote=command
182 Set this command as remote trigger.
183 --aux-path=path
185 Use the directory specified by path for generated state files
186 instead of the current working directory.
187
189 Any parameters following the options will be assumed to be job files,
190 unless they match a job file parameter. Multiple job files can be
191 listed and each job file will be regarded as a separate group. Fio will
192 stonewall execution between each group.
193 Fio accepts one or more job files describing what it is supposed to do.
195 The job file format is the classic ini file, where the names enclosed
196 in [] brackets define the job name. You are free to use any ASCII name
197 you want, except *global* which has special meaning. Following the job
198 name is a sequence of zero or more parameters, one per line, that de‐
199 fine the behavior of the job. If the first character in a line is a ';'
200 or a '#', the entire line is discarded as a comment.
201 A *global* section sets defaults for the jobs described in that file. A
203 job may override a *global* section parameter, and a job file may even
204 have several *global* sections if so desired. A job is only affected by
205 a *global* section residing above it.
206 The --cmdhelp option also lists all options. If used with an command
208 argument, --cmdhelp will detail the given command.
209 See the `examples/' directory for inspiration on how to write job
211 files. Note the copyright and license requirements currently apply to
212 `examples/' files.
213 Note that the maximum length of a line in the job file is 8192 bytes.
215
217 Some parameters take an option of a given type, such as an integer or a
218 string. Anywhere a numeric value is required, an arithmetic expression
219 may be used, provided it is surrounded by parentheses. Supported opera‐
220 tors are:
221 addition (+)
223 subtraction (-)
225 multiplication (*)
227 division (/)
229 modulus (%)
231 exponentiation (^)
233 For time values in expressions, units are microseconds by default. This
235 is different than for time values not in expressions (not enclosed in
236 parentheses).
237
239 The following parameter types are used.
240 str String. A sequence of alphanumeric characters.
242 time Integer with possible time suffix. Without a unit value is in‐
244 terpreted as seconds unless otherwise specified. Accepts a suf‐
245 fix of 'd' for days, 'h' for hours, 'm' for minutes, 's' for
246 seconds, 'ms' (or 'msec') for milliseconds and 'us' (or 'usec')
247 for microseconds. For example, use 10m for 10 minutes.
248 int Integer. A whole number value, which may contain an integer pre‐
250 fix and an integer suffix.
251 [*integer prefix*] **number** [*integer suffix*]
253 The optional *integer prefix* specifies the number's base. The
255 default is decimal. *0x* specifies hexadecimal.
256 The optional *integer suffix* specifies the number's units, and
258 includes an optional unit prefix and an optional unit. For quan‐
259 tities of data, the default unit is bytes. For quantities of
260 time, the default unit is seconds unless otherwise specified.
261 With `kb_base=1000', fio follows international standards for
263 unit prefixes. To specify power-of-10 decimal values defined in
264 the International System of Units (SI):
265 K means kilo (K) or 1000
267 M means mega (M) or 1000**2
268 G means giga (G) or 1000**3
269 T means tera (T) or 1000**4
270 P means peta (P) or 1000**5
271 To specify power-of-2 binary values defined in IEC 80000-13:
273 Ki means kibi (Ki) or 1024
275 Mi means mebi (Mi) or 1024**2
276 Gi means gibi (Gi) or 1024**3
277 Ti means tebi (Ti) or 1024**4
278 Pi means pebi (Pi) or 1024**5
279 For Zone Block Device Mode:
281 z means Zone
283 With `kb_base=1024' (the default), the unit prefixes are oppo‐
284 site from those specified in the SI and IEC 80000-13 standards
285 to provide compatibility with old scripts. For example, 4k means
286 4096.
287 For quantities of data, an optional unit of 'B' may be included
289 (e.g., 'kB' is the same as 'k').
290 The *integer suffix* is not case sensitive (e.g., m/mi mean
292 mebi/mega, not milli). 'b' and 'B' both mean byte, not bit.
293 Examples with `kb_base=1000':
295 4 KiB: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
297 1 MiB: 1048576, 1m, 1024k
298 1 MB: 1000000, 1mi, 1000ki
299 1 TiB: 1073741824, 1t, 1024m, 1048576k
300 1 TB: 1000000000, 1ti, 1000mi, 1000000ki
301 Examples with `kb_base=1024' (default):
303 4 KiB: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
305 1 MiB: 1048576, 1m, 1024k
306 1 MB: 1000000, 1mi, 1000ki
307 1 TiB: 1073741824, 1t, 1024m, 1048576k
308 1 TB: 1000000000, 1ti, 1000mi, 1000000ki
309 To specify times (units are not case sensitive):
311 D means days
313 H means hours
314 M mean minutes
315 s or sec means seconds (default)
316 ms or msec means milliseconds
317 us or usec means microseconds
318 `z' suffix specifies that the value is measured in zones. Value
320 is recalculated once block device's zone size becomes known.
321 If the option accepts an upper and lower range, use a colon ':'
323 or minus '-' to separate such values. See irange parameter type.
324 If the lower value specified happens to be larger than the upper
325 value the two values are swapped.
326 bool Boolean. Usually parsed as an integer, however only defined for
328 true and false (1 and 0).
329 irange Integer range with suffix. Allows value range to be given, such
331 as 1024-4096. A colon may also be used as the separator, e.g.
332 1k:4k. If the option allows two sets of ranges, they can be
333 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
334 int parameter type.
335 float_list
337 A list of floating point numbers, separated by a ':' character.
338
340 With the above in mind, here follows the complete list of fio job pa‐
341 rameters.
342 Units
344 kb_base=int
345 Select the interpretation of unit prefixes in input parameters.
346 1000 Inputs comply with IEC 80000-13 and the Interna‐
348 tional System of Units (SI). Use:
349 - power-of-2 values with IEC prefixes (e.g., KiB)
351 - power-of-10 values with SI prefixes (e.g., kB)
352 1024 Compatibility mode (default). To avoid breaking
354 old scripts:
355 - power-of-2 values with SI prefixes
357 - power-of-10 values with IEC prefixes
358 See bs for more details on input parameters.
360 Outputs always use correct prefixes. Most outputs include both
362 side-by-side, like:
363 bw=2383.3kB/s (2327.4KiB/s)
365 If only one value is reported, then kb_base selects the one to
367 use:
368 1000 -- SI prefixes
370 1024 -- IEC prefixes
371 unit_base=int
373 Base unit for reporting. Allowed values are:
374 0 Use auto-detection (default).
376 8 Byte based.
378 1 Bit based.
380 Job description
382 name=str
383 ASCII name of the job. This may be used to override the name
384 printed by fio for this job. Otherwise the job name is used. On
385 the command line this parameter has the special purpose of also
386 signaling the start of a new job.
387 description=str
389 Text description of the job. Doesn't do anything except dump
390 this text description when this job is run. It's not parsed.
391 loops=int
393 Run the specified number of iterations of this job. Used to re‐
394 peat the same workload a given number of times. Defaults to 1.
395 numjobs=int
397 Create the specified number of clones of this job. Each clone of
398 job is spawned as an independent thread or process. May be used
399 to setup a larger number of threads/processes doing the same
400 thing. Each thread is reported separately; to see statistics for
401 all clones as a whole, use group_reporting in conjunction with
402 new_group. See --max-jobs. Default: 1.
403 Time related parameters
405 runtime=time
406 Tell fio to terminate processing after the specified period of
407 time. It can be quite hard to determine for how long a specified
408 job will run, so this parameter is handy to cap the total run‐
409 time to a given time. When the unit is omitted, the value is in‐
410 terpreted in seconds.
411 time_based
413 If set, fio will run for the duration of the runtime specified
414 even if the file(s) are completely read or written. It will sim‐
415 ply loop over the same workload as many times as the runtime al‐
416 lows.
417 startdelay=irange(int)
419 Delay the start of job for the specified amount of time. Can be
420 a single value or a range. When given as a range, each thread
421 will choose a value randomly from within the range. Value is in
422 seconds if a unit is omitted.
423 ramp_time=time
425 If set, fio will run the specified workload for this amount of
426 time before logging any performance numbers. Useful for letting
427 performance settle before logging results, thus minimizing the
428 runtime required for stable results. Note that the ramp_time is
429 considered lead in time for a job, thus it will increase the to‐
430 tal runtime if a special timeout or runtime is specified. When
431 the unit is omitted, the value is given in seconds.
432 clocksource=str
434 Use the given clocksource as the base of timing. The supported
435 options are:
436 gettimeofday
438 gettimeofday(2)
439 clock_gettime
441 clock_gettime(2)
442 cpu Internal CPU clock source
444 cpu is the preferred clocksource if it is reliable, as it is
446 very fast (and fio is heavy on time calls). Fio will automati‐
447 cally use this clocksource if it's supported and considered re‐
448 liable on the system it is running on, unless another clock‐
449 source is specifically set. For x86/x86-64 CPUs, this means sup‐
450 porting TSC Invariant.
451 gtod_reduce=bool
453 Enable all of the gettimeofday(2) reducing options (dis‐
454 able_clat, disable_slat, disable_bw_measurement) plus reduce
455 precision of the timeout somewhat to really shrink the gettime‐
456 ofday(2) call count. With this option enabled, we only do about
457 0.4% of the gettimeofday(2) calls we would have done if all time
458 keeping was enabled.
459 gtod_cpu=int
461 Sometimes it's cheaper to dedicate a single thread of execution
462 to just getting the current time. Fio (and databases, for in‐
463 stance) are very intensive on gettimeofday(2) calls. With this
464 option, you can set one CPU aside for doing nothing but logging
465 current time to a shared memory location. Then the other
466 threads/processes that run I/O workloads need only copy that
467 segment, instead of entering the kernel with a gettimeofday(2)
468 call. The CPU set aside for doing these time calls will be ex‐
469 cluded from other uses. Fio will manually clear it from the CPU
470 mask of other jobs.
471 Target file/device
473 directory=str
474 Prefix filenames with this directory. Used to place files in a
475 different location than `./'. You can specify a number of direc‐
476 tories by separating the names with a ':' character. These di‐
477 rectories will be assigned equally distributed to job clones
478 created by numjobs as long as they are using generated file‐
479 names. If specific filename(s) are set fio will use the first
480 listed directory, and thereby matching the filename semantic
481 (which generates a file for each clone if not specified, but
482 lets all clones use the same file if set).
483 See the filename option for information on how to escape ':'
485 characters within the directory path itself.
486 Note: To control the directory fio will use for internal state
488 files use --aux-path.
489 filename=str
491 Fio normally makes up a filename based on the job name, thread
492 number, and file number (see filename_format). If you want to
493 share files between threads in a job or several jobs with fixed
494 file paths, specify a filename for each of them to override the
495 default. If the ioengine is file based, you can specify a number
496 of files by separating the names with a ':' colon. So if you
497 wanted a job to open `/dev/sda' and `/dev/sdb' as the two work‐
498 ing files, you would use `filename=/dev/sda:/dev/sdb'. This also
499 means that whenever this option is specified, nrfiles is ig‐
500 nored. The size of regular files specified by this option will
501 be size divided by number of files unless an explicit size is
502 specified by filesize.
503 Each colon in the wanted path must be escaped with a '\' charac‐
505 ter. For instance, if the path is `/dev/dsk/foo@3,0:c' then you
506 would use `filename=/dev/dsk/foo@3,0\:c' and if the path is
507 `F:\filename' then you would use `filename=F\:\filename'.
508 On Windows, disk devices are accessed as `\\.\PhysicalDrive0'
510 for the first device, `\\.\PhysicalDrive1' for the second etc.
511 Note: Windows and FreeBSD prevent write access to areas of the
512 disk containing in-use data (e.g. filesystems).
513 The filename `-' is a reserved name, meaning *stdin* or *std‐
515 out*. Which of the two depends on the read/write direction set.
516 filename_format=str
518 If sharing multiple files between jobs, it is usually necessary
519 to have fio generate the exact names that you want. By default,
520 fio will name a file based on the default file format specifica‐
521 tion of `jobname.jobnumber.filenumber'. With this option, that
522 can be customized. Fio will recognize and replace the following
523 keywords in this string:
524 $jobname
526 The name of the worker thread or process.
527 $clientuid
529 IP of the fio process when using client/server
530 mode.
531 $jobnum
533 The incremental number of the worker thread or
534 process.
535 $filenum
537 The incremental number of the file for that worker
538 thread or process.
539 To have dependent jobs share a set of files, this option can be
541 set to have fio generate filenames that are shared between the
542 two. For instance, if `testfiles.$filenum' is specified, file
543 number 4 for any job will be named `testfiles.4'. The default of
544 `$jobname.$jobnum.$filenum' will be used if no other format
545 specifier is given.
546 If you specify a path then the directories will be created up to
548 the main directory for the file. So for example if you specify
549 `a/b/c/$jobnum` then the directories a/b/c will be created be‐
550 fore the file setup part of the job. If you specify directory
551 then the path will be relative that directory, otherwise it is
552 treated as the absolute path.
553 unique_filename=bool
555 To avoid collisions between networked clients, fio defaults to
556 prefixing any generated filenames (with a directory specified)
557 with the source of the client connecting. To disable this behav‐
558 ior, set this option to 0.
559 opendir=str
561 Recursively open any files below directory str.
562 lockfile=str
564 Fio defaults to not locking any files before it does I/O to
565 them. If a file or file descriptor is shared, fio can serialize
566 I/O to that file to make the end result consistent. This is
567 usual for emulating real workloads that share files. The lock
568 modes are:
569 none No locking. The default.
571 exclusive
573 Only one thread or process may do I/O at a time,
574 excluding all others.
575 readwrite
577 Read-write locking on the file. Many readers may
578 access the file at the same time, but writes get
579 exclusive access.
580 nrfiles=int
582 Number of files to use for this job. Defaults to 1. The size of
583 files will be size divided by this unless explicit size is spec‐
584 ified by filesize. Files are created for each thread separately,
585 and each file will have a file number within its name by de‐
586 fault, as explained in filename section.
587 openfiles=int
589 Number of files to keep open at the same time. Defaults to the
590 same as nrfiles, can be set smaller to limit the number simulta‐
591 neous opens.
592 file_service_type=str
594 Defines how fio decides which file from a job to service next.
595 The following types are defined:
596 random Choose a file at random.
598 roundrobin
600 Round robin over opened files. This is the de‐
601 fault.
602 sequential
604 Finish one file before moving on to the next. Mul‐
605 tiple files can still be open depending on open‐
606 files.
607 zipf Use a Zipf distribution to decide what file to ac‐
609 cess.
610 pareto Use a Pareto distribution to decide what file to
612 access.
613 normal Use a Gaussian (normal) distribution to decide
615 what file to access.
616 gauss Alias for normal.
618 For random, roundrobin, and sequential, a postfix can be ap‐
620 pended to tell fio how many I/Os to issue before switching to a
621 new file. For example, specifying `file_service_type=random:8'
622 would cause fio to issue 8 I/Os before selecting a new file at
623 random. For the non-uniform distributions, a floating point
624 postfix can be given to influence how the distribution is
625 skewed. See random_distribution for a description of how that
626 would work.
627 ioscheduler=str
629 Attempt to switch the device hosting the file to the specified
630 I/O scheduler before running. If the file is a pipe, a character
631 device file or if device hosting the file could not be deter‐
632 mined, this option is ignored.
633 create_serialize=bool
635 If true, serialize the file creation for the jobs. This may be
636 handy to avoid interleaving of data files, which may greatly de‐
637 pend on the filesystem used and even the number of processors in
638 the system. Default: true.
639 create_fsync=bool
641 fsync(2) the data file after creation. This is the default.
642 create_on_open=bool
644 If true, don't pre-create files but allow the job's open() to
645 create a file when it's time to do I/O. Default: false -- pre-
646 create all necessary files when the job starts.
647 create_only=bool
649 If true, fio will only run the setup phase of the job. If files
650 need to be laid out or updated on disk, only that will be done
651 -- the actual job contents are not executed. Default: false.
652 allow_file_create=bool
654 If true, fio is permitted to create files as part of its work‐
655 load. If this option is false, then fio will error out if the
656 files it needs to use don't already exist. Default: true.
657 allow_mounted_write=bool
659 If this isn't set, fio will abort jobs that are destructive
660 (e.g. that write) to what appears to be a mounted device or par‐
661 tition. This should help catch creating inadvertently destruc‐
662 tive tests, not realizing that the test will destroy data on the
663 mounted file system. Note that some platforms don't allow writ‐
664 ing against a mounted device regardless of this option. Default:
665 false.
666 pre_read=bool
668 If this is given, files will be pre-read into memory before
669 starting the given I/O operation. This will also clear the in‐
670 validate flag, since it is pointless to pre-read and then drop
671 the cache. This will only work for I/O engines that are seek-
672 able, since they allow you to read the same data multiple times.
673 Thus it will not work on non-seekable I/O engines (e.g. network,
674 splice). Default: false.
675 unlink=bool
677 Unlink the job files when done. Not the default, as repeated
678 runs of that job would then waste time recreating the file set
679 again and again. Default: false.
680 unlink_each_loop=bool
682 Unlink job files after each iteration or loop. Default: false.
683 zonemode=str
685 Accepted values are:
686 none The zonerange, zonesize zonecapacity and zoneskip
688 parameters are ignored.
689 strided
691 I/O happens in a single zone until zonesize bytes
692 have been transferred. After that number of bytes
693 has been transferred processing of the next zone
694 starts. The zonecapacity parameter is ignored.
695 zbd Zoned block device mode. I/O happens sequentially
697 in each zone, even if random I/O has been se‐
698 lected. Random I/O happens across all zones in‐
699 stead of being restricted to a single zone. Trim
700 is handled using a zone reset operation. Trim only
701 considers non-empty sequential write required and
702 sequential write preferred zones.
703 zonerange=int
705 For zonemode=strided, this is the size of a single zone. See
706 also zonesize and zoneskip.
707 For zonemode=zbd, this parameter is ignored.
709 zonesize=int
711 For zonemode=strided, this is the number of bytes to transfer
712 before skipping zoneskip bytes. If this parameter is smaller
713 than zonerange then only a fraction of each zone with zonerange
714 bytes will be accessed. If this parameter is larger than zon‐
715 erange then each zone will be accessed multiple times before
716 skipping to the next zone.
717 For zonemode=zbd, this is the size of a single zone. The zon‐
719 erange parameter is ignored in this mode. For a job accessing a
720 zoned block device, the specified zonesize must be 0 or equal to
721 the device zone size. For a regular block device or file, the
722 specified zonesize must be at least 512B.
723 zonecapacity=int
725 For zonemode=zbd, this defines the capacity of a single zone,
726 which is the accessible area starting from the zone start ad‐
727 dress. This parameter only applies when using zonemode=zbd in
728 combination with regular block devices. If not specified it de‐
729 faults to the zone size. If the target device is a zoned block
730 device, the zone capacity is obtained from the device informa‐
731 tion and this option is ignored.
732 zoneskip=int[z]
734 For zonemode=strided, the number of bytes to skip after zonesize
735 bytes of data have been transferred.
736 For zonemode=zbd, the zonesize aligned number of bytes to skip
738 once a zone is fully written (write workloads) or all written
739 data in the zone have been read (read workloads). This parameter
740 is valid only for sequential workloads and ignored for random
741 workloads. For read workloads, see also read_beyond_wp.
742 read_beyond_wp=bool
745 This parameter applies to zonemode=zbd only.
746 Zoned block devices are block devices that consist of multiple
748 zones. Each zone has a type, e.g. conventional or sequential. A
749 conventional zone can be written at any offset that is a multi‐
750 ple of the block size. Sequential zones must be written sequen‐
751 tially. The position at which a write must occur is called the
752 write pointer. A zoned block device can be either host managed
753 or host aware. For host managed devices the host must ensure
754 that writes happen sequentially. Fio recognizes host managed de‐
755 vices and serializes writes to sequential zones for these de‐
756 vices.
757 If a read occurs in a sequential zone beyond the write pointer
759 then the zoned block device will complete the read without read‐
760 ing any data from the storage medium. Since such reads lead to
761 unrealistically high bandwidth and IOPS numbers fio only reads
762 beyond the write pointer if explicitly told to do so. Default:
763 false.
764 max_open_zones=int
766 When running a random write test across an entire drive many
767 more zones will be open than in a typical application workload.
768 Hence this command line option that allows to limit the number
769 of open zones. The number of open zones is defined as the number
770 of zones to which write commands are issued by all threads/pro‐
771 cesses.
772 job_max_open_zones=int
774 Limit on the number of simultaneously opened zones per single
775 thread/process.
776 ignore_zone_limits=bool
778 If this option is used, fio will ignore the maximum number of
779 open zones limit of the zoned block device in use, thus allowing
780 the option max_open_zones value to be larger than the device re‐
781 ported limit. Default: false.
782 zone_reset_threshold=float
784 A number between zero and one that indicates the ratio of logi‐
785 cal blocks with data to the total number of logical blocks in
786 the test above which zones should be reset periodically.
787 zone_reset_frequency=float
789 A number between zero and one that indicates how often a zone
790 reset should be issued if the zone reset threshold has been ex‐
791 ceeded. A zone reset is submitted after each (1 / zone_re‐
792 set_frequency) write requests. This and the previous parameter
793 can be used to simulate garbage collection activity.
794 I/O type
797 direct=bool
798 If value is true, use non-buffered I/O. This is usually O_DI‐
799 RECT. Note that OpenBSD and ZFS on Solaris don't support direct
800 I/O. On Windows the synchronous ioengines don't support direct
801 I/O. Default: false.
802 atomic=bool
804 If value is true, attempt to use atomic direct I/O. Atomic
805 writes are guaranteed to be stable once acknowledged by the op‐
806 erating system. Only Linux supports O_ATOMIC right now.
807 buffered=bool
809 If value is true, use buffered I/O. This is the opposite of the
810 direct option. Defaults to true.
811 readwrite=str, rw=str
813 Type of I/O pattern. Accepted values are:
814 read Sequential reads.
816 write Sequential writes.
818 trim Sequential trims (Linux block devices and SCSI
820 character devices only).
821 randread
823 Random reads.
824 randwrite
826 Random writes.
827 randtrim
829 Random trims (Linux block devices and SCSI charac‐
830 ter devices only).
831 rw,readwrite
833 Sequential mixed reads and writes.
834 randrw Random mixed reads and writes.
836 trimwrite
838 Sequential trim+write sequences. Blocks will be
839 trimmed first, then the same blocks will be writ‐
840 ten to.
841 Fio defaults to read if the option is not specified. For the
843 mixed I/O types, the default is to split them 50/50. For certain
844 types of I/O the result may still be skewed a bit, since the
845 speed may be different.
846 It is possible to specify the number of I/Os to do before get‐
848 ting a new offset by appending `:<nr>' to the end of the string
849 given. For a random read, it would look like `rw=randread:8' for
850 passing in an offset modifier with a value of 8. If the suffix
851 is used with a sequential I/O pattern, then the `<nr>' value
852 specified will be added to the generated offset for each I/O
853 turning sequential I/O into sequential I/O with holes. For in‐
854 stance, using `rw=write:4k' will skip 4k for every write. Also
855 see the rw_sequencer option.
856 rw_sequencer=str
858 If an offset modifier is given by appending a number to the
859 `rw=str' line, then this option controls how that number modi‐
860 fies the I/O offset being generated. Accepted values are:
861 sequential
863 Generate sequential offset.
864 identical
866 Generate the same offset.
867 sequential is only useful for random I/O, where fio would nor‐
869 mally generate a new random offset for every I/O. If you append
870 e.g. 8 to randread, you would get a new random offset for every
871 8 I/Os. The result would be a seek for only every 8 I/Os, in‐
872 stead of for every I/O. Use `rw=randread:8' to specify that. As
873 sequential I/O is already sequential, setting sequential for
874 that would not result in any differences. identical behaves in a
875 similar fashion, except it sends the same offset 8 number of
876 times before generating a new offset.
877 unified_rw_reporting=str
879 Fio normally reports statistics on a per data direction basis,
880 meaning that reads, writes, and trims are accounted and reported
881 separately. This option determines whether fio reports the re‐
882 sults normally, summed together, or as both options. Accepted
883 values are:
884 none Normal statistics reporting.
886 mixed Statistics are summed per data direction and reported to‐
888 gether.
889 both Statistics are reported normally, followed by the mixed
891 statistics.
892 0 Backward-compatible alias for none.
894 1 Backward-compatible alias for mixed.
896 2 Alias for both.
898 randrepeat=bool
900 Seed the random number generator used for random I/O patterns in
901 a predictable way so the pattern is repeatable across runs. De‐
902 fault: true.
903 allrandrepeat=bool
905 Seed all random number generators in a predictable way so re‐
906 sults are repeatable across runs. Default: false.
907 randseed=int
909 Seed the random number generators based on this seed value, to
910 be able to control what sequence of output is being generated.
911 If not set, the random sequence depends on the randrepeat set‐
912 ting.
913 fallocate=str
915 Whether pre-allocation is performed when laying down files. Ac‐
916 cepted values are:
917 none Do not pre-allocate space.
919 native Use a platform's native pre-allocation call but
921 fall back to none behavior if it fails/is not im‐
922 plemented.
923 posix Pre-allocate via posix_fallocate(3).
925 keep Pre-allocate via fallocate(2) with FAL‐
927 LOC_FL_KEEP_SIZE set.
928 truncate
930 Extend file to final size using ftruncate|(2) in‐
931 stead of allocating.
932 0 Backward-compatible alias for none.
934 1 Backward-compatible alias for posix.
936 May not be available on all supported platforms. keep is only
938 available on Linux. If using ZFS on Solaris this cannot be set
939 to posix because ZFS doesn't support pre-allocation. Default:
940 native if any pre-allocation methods except truncate are avail‐
941 able, none if not.
942 Note that using truncate on Windows will interact surprisingly
944 with non-sequential write patterns. When writing to a file that
945 has been extended by setting the end-of-file information, Win‐
946 dows will backfill the unwritten portion of the file up to that
947 offset with zeroes before issuing the new write. This means that
948 a single small write to the end of an extended file will stall
949 until the entire file has been filled with zeroes.
950 fadvise_hint=str
952 Use posix_fadvise(2) or posix_madvise(2) to advise the kernel
953 what I/O patterns are likely to be issued. Accepted values are:
954 0 Backwards compatible hint for "no hint".
956 1 Backwards compatible hint for "advise with fio
958 workload type". This uses FADV_RANDOM for a random
959 workload, and FADV_SEQUENTIAL for a sequential
960 workload.
961 sequential
963 Advise using FADV_SEQUENTIAL.
964 random Advise using FADV_RANDOM.
966 write_hint=str
968 Use fcntl(2) to advise the kernel what life time to expect from
969 a write. Only supported on Linux, as of version 4.13. Accepted
970 values are:
971 none No particular life time associated with this file.
973 short Data written to this file has a short life time.
975 medium Data written to this file has a medium life time.
977 long Data written to this file has a long life time.
979 extreme
981 Data written to this file has a very long life
982 time.
983 The values are all relative to each other, and no absolute mean‐
985 ing should be associated with them.
986 offset=int[%|z]
988 Start I/O at the provided offset in the file, given as either a
989 fixed size in bytes, zones or a percentage. If a percentage is
990 given, the generated offset will be aligned to the minimum
991 blocksize or to the value of offset_align if provided. Data be‐
992 fore the given offset will not be touched. This effectively caps
993 the file size at `real_size - offset'. Can be combined with size
994 to constrain the start and end range of the I/O workload. A
995 percentage can be specified by a number between 1 and 100 fol‐
996 lowed by '%', for example, `offset=20%' to specify 20%. In ZBD
997 mode, value can be set as number of zones using 'z'.
998 offset_align=int
1000 If set to non-zero value, the byte offset generated by a per‐
1001 centage offset is aligned upwards to this value. Defaults to 0
1002 meaning that a percentage offset is aligned to the minimum block
1003 size.
1004 offset_increment=int[%|z]
1006 If this is provided, then the real offset becomes `offset + off‐
1007 set_increment * thread_number', where the thread number is a
1008 counter that starts at 0 and is incremented for each sub-job
1009 (i.e. when numjobs option is specified). This option is useful
1010 if there are several jobs which are intended to operate on a
1011 file in parallel disjoint segments, with even spacing between
1012 the starting points. Percentages can be used for this option.
1013 If a percentage is given, the generated offset will be aligned
1014 to the minimum blocksize or to the value of offset_align if pro‐
1015 vided.In ZBD mode, value can be set as number of zones using
1016 'z'.
1017 number_ios=int
1019 Fio will normally perform I/Os until it has exhausted the size
1020 of the region set by size, or if it exhaust the allocated time
1021 (or hits an error condition). With this setting, the range/size
1022 can be set independently of the number of I/Os to perform. When
1023 fio reaches this number, it will exit normally and report sta‐
1024 tus. Note that this does not extend the amount of I/O that will
1025 be done, it will only stop fio if this condition is met before
1026 other end-of-job criteria.
1027 fsync=int
1029 If writing to a file, issue an fsync(2) (or its equivalent) of
1030 the dirty data for every number of blocks given. For example, if
1031 you give 32 as a parameter, fio will sync the file after every
1032 32 writes issued. If fio is using non-buffered I/O, we may not
1033 sync the file. The exception is the sg I/O engine, which syn‐
1034 chronizes the disk cache anyway. Defaults to 0, which means fio
1035 does not periodically issue and wait for a sync to complete.
1036 Also see end_fsync and fsync_on_close.
1037 fdatasync=int
1039 Like fsync but uses fdatasync(2) to only sync data and not meta‐
1040 data blocks. In Windows, DragonFlyBSD or OSX there is no fdata‐
1041 sync(2) so this falls back to using fsync(2). Defaults to 0,
1042 which means fio does not periodically issue and wait for a data-
1043 only sync to complete.
1044 write_barrier=int
1046 Make every N-th write a barrier write.
1047 sync_file_range=str:int
1049 Use sync_file_range(2) for every int number of write operations.
1050 Fio will track range of writes that have happened since the last
1051 sync_file_range(2) call. str can currently be one or more of:
1052 wait_before
1054 SYNC_FILE_RANGE_WAIT_BEFORE
1055 write SYNC_FILE_RANGE_WRITE
1057 wait_after
1059 SYNC_FILE_RANGE_WRITE_AFTER
1060 So if you do `sync_file_range=wait_before,write:8', fio would
1062 use `SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE' for
1063 every 8 writes. Also see the sync_file_range(2) man page. This
1064 option is Linux specific.
1065 overwrite=bool
1067 If true, writes to a file will always overwrite existing data.
1068 If the file doesn't already exist, it will be created before the
1069 write phase begins. If the file exists and is large enough for
1070 the specified write phase, nothing will be done. Default: false.
1071 end_fsync=bool
1073 If true, fsync(2) file contents when a write stage has com‐
1074 pleted. Default: false.
1075 fsync_on_close=bool
1077 If true, fio will fsync(2) a dirty file on close. This differs
1078 from end_fsync in that it will happen on every file close, not
1079 just at the end of the job. Default: false.
1080 rwmixread=int
1082 Percentage of a mixed workload that should be reads. Default:
1083 50.
1084 rwmixwrite=int
1086 Percentage of a mixed workload that should be writes. If both
1087 rwmixread and rwmixwrite is given and the values do not add up
1088 to 100%, the latter of the two will be used to override the
1089 first. This may interfere with a given rate setting, if fio is
1090 asked to limit reads or writes to a certain rate. If that is the
1091 case, then the distribution may be skewed. Default: 50.
1092 random_distribution=str:float[:float][,str:float][,str:float]
1094 By default, fio will use a completely uniform random distribu‐
1095 tion when asked to perform random I/O. Sometimes it is useful to
1096 skew the distribution in specific ways, ensuring that some parts
1097 of the data is more hot than others. fio includes the following
1098 distribution models:
1099 random Uniform random distribution
1101 zipf Zipf distribution
1103 pareto Pareto distribution
1105 normal Normal (Gaussian) distribution
1107 zoned Zoned random distribution zoned_abs Zoned absolute
1109 random distribution
1110 When using a zipf or pareto distribution, an input value is also
1112 needed to define the access pattern. For zipf, this is the `Zipf
1113 theta'. For pareto, it's the `Pareto power'. Fio includes a
1114 test program, fio-genzipf, that can be used visualize what the
1115 given input values will yield in terms of hit rates. If you
1116 wanted to use zipf with a `theta' of 1.2, you would use `ran‐
1117 dom_distribution=zipf:1.2' as the option. If a non-uniform model
1118 is used, fio will disable use of the random map. For the normal
1119 distribution, a normal (Gaussian) deviation is supplied as a
1120 value between 0 and 100.
1121 The second, optional float is allowed for pareto, zipf and nor‐
1123 mal distributions. It allows to set base of distribution in non-
1124 default place, giving more control over most probable outcome.
1125 This value is in range [0-1] which maps linearly to range of
1126 possible random values. Defaults are: random for pareto and
1127 zipf, and 0.5 for normal. If you wanted to use zipf with a
1128 `theta` of 1.2 centered on 1/4 of allowed value range, you would
1129 use `random_distribution=zipf:1.2:0.25`.
1130 For a zoned distribution, fio supports specifying percentages of
1132 I/O access that should fall within what range of the file or de‐
1133 vice. For example, given a criteria of:
1134 60% of accesses should be to the first 10%
1136 30% of accesses should be to the next 20%
1137 8% of accesses should be to the next 30%
1138 2% of accesses should be to the next 40%
1139 we can define that through zoning of the random accesses. For
1141 the above example, the user would do:
1142 random_distribution=zoned:60/10:30/20:8/30:2/40
1144 A zoned_abs distribution works exactly like thezoned, except
1146 that it takes absolute sizes. For example, let's say you wanted
1147 to define access according to the following criteria:
1148 60% of accesses should be to the first 20G
1150 30% of accesses should be to the next 100G
1151 10% of accesses should be to the next 500G
1152 we can define an absolute zoning distribution with:
1154 random_distribution=zoned:60/10:30/20:8/30:2/40
1156 For both zoned and zoned_abs, fio supports defining up to 256
1158 separate zones.
1159 Similarly to how bssplit works for setting ranges and percent‐
1161 ages of block sizes. Like bssplit, it's possible to specify sep‐
1162 arate zones for reads, writes, and trims. If just one set is
1163 given, it'll apply to all of them.
1164 percentage_random=int[,int][,int]
1166 For a random workload, set how big a percentage should be ran‐
1167 dom. This defaults to 100%, in which case the workload is fully
1168 random. It can be set from anywhere from 0 to 100. Setting it to
1169 0 would make the workload fully sequential. Any setting in be‐
1170 tween will result in a random mix of sequential and random I/O,
1171 at the given percentages. Comma-separated values may be speci‐
1172 fied for reads, writes, and trims as described in blocksize.
1173 norandommap
1175 Normally fio will cover every block of the file when doing ran‐
1176 dom I/O. If this option is given, fio will just get a new random
1177 offset without looking at past I/O history. This means that some
1178 blocks may not be read or written, and that some blocks may be
1179 read/written more than once. If this option is used with verify
1180 and multiple blocksizes (via bsrange), only intact blocks are
1181 verified, i.e., partially-overwritten blocks are ignored. With
1182 an async I/O engine and an I/O depth > 1, it is possible for the
1183 same block to be overwritten, which can cause verification er‐
1184 rors. Either do not use norandommap in this case, or also use
1185 the lfsr random generator.
1186 softrandommap=bool
1188 See norandommap. If fio runs with the random block map enabled
1189 and it fails to allocate the map, if this option is set it will
1190 continue without a random block map. As coverage will not be as
1191 complete as with random maps, this option is disabled by de‐
1192 fault.
1193 random_generator=str
1195 Fio supports the following engines for generating I/O offsets
1196 for random I/O:
1197 tausworthe
1199 Strong 2^88 cycle random number generator.
1200 lfsr Linear feedback shift register generator.
1202 tausworthe64
1204 Strong 64-bit 2^258 cycle random number generator.
1205 tausworthe is a strong random number generator, but it requires
1207 tracking on the side if we want to ensure that blocks are only
1208 read or written once. lfsr guarantees that we never generate the
1209 same offset twice, and it's also less computationally expensive.
1210 It's not a true random generator, however, though for I/O pur‐
1211 poses it's typically good enough. lfsr only works with single
1212 block sizes, not with workloads that use multiple block sizes.
1213 If used with such a workload, fio may read or write some blocks
1214 multiple times. The default value is tausworthe, unless the re‐
1215 quired space exceeds 2^32 blocks. If it does, then tausworthe64
1216 is selected automatically.
1217 Block size
1219 blocksize=int[,int][,int], bs=int[,int][,int]
1220 The block size in bytes used for I/O units. Default: 4096. A
1221 single value applies to reads, writes, and trims. Comma-sepa‐
1222 rated values may be specified for reads, writes, and trims. A
1223 value not terminated in a comma applies to subsequent types. Ex‐
1224 amples:
1225 bs=256k means 256k for reads, writes and trims.
1227 bs=8k,32k means 8k for reads, 32k for writes and
1228 trims.
1229 bs=8k,32k, means 8k for reads, 32k for writes, and
1230 default for trims.
1231 bs=,8k means default for reads, 8k for writes and
1232 trims.
1233 bs=,8k, means default for reads, 8k for writes,
1234 and default for trims.
1235 blocksize_range=irange[,irange][,irange],
1237 bsrange=irange[,irange][,irange]
1238 A range of block sizes in bytes for I/O units. The issued I/O
1239 unit will always be a multiple of the minimum size, unless
1240 blocksize_unaligned is set. Comma-separated ranges may be spec‐
1241 ified for reads, writes, and trims as described in blocksize.
1242 Example:
1243 bsrange=1k-4k,2k-8k
1245 bssplit=str[,str][,str]
1247 Sometimes you want even finer grained control of the block sizes
1248 issued, not just an even split between them. This option allows
1249 you to weight various block sizes, so that you are able to de‐
1250 fine a specific amount of block sizes issued. The format for
1251 this option is:
1252 bssplit=blocksize/percentage:blocksize/percentage
1254 for as many block sizes as needed. So if you want to define a
1256 workload that has 50% 64k blocks, 10% 4k blocks, and 40% 32k
1257 blocks, you would write:
1258 bssplit=4k/10:64k/50:32k/40
1260 Ordering does not matter. If the percentage is left blank, fio
1262 will fill in the remaining values evenly. So a bssplit option
1263 like this one:
1264 bssplit=4k/50:1k/:32k/
1266 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages
1268 always add up to 100, if bssplit is given a range that adds up
1269 to more, it will error out.
1270 Comma-separated values may be specified for reads, writes, and
1272 trims as described in blocksize.
1273 If you want a workload that has 50% 2k reads and 50% 4k reads,
1275 while having 90% 4k writes and 10% 8k writes, you would specify:
1276 bssplit=2k/50:4k/50,4k/90:8k/10
1278 Fio supports defining up to 64 different weights for each data
1280 direction.
1281 blocksize_unaligned, bs_unaligned
1283 If set, fio will issue I/O units with any size within block‐
1284 size_range, not just multiples of the minimum size. This typi‐
1285 cally won't work with direct I/O, as that normally requires sec‐
1286 tor alignment.
1287 bs_is_seq_rand=bool
1289 If this option is set, fio will use the normal read,write block‐
1290 size settings as sequential,random blocksize settings instead.
1291 Any random read or write will use the WRITE blocksize settings,
1292 and any sequential read or write will use the READ blocksize
1293 settings.
1294 blockalign=int[,int][,int], ba=int[,int][,int]
1296 Boundary to which fio will align random I/O units. Default:
1297 blocksize. Minimum alignment is typically 512b for using direct
1298 I/O, though it usually depends on the hardware block size. This
1299 option is mutually exclusive with using a random map for files,
1300 so it will turn off that option. Comma-separated values may be
1301 specified for reads, writes, and trims as described in block‐
1302 size.
1303 Buffers and memory
1305 zero_buffers
1306 Initialize buffers with all zeros. Default: fill buffers with
1307 random data.
1308 refill_buffers
1310 If this option is given, fio will refill the I/O buffers on ev‐
1311 ery submit. The default is to only fill it at init time and re‐
1312 use that data. Only makes sense if zero_buffers isn't specified,
1313 naturally. If data verification is enabled, refill_buffers is
1314 also automatically enabled.
1315 scramble_buffers=bool
1317 If refill_buffers is too costly and the target is using data
1318 deduplication, then setting this option will slightly modify the
1319 I/O buffer contents to defeat normal de-dupe attempts. This is
1320 not enough to defeat more clever block compression attempts, but
1321 it will stop naive dedupe of blocks. Default: true.
1322 buffer_compress_percentage=int
1324 If this is set, then fio will attempt to provide I/O buffer con‐
1325 tent (on WRITEs) that compresses to the specified level. Fio
1326 does this by providing a mix of random data followed by fixed
1327 pattern data. The fixed pattern is either zeros, or the pattern
1328 specified by buffer_pattern. If the buffer_pattern option is
1329 used, it might skew the compression ratio slightly. Setting buf‐
1330 fer_compress_percentage to a value other than 100 will also en‐
1331 able refill_buffers in order to reduce the likelihood that adja‐
1332 cent blocks are so similar that they over compress when seen to‐
1333 gether. See buffer_compress_chunk for how to set a finer or
1334 coarser granularity of the random/fixed data regions. Defaults
1335 to unset i.e., buffer data will not adhere to any compression
1336 level.
1337 buffer_compress_chunk=int
1339 This setting allows fio to manage how big the random/fixed data
1340 region is when using buffer_compress_percentage. When buf‐
1341 fer_compress_chunk is set to some non-zero value smaller than
1342 the block size, fio can repeat the random/fixed region through‐
1343 out the I/O buffer at the specified interval (which particularly
1344 useful when bigger block sizes are used for a job). When set to
1345 0, fio will use a chunk size that matches the block size result‐
1346 ing in a single random/fixed region within the I/O buffer. De‐
1347 faults to 512. When the unit is omitted, the value is inter‐
1348 preted in bytes.
1349 buffer_pattern=str
1351 If set, fio will fill the I/O buffers with this pattern or with
1352 the contents of a file. If not set, the contents of I/O buffers
1353 are defined by the other options related to buffer contents. The
1354 setting can be any pattern of bytes, and can be prefixed with 0x
1355 for hex values. It may also be a string, where the string must
1356 then be wrapped with "". Or it may also be a filename, where the
1357 filename must be wrapped with '' in which case the file is
1358 opened and read. Note that not all the file contents will be
1359 read if that would cause the buffers to overflow. So, for exam‐
1360 ple:
1361 buffer_pattern='filename'
1363 or:
1364 buffer_pattern="abcd"
1365 or:
1366 buffer_pattern=-12
1367 or:
1368 buffer_pattern=0xdeadface
1369 Also you can combine everything together in any order:
1371 buffer_pattern=0xdeadface"abcd"-12'filename'
1373 dedupe_percentage=int
1375 If set, fio will generate this percentage of identical buffers
1376 when writing. These buffers will be naturally dedupable. The
1377 contents of the buffers depend on what other buffer compression
1378 settings have been set. It's possible to have the individual
1379 buffers either fully compressible, or not at all -- this option
1380 only controls the distribution of unique buffers. Setting this
1381 option will also enable refill_buffers to prevent every buffer
1382 being identical.
1383 dedupe_mode=str
1385 If dedupe_percentage is given, then this option controls how fio
1386 generates the dedupe buffers.
1387 repeat
1389 Generate dedupe buffers by repeating previous
1391 writes
1392 working_set
1394 Generate dedupe buffers from working set
1396 repeat is the default option for fio. Dedupe buffers are gener‐
1398 ated by repeating previous unique write.
1399 working_set is a more realistic workload. With working_set,
1401 dedupe_working_set_percentage should be provided. Given that,
1402 fio will use the initial unique write buffers as its working
1403 set. Upon deciding to dedupe, fio will randomly choose a buffer
1404 from the working set. Note that by using working_set the dedupe
1405 percentage will converge to the desired over time while repeat
1406 maintains the desired percentage throughout the job.
1407 dedupe_working_set_percentage=int
1409 If dedupe_mode is set to working_set, then this controls the
1410 percentage of size of the file or device used as the buffers fio
1411 will choose to generate the dedupe buffers from
1412 Note that size needs to be explicitly provided and only 1 file
1414 per job is supported
1415 invalidate=bool
1417 Invalidate the buffer/page cache parts of the files to be used
1418 prior to starting I/O if the platform and file type support it.
1419 Defaults to true. This will be ignored if pre_read is also
1420 specified for the same job.
1421 sync=str
1423 Whether, and what type, of synchronous I/O to use for writes.
1424 The allowed values are:
1425 none Do not use synchronous IO, the default.
1427 0 Same as none.
1429 sync Use synchronous file IO. For the majority of I/O
1431 engines, this means using O_SYNC.
1432 1 Same as sync.
1434 dsync Use synchronous data IO. For the majority of I/O
1436 engines, this means using O_DSYNC.
1437 iomem=str, mem=str
1439 Fio can use various types of memory as the I/O unit buffer. The
1440 allowed values are:
1441 malloc Use memory from malloc(3) as the buffers. Default
1443 memory type.
1444 shm Use shared memory as the buffers. Allocated
1446 through shmget(2).
1447 shmhuge
1449 Same as shm, but use huge pages as backing.
1450 mmap Use mmap(2) to allocate buffers. May either be
1452 anonymous memory, or can be file backed if a file‐
1453 name is given after the option. The format is
1454 `mem=mmap:/path/to/file'.
1455 mmaphuge
1457 Use a memory mapped huge file as the buffer back‐
1458 ing. Append filename after mmaphuge, ala `mem=mma‐
1459 phuge:/hugetlbfs/file'.
1460 mmapshared
1462 Same as mmap, but use a MMAP_SHARED mapping.
1463 cudamalloc
1465 Use GPU memory as the buffers for GPUDirect RDMA
1466 benchmark. The ioengine must be rdma.
1467 The area allocated is a function of the maximum allowed bs size
1469 for the job, multiplied by the I/O depth given. Note that for
1470 shmhuge and mmaphuge to work, the system must have free huge
1471 pages allocated. This can normally be checked and set by read‐
1472 ing/writing `/proc/sys/vm/nr_hugepages' on a Linux system. Fio
1473 assumes a huge page is 4MiB in size. So to calculate the number
1474 of huge pages you need for a given job file, add up the I/O
1475 depth of all jobs (normally one unless iodepth is used) and mul‐
1476 tiply by the maximum bs set. Then divide that number by the huge
1477 page size. You can see the size of the huge pages in `/proc/mem‐
1478 info'. If no huge pages are allocated by having a non-zero num‐
1479 ber in `nr_hugepages', using mmaphuge or shmhuge will fail. Also
1480 see hugepage-size.
1481 mmaphuge also needs to have hugetlbfs mounted and the file loca‐
1483 tion should point there. So if it's mounted in `/huge', you
1484 would use `mem=mmaphuge:/huge/somefile'.
1485 iomem_align=int, mem_align=int
1487 This indicates the memory alignment of the I/O memory buffers.
1488 Note that the given alignment is applied to the first I/O unit
1489 buffer, if using iodepth the alignment of the following buffers
1490 are given by the bs used. In other words, if using a bs that is
1491 a multiple of the page sized in the system, all buffers will be
1492 aligned to this value. If using a bs that is not page aligned,
1493 the alignment of subsequent I/O memory buffers is the sum of the
1494 iomem_align and bs used.
1495 hugepage-size=int
1497 Defines the size of a huge page. Must at least be equal to the
1498 system setting, see `/proc/meminfo'. Defaults to 4MiB. Should
1499 probably always be a multiple of megabytes, so using
1500 `hugepage-size=Xm' is the preferred way to set this to avoid
1501 setting a non-pow-2 bad value.
1502 lockmem=int
1504 Pin the specified amount of memory with mlock(2). Can be used to
1505 simulate a smaller amount of memory. The amount specified is per
1506 worker.
1507 I/O size
1509 size=int[%|z]
1510 The total size of file I/O for each thread of this job. Fio will
1511 run until this many bytes has been transferred, unless runtime
1512 is limited by other options (such as runtime, for instance, or
1513 increased/decreased by io_size). Fio will divide this size be‐
1514 tween the available files determined by options such as nrfiles,
1515 filename, unless filesize is specified by the job. If the result
1516 of division happens to be 0, the size is set to the physical
1517 size of the given files or devices if they exist. If this op‐
1518 tion is not specified, fio will use the full size of the given
1519 files or devices. If the files do not exist, size must be given.
1520 It is also possible to give size as a percentage between 1 and
1521 100. If `size=20%' is given, fio will use 20% of the full size
1522 of the given files or devices. In ZBD mode, size can be given in
1523 units of number of zones using 'z'. Can be combined with offset
1524 to constrain the start and end range that I/O will be done
1525 within.
1526 io_size=int[%|z], io_limit=int[%|z]
1528 Normally fio operates within the region set by size, which means
1529 that the size option sets both the region and size of I/O to be
1530 performed. Sometimes that is not what you want. With this op‐
1531 tion, it is possible to define just the amount of I/O that fio
1532 should do. For instance, if size is set to 20GiB and io_size is
1533 set to 5GiB, fio will perform I/O within the first 20GiB but
1534 exit when 5GiB have been done. The opposite is also possible --
1535 if size is set to 20GiB, and io_size is set to 40GiB, then fio
1536 will do 40GiB of I/O within the 0..20GiB region. Value can be
1537 set as percentage: io_size=N%. In this case io_size multiplies
1538 size= value. In ZBD mode, value can also be set as number of
1539 zones using 'z'.
1540 filesize=irange(int)
1542 Individual file sizes. May be a range, in which case fio will
1543 select sizes for files at random within the given range. If not
1544 given, each created file is the same size. This option overrides
1545 size in terms of file size, i.e. size becomes merely the default
1546 for io_size (and has no effect it all if io_size is set explic‐
1547 itly).
1548 file_append=bool
1550 Perform I/O after the end of the file. Normally fio will operate
1551 within the size of a file. If this option is set, then fio will
1552 append to the file instead. This has identical behavior to set‐
1553 ting offset to the size of a file. This option is ignored on
1554 non-regular files.
1555 fill_device=bool, fill_fs=bool
1557 Sets size to something really large and waits for ENOSPC (no
1558 space left on device) or EDQUOT (disk quota exceeded) as the
1559 terminating condition. Only makes sense with sequential write.
1560 For a read workload, the mount point will be filled first then
1561 I/O started on the result.
1562 I/O engine
1564 ioengine=str
1565 Defines how the job issues I/O to the file. The following types
1566 are defined:
1567 sync Basic read(2) or write(2) I/O. lseek(2) is used to
1569 position the I/O location. See fsync and fdata‐
1570 sync for syncing write I/Os.
1571 psync Basic pread(2) or pwrite(2) I/O. Default on all
1573 supported operating systems except for Windows.
1574 vsync Basic readv(2) or writev(2) I/O. Will emulate
1576 queuing by coalescing adjacent I/Os into a single
1577 submission.
1578 pvsync Basic preadv(2) or pwritev(2) I/O.
1580 pvsync2
1582 Basic preadv2(2) or pwritev2(2) I/O.
1583 libaio Linux native asynchronous I/O. Note that Linux may
1585 only support queued behavior with non-buffered I/O
1586 (set `direct=1' or `buffered=0'). This engine de‐
1587 fines engine specific options.
1588 posixaio
1590 POSIX asynchronous I/O using aio_read(3) and
1591 aio_write(3).
1592 solarisaio
1594 Solaris native asynchronous I/O.
1595 windowsaio
1597 Windows native asynchronous I/O. Default on Win‐
1598 dows.
1599 mmap File is memory mapped with mmap(2) and data copied
1601 to/from using memcpy(3).
1602 splice splice(2) is used to transfer the data and vm‐
1604 splice(2) to transfer data from user space to the
1605 kernel.
1606 sg SCSI generic sg v3 I/O. May either be synchronous
1608 using the SG_IO ioctl, or if the target is an sg
1609 character device we use read(2) and write(2) for
1610 asynchronous I/O. Requires filename option to
1611 specify either block or character devices. This
1612 engine supports trim operations. The sg engine in‐
1613 cludes engine specific options.
1614 libzbc Read, write, trim and ZBC/ZAC operations to a
1616 zoned block device using libzbc library. The tar‐
1617 get can be either an SG character device or a
1618 block device file.
1619 null Doesn't transfer any data, just pretends to. This
1621 is mainly used to exercise fio itself and for de‐
1622 bugging/testing purposes.
1623 net Transfer over the network to given `host:port'.
1625 Depending on the protocol used, the hostname,
1626 port, listen and filename options are used to
1627 specify what sort of connection to make, while the
1628 protocol option determines which protocol will be
1629 used. This engine defines engine specific options.
1630 netsplice
1632 Like net, but uses splice(2) and vmsplice(2) to
1633 map data and send/receive. This engine defines
1634 engine specific options.
1635 cpuio Doesn't transfer any data, but burns CPU cycles
1637 according to the cpuload, cpuchunks and cpumode
1638 options. A job never finishes unless there is at
1639 least one non-cpuio job.
1640 cpuload=85 will cause that job to do nothing but
1642 burn 85% of the CPU. In case of SMP machines, use
1643 numjobs=<nr_of_cpu> to get desired CPU usage, as
1644 the cpuload only loads a single CPU at the desired
1645 rate.
1646 cpumode=qsort replace the default noop instruc‐
1648 tions loop by a qsort algorithm to consume more
1649 energy.
1650 rdma The RDMA I/O engine supports both RDMA memory se‐
1652 mantics (RDMA_WRITE/RDMA_READ) and channel seman‐
1653 tics (Send/Recv) for the InfiniBand, RoCE and
1654 iWARP protocols. This engine defines engine spe‐
1655 cific options.
1656 falloc I/O engine that does regular fallocate to simulate
1657 data transfer as fio ioengine.
1658 DDIR_READ does fallocate(,mode = FAL‐
1659 LOC_FL_KEEP_SIZE,).
1660 DIR_WRITE does fallocate(,mode = 0).
1661 DDIR_TRIM does fallocate(,mode = FAL‐
1662 LOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE).
1663 ftruncate
1665 I/O engine that sends ftruncate(2) operations in
1666 response to write (DDIR_WRITE) events. Each ftrun‐
1667 cate issued sets the file's size to the current
1668 block offset. blocksize is ignored.
1669 e4defrag
1671 I/O engine that does regular EXT4_IOC_MOVE_EXT
1672 ioctls to simulate defragment activity in request
1673 to DDIR_WRITE event.
1674 rados I/O engine supporting direct access to Ceph Reli‐
1676 able Autonomic Distributed Object Store (RADOS)
1677 via librados. This ioengine defines engine spe‐
1678 cific options.
1679 rbd I/O engine supporting direct access to Ceph Rados
1681 Block Devices (RBD) via librbd without the need to
1682 use the kernel rbd driver. This ioengine defines
1683 engine specific options.
1684 http I/O engine supporting GET/PUT requests over
1686 HTTP(S) with libcurl to a WebDAV or S3 endpoint.
1687 This ioengine defines engine specific options.
1688 This engine only supports direct IO of iodepth=1;
1690 you need to scale this via numjobs. blocksize de‐
1691 fines the size of the objects to be created.
1692 TRIM is translated to object deletion.
1694 gfapi Using GlusterFS libgfapi sync interface to direct
1696 access to GlusterFS volumes without having to go
1697 through FUSE. This ioengine defines engine spe‐
1698 cific options.
1699 gfapi_async
1701 Using GlusterFS libgfapi async interface to direct
1702 access to GlusterFS volumes without having to go
1703 through FUSE. This ioengine defines engine spe‐
1704 cific options.
1705 libhdfs
1707 Read and write through Hadoop (HDFS). The filename
1708 option is used to specify host,port of the hdfs
1709 name-node to connect. This engine interprets off‐
1710 sets a little differently. In HDFS, files once
1711 created cannot be modified so random writes are
1712 not possible. To imitate this the libhdfs engine
1713 expects a bunch of small files to be created over
1714 HDFS and will randomly pick a file from them based
1715 on the offset generated by fio backend (see the
1716 example job file to create such files, use
1717 `rw=write' option). Please note, it may be neces‐
1718 sary to set environment variables to work with
1719 HDFS/libhdfs properly. Each job uses its own con‐
1720 nection to HDFS.
1721 mtd Read, write and erase an MTD character device
1723 (e.g., `/dev/mtd0'). Discards are treated as
1724 erases. Depending on the underlying device type,
1725 the I/O may have to go in a certain pattern, e.g.,
1726 on NAND, writing sequentially to erase blocks and
1727 discarding before overwriting. The trimwrite mode
1728 works well for this constraint.
1729 pmemblk
1731 Read and write using filesystem DAX to a file on a
1732 filesystem mounted with DAX on a persistent memory
1733 device through the PMDK libpmemblk library.
1734 dev-dax
1736 Read and write using device DAX to a persistent
1737 memory device (e.g., /dev/dax0.0) through the PMDK
1738 libpmem library.
1739 external
1741 Prefix to specify loading an external I/O engine
1742 object file. Append the engine filename, e.g. `io‐
1743 engine=external:/tmp/foo.o' to load ioengine
1744 `foo.o' in `/tmp'. The path can be either absolute
1745 or relative. See `engines/skeleton_external.c' in
1746 the fio source for details of writing an external
1747 I/O engine.
1748 filecreate
1750 Simply create the files and do no I/O to them.
1751 You still need to set filesize so that all the ac‐
1752 counting still occurs, but no actual I/O will be
1753 done other than creating the file.
1754 filestat
1756 Simply do stat() and do no I/O to the file. You
1757 need to set 'filesize' and 'nrfiles', so that
1758 files will be created. This engine is to measure
1759 file lookup and meta data access.
1760 filedelete
1762 Simply delete files by unlink() and do no I/O to
1763 the file. You need to set 'filesize' and 'nr‐
1764 files', so that files will be created. This en‐
1765 gine is to measure file delete.
1766 libpmem
1768 Read and write using mmap I/O to a file on a
1769 filesystem mounted with DAX on a persistent memory
1770 device through the PMDK libpmem library.
1771 ime_psync
1773 Synchronous read and write using DDN's Infinite
1774 Memory Engine (IME). This engine is very basic and
1775 issues calls to IME whenever an IO is queued.
1776 ime_psyncv
1778 Synchronous read and write using DDN's Infinite
1779 Memory Engine (IME). This engine uses iovecs and
1780 will try to stack as much IOs as possible (if the
1781 IOs are "contiguous" and the IO depth is not ex‐
1782 ceeded) before issuing a call to IME.
1783 ime_aio
1785 Asynchronous read and write using DDN's Infinite
1786 Memory Engine (IME). This engine will try to stack
1787 as much IOs as possible by creating requests for
1788 IME. FIO will then decide when to commit these
1789 requests.
1790 libiscsi
1792 Read and write iscsi lun with libiscsi.
1793 nbd Synchronous read and write a Network Block Device
1795 (NBD).
1796 libcufile
1798 I/O engine supporting libcufile synchronous access
1799 to nvidia-fs and a GPUDirect Storage-supported
1800 filesystem. This engine performs I/O without
1801 transferring buffers between user-space and the
1802 kernel, unless verify is set or cuda_io is posix.
1803 iomem must not be cudamalloc. This ioengine de‐
1804 fines engine specific options.
1805 dfs I/O engine supporting asynchronous read and write
1807 operations to the DAOS File System (DFS) via
1808 libdfs.
1809 nfs I/O engine supporting asynchronous read and write
1811 operations to NFS filesystems from userspace via
1812 libnfs. This is useful for achieving higher con‐
1813 currency and thus throughput than is possible via
1814 kernel NFS.
1815 exec Execute 3rd party tools. Could be used to perform
1817 monitoring during jobs runtime.
1818 I/O engine specific parameters
1820 In addition, there are some parameters which are only valid when a spe‐
1821 cific ioengine is in use. These are used identically to normal parame‐
1822 ters, with the caveat that when used on the command line, they must
1823 come after the ioengine that defines them is selected.
1824 (io_uring,libaio)cmdprio_percentage=int[,int]
1826 Set the percentage of I/O that will be issued with the highest
1827 priority. Default: 0. A single value applies to reads and
1828 writes. Comma-separated values may be specified for reads and
1829 writes. For this option to be effective, NCQ priority must be
1830 supported and enabled, and `direct=1' option must be used. fio
1831 must also be run as the root user. Unlike slat/clat/lat stats,
1832 which can be tracked and reported independently, per priority
1833 stats only track and report a single type of latency. By de‐
1834 fault, completion latency (clat) will be reported, if lat_per‐
1835 centiles is set, total latency (lat) will be reported.
1836 (io_uring,libaio)cmdprio_class=int[,int]
1838 Set the I/O priority class to use for I/Os that must be issued
1839 with a priority when cmdprio_percentage or cmdprio_bssplit is
1840 set. If not specified when cmdprio_percentage or cmdprio_bss‐
1841 plit is set, this defaults to the highest priority class. A sin‐
1842 gle value applies to reads and writes. Comma-separated values
1843 may be specified for reads and writes. See man ionice(1). See
1844 also the prioclass option.
1845 (io_uring,libaio)cmdprio=int[,int]
1847 Set the I/O priority value to use for I/Os that must be issued
1848 with a priority when cmdprio_percentage or cmdprio_bssplit is
1849 set. If not specified when cmdprio_percentage or cmdprio_bss‐
1850 plit is set, this defaults to 0. Linux limits us to a positive
1851 value between 0 and 7, with 0 being the highest. A single value
1852 applies to reads and writes. Comma-separated values may be
1853 specified for reads and writes. See man ionice(1). Refer to an
1854 appropriate manpage for other operating systems since the mean‐
1855 ing of priority may differ. See also the prio option.
1856 (io_uring,libaio)cmdprio_bssplit=str[,str]
1858 To get a finer control over I/O priority, this option allows
1859 specifying the percentage of IOs that must have a priority set
1860 depending on the block size of the IO. This option is useful
1861 only when used together with the option bssplit, that is, multi‐
1862 ple different block sizes are used for reads and writes.
1863 The first accepted format for this option is the same as the
1865 format of the bssplit option:
1866 cmdprio_bssplit=blocksize/percentage:blocksize/percentage
1868 In this case, each entry will use the priority class and prior‐
1870 ity level defined by the options cmdprio_class and cmdprio re‐
1871 spectively.
1872 The second accepted format for this option is:
1874 cmdprio_bssplit=blocksize/percentage/class/level:block‐
1876 size/percentage/class/level
1877 In this case, the priority class and priority level is defined
1879 inside each entry. In comparison with the first accepted format,
1880 the second accepted format does not restrict all entries to have
1881 the same priority class and priority level.
1882 For both formats, only the read and write data directions are
1884 supported, values for trim IOs are ignored. This option is mutu‐
1885 ally exclusive with the cmdprio_percentage option.
1886 (io_uring)fixedbufs
1888 If fio is asked to do direct IO, then Linux will map pages for
1889 each IO call, and release them when IO is done. If this option
1890 is set, the pages are pre-mapped before IO is started. This
1891 eliminates the need to map and release for each IO. This is
1892 more efficient, and reduces the IO latency as well.
1893 (io_uring)hipri
1895 If this option is set, fio will attempt to use polled IO comple‐
1896 tions. Normal IO completions generate interrupts to signal the
1897 completion of IO, polled completions do not. Hence they are re‐
1898 quire active reaping by the application. The benefits are more
1899 efficient IO for high IOPS scenarios, and lower latencies for
1900 low queue depth IO.
1901 (io_uring)registerfiles
1903 With this option, fio registers the set of files being used with
1904 the kernel. This avoids the overhead of managing file counts in
1905 the kernel, making the submission and completion part more
1906 lightweight. Required for the below sqthread_poll option.
1907 (io_uring)sqthread_poll
1909 Normally fio will submit IO by issuing a system call to notify
1910 the kernel of available items in the SQ ring. If this option is
1911 set, the act of submitting IO will be done by a polling thread
1912 in the kernel. This frees up cycles for fio, at the cost of us‐
1913 ing more CPU in the system.
1914 (io_uring)sqthread_poll_cpu
1916 When `sqthread_poll` is set, this option provides a way to de‐
1917 fine which CPU should be used for the polling thread.
1918 (libaio)userspace_reap
1920 Normally, with the libaio engine in use, fio will use the
1921 io_getevents(3) system call to reap newly returned events. With
1922 this flag turned on, the AIO ring will be read directly from
1923 user-space to reap events. The reaping mode is only enabled when
1924 polling for a minimum of 0 events (e.g. when `iodepth_batch_com‐
1925 plete=0').
1926 (pvsync2)hipri
1928 Set RWF_HIPRI on I/O, indicating to the kernel that it's of
1929 higher priority than normal.
1930 (pvsync2)hipri_percentage
1932 When hipri is set this determines the probability of a pvsync2
1933 I/O being high priority. The default is 100%.
1934 (pvsync2,libaio,io_uring)nowait
1936 By default if a request cannot be executed immediately (e.g. re‐
1937 source starvation, waiting on locks) it is queued and the initi‐
1938 ating process will be blocked until the required resource be‐
1939 comes free. This option sets the RWF_NOWAIT flag (supported
1940 from the 4.14 Linux kernel) and the call will return instantly
1941 with EAGAIN or a partial result rather than waiting.
1942 It is useful to also use ignore_error=EAGAIN when using this op‐
1944 tion. Note: glibc 2.27, 2.28 have a bug in syscall wrappers
1945 preadv2, pwritev2. They return EOPNOTSUP instead of EAGAIN.
1946 For cached I/O, using this option usually means a request oper‐
1948 ates only with cached data. Currently the RWF_NOWAIT flag does
1949 not supported for cached write. For direct I/O, requests will
1950 only succeed if cache invalidation isn't required, file blocks
1951 are fully allocated and the disk request could be issued immedi‐
1952 ately.
1953 (cpuio)cpuload=int
1955 Attempt to use the specified percentage of CPU cycles. This is a
1956 mandatory option when using cpuio I/O engine.
1957 (cpuio)cpuchunks=int
1959 Split the load into cycles of the given time. In microseconds.
1960 (cpuio)cpumode=str
1962 Specify how to stress the CPU. It can take these two values:
1963 noop This is the default and directs the CPU to execute
1965 noop instructions.
1966 qsort Replace the default noop instructions with a qsort
1968 algorithm to consume more energy.
1969 (cpuio)exit_on_io_done=bool
1971 Detect when I/O threads are done, then exit.
1972 (libhdfs)namenode=str
1974 The hostname or IP address of a HDFS cluster namenode to con‐
1975 tact.
1976 (libhdfs)port=int
1978 The listening port of the HFDS cluster namenode.
1979 (netsplice,net)port=int
1981 The TCP or UDP port to bind to or connect to. If this is used
1982 with numjobs to spawn multiple instances of the same job type,
1983 then this will be the starting port number since fio will use a
1984 range of ports.
1985 (rdma,librpma_*)port=int
1987 The port to use for RDMA-CM communication. This should be the
1988 same value on the client and the server side.
1989 (netsplice,net,rdma)hostname=str
1991 The hostname or IP address to use for TCP, UDP or RDMA-CM based
1992 I/O. If the job is a TCP listener or UDP reader, the hostname
1993 is not used and must be omitted unless it is a valid UDP multi‐
1994 cast address.
1995 (librpma_*)serverip=str
1997 The IP address to be used for RDMA-CM based I/O.
1998 (librpma_*_server)direct_write_to_pmem=bool
2000 Set to 1 only when Direct Write to PMem from the remote host is
2001 possible. Otherwise, set to 0.
2002 (librpma_*_server)busy_wait_polling=bool
2004 Set to 0 to wait for completion instead of busy-wait polling
2005 completion. Default: 1.
2006 (netsplice,net)interface=str
2008 The IP address of the network interface used to send or receive
2009 UDP multicast.
2010 (netsplice,net)ttl=int
2012 Time-to-live value for outgoing UDP multicast packets. Default:
2013 1.
2014 (netsplice,net)nodelay=bool
2016 Set TCP_NODELAY on TCP connections.
2017 (netsplice,net)protocol=str, proto=str
2019 The network protocol to use. Accepted values are:
2020 tcp Transmission control protocol.
2022 tcpv6 Transmission control protocol V6.
2024 udp User datagram protocol.
2026 udpv6 User datagram protocol V6.
2028 unix UNIX domain socket.
2030 When the protocol is TCP or UDP, the port must also be given, as
2032 well as the hostname if the job is a TCP listener or UDP reader.
2033 For unix sockets, the normal filename option should be used and
2034 the port is invalid.
2035 (netsplice,net)listen
2037 For TCP network connections, tell fio to listen for incoming
2038 connections rather than initiating an outgoing connection. The
2039 hostname must be omitted if this option is used.
2040 (netsplice,net)pingpong
2042 Normally a network writer will just continue writing data, and a
2043 network reader will just consume packages. If `pingpong=1' is
2044 set, a writer will send its normal payload to the reader, then
2045 wait for the reader to send the same payload back. This allows
2046 fio to measure network latencies. The submission and completion
2047 latencies then measure local time spent sending or receiving,
2048 and the completion latency measures how long it took for the
2049 other end to receive and send back. For UDP multicast traffic
2050 `pingpong=1' should only be set for a single reader when multi‐
2051 ple readers are listening to the same address.
2052 (netsplice,net)window_size=int
2054 Set the desired socket buffer size for the connection.
2055 (netsplice,net)mss=int
2057 Set the TCP maximum segment size (TCP_MAXSEG).
2058 (e4defrag)donorname=str
2060 File will be used as a block donor (swap extents between files).
2061 (e4defrag)inplace=int
2063 Configure donor file blocks allocation strategy:
2064 0 Default. Preallocate donor's file on init.
2066 1 Allocate space immediately inside defragment
2068 event, and free right after event.
2069 (rbd,rados)clustername=str
2071 Specifies the name of the Ceph cluster.
2072 (rbd)rbdname=str
2074 Specifies the name of the RBD.
2075 (rbd,rados)pool=str
2077 Specifies the name of the Ceph pool containing RBD or RADOS
2078 data.
2079 (rbd,rados)clientname=str
2081 Specifies the username (without the 'client.' prefix) used to
2082 access the Ceph cluster. If the clustername is specified, the
2083 clientname shall be the full *type.id* string. If no type. pre‐
2084 fix is given, fio will add 'client.' by default.
2085 (rbd,rados)busy_poll=bool
2087 Poll store instead of waiting for completion. Usually this pro‐
2088 vides better throughput at cost of higher(up to 100%) CPU uti‐
2089 lization.
2090 (rados)touch_objects=bool
2092 During initialization, touch (create if do not exist) all ob‐
2093 jects (files). Touching all objects affects ceph caches and
2094 likely impacts test results. Enabled by default.
2095 (http)http_host=str
2097 Hostname to connect to. For S3, this could be the bucket name.
2098 Default is localhost
2099 (http)http_user=str
2101 Username for HTTP authentication.
2102 (http)http_pass=str
2104 Password for HTTP authentication.
2105 (http)https=str
2107 Whether to use HTTPS instead of plain HTTP. on enables HTTPS;
2108 insecure will enable HTTPS, but disable SSL peer verification
2109 (use with caution!). Default is off.
2110 (http)http_mode=str
2112 Which HTTP access mode to use: webdav, swift, or s3. Default is
2113 webdav.
2114 (http)http_s3_region=str
2116 The S3 region/zone to include in the request. Default is us-
2117 east-1.
2118 (http)http_s3_key=str
2120 The S3 secret key.
2121 (http)http_s3_keyid=str
2123 The S3 key/access id.
2124 (http)http_swift_auth_token=str
2126 The Swift auth token. See the example configuration file on how
2127 to retrieve this.
2128 (http)http_verbose=int
2130 Enable verbose requests from libcurl. Useful for debugging. 1
2131 turns on verbose logging from libcurl, 2 additionally enables
2132 HTTP IO tracing. Default is 0
2133 (mtd)skip_bad=bool
2135 Skip operations against known bad blocks.
2136 (libhdfs)hdfsdirectory
2138 libhdfs will create chunk in this HDFS directory.
2139 (libhdfs)chunk_size
2141 The size of the chunk to use for each file.
2142 (rdma)verb=str
2144 The RDMA verb to use on this side of the RDMA ioengine connec‐
2145 tion. Valid values are write, read, send and recv. These corre‐
2146 spond to the equivalent RDMA verbs (e.g. write = rdma_write
2147 etc.). Note that this only needs to be specified on the client
2148 side of the connection. See the examples folder.
2149 (rdma)bindname=str
2151 The name to use to bind the local RDMA-CM connection to a local
2152 RDMA device. This could be a hostname or an IPv4 or IPv6 ad‐
2153 dress. On the server side this will be passed into the
2154 rdma_bind_addr() function and on the client site it will be used
2155 in the rdma_resolve_add() function. This can be useful when mul‐
2156 tiple paths exist between the client and the server or in cer‐
2157 tain loopback configurations.
2158 (filestat)stat_type=str
2160 Specify stat system call type to measure lookup/getattr perfor‐
2161 mance. Default is stat for stat(2).
2162 (sg)hipri
2164 If this option is set, fio will attempt to use polled IO comple‐
2165 tions. This will have a similar effect as (io_uring)hipri. Only
2166 SCSI READ and WRITE commands will have the SGV4_FLAG_HIPRI set
2167 (not UNMAP (trim) nor VERIFY). Older versions of the Linux sg
2168 driver that do not support hipri will simply ignore this flag
2169 and do normal IO. The Linux SCSI Low Level Driver (LLD) that
2170 "owns" the device also needs to support hipri (also known as
2171 iopoll and mq_poll). The MegaRAID driver is an example of a SCSI
2172 LLD. Default: clear (0) which does normal (interrupted based)
2173 IO.
2174 (sg)readfua=bool
2176 With readfua option set to 1, read operations include the force
2177 unit access (fua) flag. Default: 0.
2178 (sg)writefua=bool
2180 With writefua option set to 1, write operations include the
2181 force unit access (fua) flag. Default: 0.
2182 (sg)sg_write_mode=str
2184 Specify the type of write commands to issue. This option can
2185 take multiple values:
2186 write (default)
2188 Write opcodes are issued as usual
2189 write_and_verify
2191 Issue WRITE AND VERIFY commands. The BYTCHK bit is
2192 set to 00b. This directs the device to carry out a
2193 medium verification with no data comparison for
2194 the data that was written. The writefua option is
2195 ignored with this selection.
2196 verify This option is deprecated. Use write_and_verify
2198 instead.
2199 write_same
2201 Issue WRITE SAME commands. This transfers a single
2202 block to the device and writes this same block of
2203 data to a contiguous sequence of LBAs beginning at
2204 the specified offset. fio's block size parameter
2205 specifies the amount of data written with each
2206 command. However, the amount of data actually
2207 transferred to the device is equal to the device's
2208 block (sector) size. For a device with 512 byte
2209 sectors, blocksize=8k will write 16 sectors with
2210 each command. fio will still generate 8k of data
2211 for each command butonly the first 512 bytes will
2212 be used and transferred to the device. The write‐
2213 fua option is ignored with this selection.
2214 same This option is deprecated. Use write_same instead.
2216 write_same_ndob
2218 Issue WRITE SAME(16) commands as above but with
2219 the No Data Output Buffer (NDOB) bit set. No data
2220 will be transferred to the device with this bit
2221 set. Data written will be a pre-determined pattern
2222 such as all zeroes.
2223 write_stream
2225 Issue WRITE STREAM(16) commands. Use the stream_id
2226 option to specify the stream identifier.
2227 verify_bytchk_00
2229 Issue VERIFY commands with BYTCHK set to 00. This
2230 directs the device to carry out a medium verifica‐
2231 tion with no data comparison.
2232 verify_bytchk_01
2234 Issue VERIFY commands with BYTCHK set to 01. This
2235 directs the device to compare the data on the de‐
2236 vice with the data transferred to the device.
2237 verify_bytchk_11
2239 Issue VERIFY commands with BYTCHK set to 11. This
2240 transfers a single block to the device and com‐
2241 pares the contents of this block with the data on
2242 the device beginning at the specified offset.
2243 fio's block size parameter specifies the total
2244 amount of data compared with this command. How‐
2245 ever, only one block (sector) worth of data is
2246 transferred to the device. This is similar to the
2247 WRITE SAME command except that data is compared
2248 instead of written.
2249 (sg)stream_id=int
2251 Set the stream identifier for WRITE STREAM commands. If this is
2252 set to 0 (which is not a valid stream identifier) fio will open
2253 a stream and then close it when done. Default is 0.
2254 (nbd)uri=str
2256 Specify the NBD URI of the server to test. The string is a
2257 standard NBD URI (see https://github.com/NetworkBlockDe‐
2258 vice/nbd/tree/master/doc). Example URIs:
2259 nbd://localhost:10809
2261 nbd+unix:///?socket=/tmp/socket
2263 nbds://tlshost/exportname
2265 (libcufile)gpu_dev_ids=str
2267 Specify the GPU IDs to use with CUDA. This is a colon-separated
2268 list of int. GPUs are assigned to workers roundrobin. Default
2269 is 0.
2270 (libcufile)cuda_io=str
2272 Specify the type of I/O to use with CUDA. This option takes the
2273 following values:
2274 cufile (default)
2276 Use libcufile and nvidia-fs. This option performs
2277 I/O directly between a GPUDirect Storage filesys‐
2278 tem and GPU buffers, avoiding use of a bounce buf‐
2279 fer. If verify is set, cudaMemcpy is used to copy
2280 verification data between RAM and GPU(s). Verifi‐
2281 cation data is copied from RAM to GPU before a
2282 write and from GPU to RAM after a read. direct
2283 must be 1.
2284 posix Use POSIX to perform I/O with a RAM buffer, and
2286 use cudaMemcpy to transfer data between RAM and
2287 the GPU(s). Data is copied from GPU to RAM before
2288 a write and copied from RAM to GPU after a read.
2289 verify does not affect the use of cudaMemcpy.
2290 (dfs)pool
2292 Specify the label or UUID of the DAOS pool to connect to.
2293 (dfs)cont
2295 Specify the label or UUID of the DAOS container to open.
2296 (dfs)chunk_size
2298 Specificy a different chunk size (in bytes) for the dfs file.
2299 Use DAOS container's chunk size by default.
2300 (dfs)object_class
2302 Specificy a different object class for the dfs file. Use DAOS
2303 container's object class by default.
2304 (nfs)nfs_url
2306 URL in libnfs format, eg
2307 nfs://<server|ipv4|ipv6>/path[?arg=val[&arg=val]*] Refer to the
2308 libnfs README for more details.
2309 (exec)program=str
2311 Specify the program to execute. Note the program will receive a
2312 SIGTERM when the job is reaching the time limit. A SIGKILL is
2313 sent once the job is over. The delay between the two signals is
2314 defined by grace_time option.
2315 (exec)arguments=str
2317 Specify arguments to pass to program. Some special variables
2318 can be expanded to pass fio's job details to the program :
2319 %r replaced by the duration of the job in seconds
2321 %n replaced by the name of the job
2323 (exec)grace_time=int
2325 Defines the time between the SIGTERM and SIGKILL signals. De‐
2326 fault is 1 second.
2327 (exec)std_redirect=ool
2329 If set, stdout and stderr streams are redirected to files named
2330 from the job name. Default is true.
2331 I/O depth
2333 iodepth=int
2334 Number of I/O units to keep in flight against the file. Note
2335 that increasing iodepth beyond 1 will not affect synchronous io‐
2336 engines (except for small degrees when verify_async is in use).
2337 Even async engines may impose OS restrictions causing the de‐
2338 sired depth not to be achieved. This may happen on Linux when
2339 using libaio and not setting `direct=1', since buffered I/O is
2340 not async on that OS. Keep an eye on the I/O depth distribution
2341 in the fio output to verify that the achieved depth is as ex‐
2342 pected. Default: 1.
2343 iodepth_batch_submit=int, iodepth_batch=int
2345 This defines how many pieces of I/O to submit at once. It de‐
2346 faults to 1 which means that we submit each I/O as soon as it is
2347 available, but can be raised to submit bigger batches of I/O at
2348 the time. If it is set to 0 the iodepth value will be used.
2349 iodepth_batch_complete_min=int, iodepth_batch_complete=int
2351 This defines how many pieces of I/O to retrieve at once. It de‐
2352 faults to 1 which means that we'll ask for a minimum of 1 I/O in
2353 the retrieval process from the kernel. The I/O retrieval will go
2354 on until we hit the limit set by iodepth_low. If this variable
2355 is set to 0, then fio will always check for completed events be‐
2356 fore queuing more I/O. This helps reduce I/O latency, at the
2357 cost of more retrieval system calls.
2358 iodepth_batch_complete_max=int
2360 This defines maximum pieces of I/O to retrieve at once. This
2361 variable should be used along with iodepth_batch_com‐
2362 plete_min=int variable, specifying the range of min and max
2363 amount of I/O which should be retrieved. By default it is equal
2364 to iodepth_batch_complete_min value. Example #1:
2365 iodepth_batch_complete_min=1
2367 iodepth_batch_complete_max=<iodepth>
2368 which means that we will retrieve at least 1 I/O and up to the
2370 whole submitted queue depth. If none of I/O has been completed
2371 yet, we will wait. Example #2:
2372 iodepth_batch_complete_min=0
2374 iodepth_batch_complete_max=<iodepth>
2375 which means that we can retrieve up to the whole submitted queue
2377 depth, but if none of I/O has been completed yet, we will NOT
2378 wait and immediately exit the system call. In this example we
2379 simply do polling.
2380 iodepth_low=int
2382 The low water mark indicating when to start filling the queue
2383 again. Defaults to the same as iodepth, meaning that fio will
2384 attempt to keep the queue full at all times. If iodepth is set
2385 to e.g. 16 and iodepth_low is set to 4, then after fio has
2386 filled the queue of 16 requests, it will let the depth drain
2387 down to 4 before starting to fill it again.
2388 serialize_overlap=bool
2390 Serialize in-flight I/Os that might otherwise cause or suffer
2391 from data races. When two or more I/Os are submitted simultane‐
2392 ously, there is no guarantee that the I/Os will be processed or
2393 completed in the submitted order. Further, if two or more of
2394 those I/Os are writes, any overlapping region between them can
2395 become indeterminate/undefined on certain storage. These issues
2396 can cause verification to fail erratically when at least one of
2397 the racing I/Os is changing data and the overlapping region has
2398 a non-zero size. Setting serialize_overlap tells fio to avoid
2399 provoking this behavior by explicitly serializing in-flight I/Os
2400 that have a non-zero overlap. Note that setting this option can
2401 reduce both performance and the iodepth achieved.
2402 This option only applies to I/Os issued for a single job except
2404 when it is enabled along with io_submit_mode=offload. In offload
2405 mode, fio will check for overlap among all I/Os submitted by
2406 offload jobs with serialize_overlap enabled.
2407 Default: false.
2409 io_submit_mode=str
2411 This option controls how fio submits the I/O to the I/O engine.
2412 The default is `inline', which means that the fio job threads
2413 submit and reap I/O directly. If set to `offload', the job
2414 threads will offload I/O submission to a dedicated pool of I/O
2415 threads. This requires some coordination and thus has a bit of
2416 extra overhead, especially for lower queue depth I/O where it
2417 can increase latencies. The benefit is that fio can manage sub‐
2418 mission rates independently of the device completion rates. This
2419 avoids skewed latency reporting if I/O gets backed up on the de‐
2420 vice side (the coordinated omission problem). Note that this op‐
2421 tion cannot reliably be used with async IO engines.
2422 I/O rate
2424 thinktime=time
2425 Stall the job for the specified period of time after an I/O has
2426 completed before issuing the next. May be used to simulate pro‐
2427 cessing being done by an application. When the unit is omitted,
2428 the value is interpreted in microseconds. See thinktime_blocks,
2429 thinktime_iotime and thinktime_spin.
2430 thinktime_spin=time
2432 Only valid if thinktime is set - pretend to spend CPU time doing
2433 something with the data received, before falling back to sleep‐
2434 ing for the rest of the period specified by thinktime. When the
2435 unit is omitted, the value is interpreted in microseconds.
2436 thinktime_blocks=int
2438 Only valid if thinktime is set - control how many blocks to is‐
2439 sue, before waiting thinktime usecs. If not set, defaults to 1
2440 which will make fio wait thinktime usecs after every block. This
2441 effectively makes any queue depth setting redundant, since no
2442 more than 1 I/O will be queued before we have to complete it and
2443 do our thinktime. In other words, this setting effectively caps
2444 the queue depth if the latter is larger.
2445 thinktime_blocks_type=str
2447 Only valid if thinktime is set - control how thinktime_blocks
2448 triggers. The default is `complete', which triggers thinktime
2449 when fio completes thinktime_blocks blocks. If this is set to
2450 `issue', then the trigger happens at the issue side.
2451 thinktime_iotime=time
2453 Only valid if thinktime is set - control thinktime interval by
2454 time. The thinktime stall is repeated after IOs are executed
2455 for thinktime_iotime. For example, `--thinktime_iotime=9s
2456 --thinktime=1s' repeat 10-second cycle with IOs for 9 seconds
2457 and stall for 1 second. When the unit is omitted, thinktime_io‐
2458 time is interpreted as a number of seconds. If this option is
2459 used together with thinktime_blocks, the thinktime stall is re‐
2460 peated after thinktime_iotime or after thinktime_blocks IOs,
2461 whichever happens first.
2462 rate=int[,int][,int]
2465 Cap the bandwidth used by this job. The number is in bytes/sec,
2466 the normal suffix rules apply. Comma-separated values may be
2467 specified for reads, writes, and trims as described in block‐
2468 size.
2469 For example, using `rate=1m,500k' would limit reads to 1MiB/sec
2471 and writes to 500KiB/sec. Capping only reads or writes can be
2472 done with `rate=,500k' or `rate=500k,' where the former will
2473 only limit writes (to 500KiB/sec) and the latter will only limit
2474 reads.
2475 rate_min=int[,int][,int]
2477 Tell fio to do whatever it can to maintain at least this band‐
2478 width. Failing to meet this requirement will cause the job to
2479 exit. Comma-separated values may be specified for reads, writes,
2480 and trims as described in blocksize.
2481 rate_iops=int[,int][,int]
2483 Cap the bandwidth to this number of IOPS. Basically the same as
2484 rate, just specified independently of bandwidth. If the job is
2485 given a block size range instead of a fixed value, the smallest
2486 block size is used as the metric. Comma-separated values may be
2487 specified for reads, writes, and trims as described in block‐
2488 size.
2489 rate_iops_min=int[,int][,int]
2491 If fio doesn't meet this rate of I/O, it will cause the job to
2492 exit. Comma-separated values may be specified for reads,
2493 writes, and trims as described in blocksize.
2494 rate_process=str
2496 This option controls how fio manages rated I/O submissions. The
2497 default is `linear', which submits I/O in a linear fashion with
2498 fixed delays between I/Os that gets adjusted based on I/O com‐
2499 pletion rates. If this is set to `poisson', fio will submit I/O
2500 based on a more real world random request flow, known as the
2501 Poisson process (https://en.wikipedia.org/wiki/Pois‐
2502 son_point_process). The lambda will be 10^6 / IOPS for the given
2503 workload.
2504 rate_ignore_thinktime=bool
2506 By default, fio will attempt to catch up to the specified rate
2507 setting, if any kind of thinktime setting was used. If this op‐
2508 tion is set, then fio will ignore the thinktime and continue do‐
2509 ing IO at the specified rate, instead of entering a catch-up
2510 mode after thinktime is done.
2511 I/O latency
2513 latency_target=time
2514 If set, fio will attempt to find the max performance point that
2515 the given workload will run at while maintaining a latency below
2516 this target. When the unit is omitted, the value is interpreted
2517 in microseconds. See latency_window and latency_percentile.
2518 latency_window=time
2520 Used with latency_target to specify the sample window that the
2521 job is run at varying queue depths to test the performance. When
2522 the unit is omitted, the value is interpreted in microseconds.
2523 latency_percentile=float
2525 The percentage of I/Os that must fall within the criteria speci‐
2526 fied by latency_target and latency_window. If not set, this de‐
2527 faults to 100.0, meaning that all I/Os must be equal or below to
2528 the value set by latency_target.
2529 latency_run=bool
2531 Used with latency_target. If false (default), fio will find the
2532 highest queue depth that meets latency_target and exit. If true,
2533 fio will continue running and try to meet latency_target by ad‐
2534 justing queue depth.
2535 max_latency=time[,time][,time]
2537 If set, fio will exit the job with an ETIMEDOUT error if it ex‐
2538 ceeds this maximum latency. When the unit is omitted, the value
2539 is interpreted in microseconds. Comma-separated values may be
2540 specified for reads, writes, and trims as described in block‐
2541 size.
2542 rate_cycle=int
2544 Average bandwidth for rate and rate_min over this number of mil‐
2545 liseconds. Defaults to 1000.
2546 I/O replay
2548 write_iolog=str
2549 Write the issued I/O patterns to the specified file. See
2550 read_iolog. Specify a separate file for each job, otherwise the
2551 iologs will be interspersed and the file may be corrupt.
2552 read_iolog=str
2554 Open an iolog with the specified filename and replay the I/O
2555 patterns it contains. This can be used to store a workload and
2556 replay it sometime later. The iolog given may also be a blktrace
2557 binary file, which allows fio to replay a workload captured by
2558 blktrace. See blktrace(8) for how to capture such logging data.
2559 For blktrace replay, the file needs to be turned into a blkparse
2560 binary data file first (`blkparse <device> -o /dev/null -d
2561 file_for_fio.bin'). You can specify a number of files by sepa‐
2562 rating the names with a ':' character. See the filename option
2563 for information on how to escape ':' characters within the file
2564 names. These files will be sequentially assigned to job clones
2565 created by numjobs. '-' is a reserved name, meaning read from
2566 stdin, notably if filename is set to '-' which means stdin as
2567 well, then this flag can't be set to '-'.
2568 read_iolog_chunked=bool
2570 Determines how iolog is read. If false (default) entire
2571 read_iolog will be read at once. If selected true, input from
2572 iolog will be read gradually. Useful when iolog is very large,
2573 or it is generated.
2574 merge_blktrace_file=str
2576 When specified, rather than replaying the logs passed to
2577 read_iolog, the logs go through a merge phase which aggregates
2578 them into a single blktrace. The resulting file is then passed
2579 on as the read_iolog parameter. The intention here is to make
2580 the order of events consistent. This limits the influence of the
2581 scheduler compared to replaying multiple blktraces via concur‐
2582 rent jobs.
2583 merge_blktrace_scalars=float_list
2585 This is a percentage based option that is index paired with the
2586 list of files passed to read_iolog. When merging is performed,
2587 scale the time of each event by the corresponding amount. For
2588 example, `--merge_blktrace_scalars="50:100"' runs the first
2589 trace in halftime and the second trace in realtime. This knob is
2590 separately tunable from replay_time_scale which scales the trace
2591 during runtime and will not change the output of the merge un‐
2592 like this option.
2593 merge_blktrace_iters=float_list
2595 This is a whole number option that is index paired with the list
2596 of files passed to read_iolog. When merging is performed, run
2597 each trace for the specified number of iterations. For example,
2598 `--merge_blktrace_iters="2:1"' runs the first trace for two it‐
2599 erations and the second trace for one iteration.
2600 replay_no_stall=bool
2602 When replaying I/O with read_iolog the default behavior is to
2603 attempt to respect the timestamps within the log and replay them
2604 with the appropriate delay between IOPS. By setting this vari‐
2605 able fio will not respect the timestamps and attempt to replay
2606 them as fast as possible while still respecting ordering. The
2607 result is the same I/O pattern to a given device, but different
2608 timings.
2609 replay_time_scale=int
2611 When replaying I/O with read_iolog, fio will honor the original
2612 timing in the trace. With this option, it's possible to scale
2613 the time. It's a percentage option, if set to 50 it means run at
2614 50% the original IO rate in the trace. If set to 200, run at
2615 twice the original IO rate. Defaults to 100.
2616 replay_redirect=str
2618 While replaying I/O patterns using read_iolog the default behav‐
2619 ior is to replay the IOPS onto the major/minor device that each
2620 IOP was recorded from. This is sometimes undesirable because on
2621 a different machine those major/minor numbers can map to a dif‐
2622 ferent device. Changing hardware on the same system can also re‐
2623 sult in a different major/minor mapping. replay_redirect causes
2624 all I/Os to be replayed onto the single specified device regard‐
2625 less of the device it was recorded from. i.e. `replay_redi‐
2626 rect=/dev/sdc' would cause all I/O in the blktrace or iolog to
2627 be replayed onto `/dev/sdc'. This means multiple devices will be
2628 replayed onto a single device, if the trace contains multiple
2629 devices. If you want multiple devices to be replayed concur‐
2630 rently to multiple redirected devices you must blkparse your
2631 trace into separate traces and replay them with independent fio
2632 invocations. Unfortunately this also breaks the strict time or‐
2633 dering between multiple device accesses.
2634 replay_align=int
2636 Force alignment of the byte offsets in a trace to this value.
2637 The value must be a power of 2.
2638 replay_scale=int
2640 Scale bye offsets down by this factor when replaying traces.
2641 Should most likely use replay_align as well.
2642 Threads, processes and job synchronization
2644 replay_skip=str
2645 Sometimes it's useful to skip certain IO types in a replay
2646 trace. This could be, for instance, eliminating the writes in
2647 the trace. Or not replaying the trims/discards, if you are redi‐
2648 recting to a device that doesn't support them. This option
2649 takes a comma separated list of read, write, trim, sync.
2650 thread Fio defaults to creating jobs by using fork, however if this op‐
2652 tion is given, fio will create jobs by using POSIX Threads'
2653 function pthread_create(3) to create threads instead.
2654 wait_for=str
2656 If set, the current job won't be started until all workers of
2657 the specified waitee job are done. wait_for operates on the job
2658 name basis, so there are a few limitations. First, the waitee
2659 must be defined prior to the waiter job (meaning no forward ref‐
2660 erences). Second, if a job is being referenced as a waitee, it
2661 must have a unique name (no duplicate waitees).
2662 nice=int
2664 Run the job with the given nice value. See man nice(2). On Win‐
2665 dows, values less than -15 set the process class to "High"; -1
2666 through -15 set "Above Normal"; 1 through 15 "Below Normal"; and
2667 above 15 "Idle" priority class.
2668 prio=int
2670 Set the I/O priority value of this job. Linux limits us to a
2671 positive value between 0 and 7, with 0 being the highest. See
2672 man ionice(1). Refer to an appropriate manpage for other operat‐
2673 ing systems since meaning of priority may differ. For per-com‐
2674 mand priority setting, see the I/O engine specific `cmdprio_per‐
2675 centage` and `cmdprio` options.
2676 prioclass=int
2678 Set the I/O priority class. See man ionice(1). For per-command
2679 priority setting, see the I/O engine specific `cmdprio_percent‐
2680 age` and `cmdprio_class` options.
2681 cpus_allowed=str
2683 Controls the same options as cpumask, but accepts a textual
2684 specification of the permitted CPUs instead and CPUs are indexed
2685 from 0. So to use CPUs 0 and 5 you would specify `cpus_al‐
2686 lowed=0,5'. This option also allows a range of CPUs to be speci‐
2687 fied -- say you wanted a binding to CPUs 0, 5, and 8 to 15, you
2688 would set `cpus_allowed=0,5,8-15'.
2689 On Windows, when `cpus_allowed' is unset only CPUs from fio's
2691 current processor group will be used and affinity settings are
2692 inherited from the system. An fio build configured to target
2693 Windows 7 makes options that set CPUs processor group aware and
2694 values will set both the processor group and a CPU from within
2695 that group. For example, on a system where processor group 0 has
2696 40 CPUs and processor group 1 has 32 CPUs, `cpus_allowed' values
2697 between 0 and 39 will bind CPUs from processor group 0 and
2698 `cpus_allowed' values between 40 and 71 will bind CPUs from pro‐
2699 cessor group 1. When using `cpus_allowed_policy=shared' all CPUs
2700 specified by a single `cpus_allowed' option must be from the
2701 same processor group. For Windows fio builds not built for Win‐
2702 dows 7, CPUs will only be selected from (and be relative to)
2703 whatever processor group fio happens to be running in and CPUs
2704 from other processor groups cannot be used.
2705 cpus_allowed_policy=str
2707 Set the policy of how fio distributes the CPUs specified by
2708 cpus_allowed or cpumask. Two policies are supported:
2709 shared All jobs will share the CPU set specified.
2711 split Each job will get a unique CPU from the CPU set.
2713 shared is the default behavior, if the option isn't specified.
2715 If split is specified, then fio will assign one cpu per job. If
2716 not enough CPUs are given for the jobs listed, then fio will
2717 roundrobin the CPUs in the set.
2718 cpumask=int
2720 Set the CPU affinity of this job. The parameter given is a bit
2721 mask of allowed CPUs the job may run on. So if you want the al‐
2722 lowed CPUs to be 1 and 5, you would pass the decimal value of (1
2723 << 1 | 1 << 5), or 34. See man sched_setaffinity(2). This may
2724 not work on all supported operating systems or kernel versions.
2725 This option doesn't work well for a higher CPU count than what
2726 you can store in an integer mask, so it can only control cpus
2727 1-32. For boxes with larger CPU counts, use cpus_allowed.
2728 numa_cpu_nodes=str
2730 Set this job running on specified NUMA nodes' CPUs. The argu‐
2731 ments allow comma delimited list of cpu numbers, A-B ranges, or
2732 `all'. Note, to enable NUMA options support, fio must be built
2733 on a system with libnuma-dev(el) installed.
2734 numa_mem_policy=str
2736 Set this job's memory policy and corresponding NUMA nodes. For‐
2737 mat of the arguments:
2738 <mode>[:<nodelist>]
2740 `mode' is one of the following memory policies: `default', `pre‐
2742 fer', `bind', `interleave' or `local'. For `default' and `local'
2743 memory policies, no node needs to be specified. For `prefer',
2744 only one node is allowed. For `bind' and `interleave' the `node‐
2745 list' may be as follows: a comma delimited list of numbers, A-B
2746 ranges, or `all'.
2747 cgroup=str
2749 Add job to this control group. If it doesn't exist, it will be
2750 created. The system must have a mounted cgroup blkio mount point
2751 for this to work. If your system doesn't have it mounted, you
2752 can do so with:
2753 # mount -t cgroup -o blkio none /cgroup
2755 cgroup_weight=int
2757 Set the weight of the cgroup to this value. See the documenta‐
2758 tion that comes with the kernel, allowed values are in the range
2759 of 100..1000.
2760 cgroup_nodelete=bool
2762 Normally fio will delete the cgroups it has created after the
2763 job completion. To override this behavior and to leave cgroups
2764 around after the job completion, set `cgroup_nodelete=1'. This
2765 can be useful if one wants to inspect various cgroup files after
2766 job completion. Default: false.
2767 flow_id=int
2769 The ID of the flow. If not specified, it defaults to being a
2770 global flow. See flow.
2771 flow=int
2773 Weight in token-based flow control. If this value is used, then
2774 fio regulates the activity between two or more jobs sharing the
2775 same flow_id. Fio attempts to keep each job activity propor‐
2776 tional to other jobs' activities in the same flow_id group, with
2777 respect to requested weight per job. That is, if one job has
2778 `flow=3', another job has `flow=2' and another with `flow=1`,
2779 then there will be a roughly 3:2:1 ratio in how much one runs vs
2780 the others.
2781 flow_sleep=int
2783 The period of time, in microseconds, to wait after the flow
2784 counter has exceeded its proportion before retrying operations.
2785 stonewall, wait_for_previous
2787 Wait for preceding jobs in the job file to exit, before starting
2788 this one. Can be used to insert serialization points in the job
2789 file. A stone wall also implies starting a new reporting group,
2790 see group_reporting. Optionally you can use `stonewall=0` to
2791 disable or `stonewall=1` to enable it.
2792 exitall
2794 By default, fio will continue running all other jobs when one
2795 job finishes. Sometimes this is not the desired action. Setting
2796 exitall will instead make fio terminate all jobs in the same
2797 group, as soon as one job of that group finishes.
2798 exit_what=str
2800 By default, fio will continue running all other jobs when one
2801 job finishes. Sometimes this is not the desired action. Setting
2802 exitall will instead make fio terminate all jobs in the same
2803 group. The option exit_what allows you to control which jobs get
2804 terminated when exitall is enabled. The default value is group.
2805 The allowed values are:
2806 all terminates all jobs.
2808 group is the default and does not change the behaviour
2810 of exitall.
2811 stonewall
2813 terminates all currently running jobs across all
2814 groups and continues execution with the next
2815 stonewalled group.
2816 exec_prerun=str
2818 Before running this job, issue the command specified through
2819 system(3). Output is redirected in a file called `jobname.pre‐
2820 run.txt'.
2821 exec_postrun=str
2823 After the job completes, issue the command specified though sys‐
2824 tem(3). Output is redirected in a file called `job‐
2825 name.postrun.txt'.
2826 uid=int
2828 Instead of running as the invoking user, set the user ID to this
2829 value before the thread/process does any work.
2830 gid=int
2832 Set group ID, see uid.
2833 Verification
2835 verify_only
2836 Do not perform specified workload, only verify data still
2837 matches previous invocation of this workload. This option allows
2838 one to check data multiple times at a later date without over‐
2839 writing it. This option makes sense only for workloads that
2840 write data, and does not support workloads with the time_based
2841 option set.
2842 do_verify=bool
2844 Run the verify phase after a write phase. Only valid if verify
2845 is set. Default: true.
2846 verify=str
2848 If writing to a file, fio can verify the file contents after
2849 each iteration of the job. Each verification method also implies
2850 verification of special header, which is written to the begin‐
2851 ning of each block. This header also includes meta information,
2852 like offset of the block, block number, timestamp when block was
2853 written, etc. verify can be combined with verify_pattern option.
2854 The allowed values are:
2855 md5 Use an md5 sum of the data area and store it in
2857 the header of each block.
2858 crc64 Use an experimental crc64 sum of the data area and
2860 store it in the header of each block.
2861 crc32c Use a crc32c sum of the data area and store it in
2863 the header of each block. This will automatically
2864 use hardware acceleration (e.g. SSE4.2 on an x86
2865 or CRC crypto extensions on ARM64) but will fall
2866 back to software crc32c if none is found. Gener‐
2867 ally the fastest checksum fio supports when hard‐
2868 ware accelerated.
2869 crc32c-intel
2871 Synonym for crc32c.
2872 crc32 Use a crc32 sum of the data area and store it in
2874 the header of each block.
2875 crc16 Use a crc16 sum of the data area and store it in
2877 the header of each block.
2878 crc7 Use a crc7 sum of the data area and store it in
2880 the header of each block.
2881 xxhash Use xxhash as the checksum function. Generally the
2883 fastest software checksum that fio supports.
2884 sha512 Use sha512 as the checksum function.
2886 sha256 Use sha256 as the checksum function.
2888 sha1 Use optimized sha1 as the checksum function.
2890 sha3-224
2892 Use optimized sha3-224 as the checksum function.
2893 sha3-256
2895 Use optimized sha3-256 as the checksum function.
2896 sha3-384
2898 Use optimized sha3-384 as the checksum function.
2899 sha3-512
2901 Use optimized sha3-512 as the checksum function.
2902 meta This option is deprecated, since now meta informa‐
2904 tion is included in generic verification header
2905 and meta verification happens by default. For de‐
2906 tailed information see the description of the ver‐
2907 ify setting. This option is kept because of com‐
2908 patibility's sake with old configurations. Do not
2909 use it.
2910 pattern
2912 Verify a strict pattern. Normally fio includes a
2913 header with some basic information and checksum‐
2914 ming, but if this option is set, only the specific
2915 pattern set with verify_pattern is verified.
2916 null Only pretend to verify. Useful for testing inter‐
2918 nals with `ioengine=null', not for much else.
2919 This option can be used for repeated burn-in tests of a system
2921 to make sure that the written data is also correctly read back.
2922 If the data direction given is a read or random read, fio will
2923 assume that it should verify a previously written file. If the
2924 data direction includes any form of write, the verify will be of
2925 the newly written data.
2926 To avoid false verification errors, do not use the norandommap
2928 option when verifying data with async I/O engines and I/O depths
2929 > 1. Or use the norandommap and the lfsr random generator to‐
2930 gether to avoid writing to the same offset with multiple out‐
2931 standing I/Os.
2932 verify_offset=int
2934 Swap the verification header with data somewhere else in the
2935 block before writing. It is swapped back before verifying.
2936 verify_interval=int
2938 Write the verification header at a finer granularity than the
2939 blocksize. It will be written for chunks the size of verify_in‐
2940 terval. blocksize should divide this evenly.
2941 verify_pattern=str
2943 If set, fio will fill the I/O buffers with this pattern. Fio de‐
2944 faults to filling with totally random bytes, but sometimes it's
2945 interesting to fill with a known pattern for I/O verification
2946 purposes. Depending on the width of the pattern, fio will fill
2947 1/2/3/4 bytes of the buffer at the time (it can be either a dec‐
2948 imal or a hex number). The verify_pattern if larger than a
2949 32-bit quantity has to be a hex number that starts with either
2950 "0x" or "0X". Use with verify. Also, verify_pattern supports %o
2951 format, which means that for each block offset will be written
2952 and then verified back, e.g.:
2953 verify_pattern=%o
2955 Or use combination of everything:
2957 verify_pattern=0xff%o"abcd"-12
2959 verify_fatal=bool
2961 Normally fio will keep checking the entire contents before quit‐
2962 ting on a block verification failure. If this option is set, fio
2963 will exit the job on the first observed failure. Default: false.
2964 verify_dump=bool
2966 If set, dump the contents of both the original data block and
2967 the data block we read off disk to files. This allows later
2968 analysis to inspect just what kind of data corruption occurred.
2969 Off by default.
2970 verify_async=int
2972 Fio will normally verify I/O inline from the submitting thread.
2973 This option takes an integer describing how many async offload
2974 threads to create for I/O verification instead, causing fio to
2975 offload the duty of verifying I/O contents to one or more sepa‐
2976 rate threads. If using this offload option, even sync I/O en‐
2977 gines can benefit from using an iodepth setting higher than 1,
2978 as it allows them to have I/O in flight while verifies are run‐
2979 ning. Defaults to 0 async threads, i.e. verification is not
2980 asynchronous.
2981 verify_async_cpus=str
2983 Tell fio to set the given CPU affinity on the async I/O verifi‐
2984 cation threads. See cpus_allowed for the format used.
2985 verify_backlog=int
2987 Fio will normally verify the written contents of a job that uti‐
2988 lizes verify once that job has completed. In other words, every‐
2989 thing is written then everything is read back and verified. You
2990 may want to verify continually instead for a variety of reasons.
2991 Fio stores the meta data associated with an I/O block in memory,
2992 so for large verify workloads, quite a bit of memory would be
2993 used up holding this meta data. If this option is enabled, fio
2994 will write only N blocks before verifying these blocks.
2995 verify_backlog_batch=int
2997 Control how many blocks fio will verify if verify_backlog is
2998 set. If not set, will default to the value of verify_backlog
2999 (meaning the entire queue is read back and verified). If ver‐
3000 ify_backlog_batch is less than verify_backlog then not all
3001 blocks will be verified, if verify_backlog_batch is larger than
3002 verify_backlog, some blocks will be verified more than once.
3003 verify_state_save=bool
3005 When a job exits during the write phase of a verify workload,
3006 save its current state. This allows fio to replay up until that
3007 point, if the verify state is loaded for the verify read phase.
3008 The format of the filename is, roughly:
3009 <type>-<jobname>-<jobindex>-verify.state.
3011 <type> is "local" for a local run, "sock" for a client/server
3013 socket connection, and "ip" (192.168.0.1, for instance) for a
3014 networked client/server connection. Defaults to true.
3015 verify_state_load=bool
3017 If a verify termination trigger was used, fio stores the current
3018 write state of each thread. This can be used at verification
3019 time so that fio knows how far it should verify. Without this
3020 information, fio will run a full verification pass, according to
3021 the settings in the job file used. Default false.
3022 trim_percentage=int
3024 Number of verify blocks to discard/trim.
3025 trim_verify_zero=bool
3027 Verify that trim/discarded blocks are returned as zeros.
3028 trim_backlog=int
3030 Verify that trim/discarded blocks are returned as zeros.
3031 trim_backlog_batch=int
3033 Trim this number of I/O blocks.
3034 experimental_verify=bool
3036 Enable experimental verification.
3037 Steady state
3039 steadystate=str:float, ss=str:float
3040 Define the criterion and limit for assessing steady state per‐
3041 formance. The first parameter designates the criterion whereas
3042 the second parameter sets the threshold. When the criterion
3043 falls below the threshold for the specified duration, the job
3044 will stop. For example, `iops_slope:0.1%' will direct fio to
3045 terminate the job when the least squares regression slope falls
3046 below 0.1% of the mean IOPS. If group_reporting is enabled this
3047 will apply to all jobs in the group. Below is the list of avail‐
3048 able steady state assessment criteria. All assessments are car‐
3049 ried out using only data from the rolling collection window.
3050 Threshold limits can be expressed as a fixed value or as a per‐
3051 centage of the mean in the collection window.
3052 When using this feature, most jobs should include the time_based
3054 and runtime options or the loops option so that fio does not
3055 stop running after it has covered the full size of the specified
3056 file(s) or device(s).
3057 iops Collect IOPS data. Stop the job if all in‐
3059 dividual IOPS measurements are within the
3060 specified limit of the mean IOPS (e.g.,
3061 `iops:2' means that all individual IOPS
3062 values must be within 2 of the mean,
3063 whereas `iops:0.2%' means that all individ‐
3064 ual IOPS values must be within 0.2% of the
3065 mean IOPS to terminate the job).
3066 iops_slope
3068 Collect IOPS data and calculate the least
3069 squares regression slope. Stop the job if
3070 the slope falls below the specified limit.
3071 bw Collect bandwidth data. Stop the job if all
3073 individual bandwidth measurements are
3074 within the specified limit of the mean
3075 bandwidth.
3076 bw_slope
3078 Collect bandwidth data and calculate the
3079 least squares regression slope. Stop the
3080 job if the slope falls below the specified
3081 limit.
3082 steadystate_duration=time, ss_dur=time
3084 A rolling window of this duration will be used to judge
3085 whether steady state has been reached. Data will be col‐
3086 lected once per second. The default is 0 which disables
3087 steady state detection. When the unit is omitted, the
3088 value is interpreted in seconds.
3089 steadystate_ramp_time=time, ss_ramp=time
3091 Allow the job to run for the specified duration before
3092 beginning data collection for checking the steady state
3093 job termination criterion. The default is 0. When the
3094 unit is omitted, the value is interpreted in seconds.
3095 Measurements and reporting
3097 per_job_logs=bool
3098 If set, this generates bw/clat/iops log with per file private
3099 filenames. If not set, jobs with identical names will share the
3100 log filename. Default: true.
3101 group_reporting
3103 It may sometimes be interesting to display statistics for groups
3104 of jobs as a whole instead of for each individual job. This is
3105 especially true if numjobs is used; looking at individual
3106 thread/process output quickly becomes unwieldy. To see the final
3107 report per-group instead of per-job, use group_reporting. Jobs
3108 in a file will be part of the same reporting group, unless if
3109 separated by a stonewall, or by using new_group.
3110 new_group
3112 Start a new reporting group. See: group_reporting. If not given,
3113 all jobs in a file will be part of the same reporting group, un‐
3114 less separated by a stonewall.
3115 stats=bool
3117 By default, fio collects and shows final output results for all
3118 jobs that run. If this option is set to 0, then fio will ignore
3119 it in the final stat output.
3120 write_bw_log=str
3122 If given, write a bandwidth log for this job. Can be used to
3123 store data of the bandwidth of the jobs in their lifetime.
3124 If no str argument is given, the default filename of `job‐
3126 name_type.x.log' is used. Even when the argument is given, fio
3127 will still append the type of log. So if one specifies:
3128 write_bw_log=foo
3130 The actual log name will be `foo_bw.x.log' where `x' is the in‐
3132 dex of the job (1..N, where N is the number of jobs). If
3133 per_job_logs is false, then the filename will not include the
3134 `.x` job index.
3135 The included fio_generate_plots script uses gnuplot to turn
3137 these text files into nice graphs. See the LOG FILE FORMATS sec‐
3138 tion for how data is structured within the file.
3139 write_lat_log=str
3141 Same as write_bw_log, except this option creates I/O submission
3142 (e.g., `name_slat.x.log'), completion (e.g., `name_clat.x.log'),
3143 and total (e.g., `name_lat.x.log') latency files instead. See
3144 write_bw_log for details about the filename format and the LOG
3145 FILE FORMATS section for how data is structured within the
3146 files.
3147 write_hist_log=str
3149 Same as write_bw_log but writes an I/O completion latency his‐
3150 togram file (e.g., `name_hist.x.log') instead. Note that this
3151 file will be empty unless log_hist_msec has also been set. See
3152 write_bw_log for details about the filename format and the LOG
3153 FILE FORMATS section for how data is structured within the file.
3154 write_iops_log=str
3156 Same as write_bw_log, but writes an IOPS file (e.g.
3157 `name_iops.x.log`) instead. Because fio defaults to individual
3158 I/O logging, the value entry in the IOPS log will be 1 unless
3159 windowed logging (see log_avg_msec) has been enabled. See
3160 write_bw_log for details about the filename format and LOG FILE
3161 FORMATS for how data is structured within the file.
3162 log_entries=int
3164 By default, fio will log an entry in the iops, latency, or bw
3165 log for every I/O that completes. The initial number of I/O log
3166 entries is 1024. When the log entries are all used, new log en‐
3167 tries are dynamically allocated. This dynamic log entry alloca‐
3168 tion may negatively impact time-related statistics such as I/O
3169 tail latencies (e.g. 99.9th percentile completion latency). This
3170 option allows specifying a larger initial number of log entries
3171 to avoid run-time allocation of new log entries, resulting in
3172 more precise time-related I/O statistics. Also see log_avg_msec
3173 as well. Defaults to 1024.
3174 log_avg_msec=int
3176 By default, fio will log an entry in the iops, latency, or bw
3177 log for every I/O that completes. When writing to the disk log,
3178 that can quickly grow to a very large size. Setting this option
3179 makes fio average the each log entry over the specified period
3180 of time, reducing the resolution of the log. See log_max_value
3181 as well. Defaults to 0, logging all entries. Also see LOG FILE
3182 FORMATS section.
3183 log_hist_msec=int
3185 Same as log_avg_msec, but logs entries for completion latency
3186 histograms. Computing latency percentiles from averages of in‐
3187 tervals using log_avg_msec is inaccurate. Setting this option
3188 makes fio log histogram entries over the specified period of
3189 time, reducing log sizes for high IOPS devices while retaining
3190 percentile accuracy. See log_hist_coarseness and write_hist_log
3191 as well. Defaults to 0, meaning histogram logging is disabled.
3192 log_hist_coarseness=int
3194 Integer ranging from 0 to 6, defining the coarseness of the res‐
3195 olution of the histogram logs enabled with log_hist_msec. For
3196 each increment in coarseness, fio outputs half as many bins. De‐
3197 faults to 0, for which histogram logs contain 1216 latency bins.
3198 See LOG FILE FORMATS section.
3199 log_max_value=bool
3201 If log_avg_msec is set, fio logs the average over that window.
3202 If you instead want to log the maximum value, set this option to
3203 1. Defaults to 0, meaning that averaged values are logged.
3204 log_offset=bool
3206 If this is set, the iolog options will include the byte offset
3207 for the I/O entry as well as the other data values. Defaults to
3208 0 meaning that offsets are not present in logs. Also see LOG
3209 FILE FORMATS section.
3210 log_prio=bool
3212 If this is set, the iolog options will include the I/O priority
3213 for the I/O entry as well as the other data values. Defaults to
3214 0 meaning that I/O priorities are not present in logs. Also see
3215 LOG FILE FORMATS section.
3216 log_compression=int
3218 If this is set, fio will compress the I/O logs as it goes, to
3219 keep the memory footprint lower. When a log reaches the speci‐
3220 fied size, that chunk is removed and compressed in the back‐
3221 ground. Given that I/O logs are fairly highly compressible, this
3222 yields a nice memory savings for longer runs. The downside is
3223 that the compression will consume some background CPU cycles, so
3224 it may impact the run. This, however, is also true if the log‐
3225 ging ends up consuming most of the system memory. So pick your
3226 poison. The I/O logs are saved normally at the end of a run, by
3227 decompressing the chunks and storing them in the specified log
3228 file. This feature depends on the availability of zlib.
3229 log_compression_cpus=str
3231 Define the set of CPUs that are allowed to handle online log
3232 compression for the I/O jobs. This can provide better isolation
3233 between performance sensitive jobs, and background compression
3234 work. See cpus_allowed for the format used.
3235 log_store_compressed=bool
3237 If set, fio will store the log files in a compressed format.
3238 They can be decompressed with fio, using the --inflate-log com‐
3239 mand line parameter. The files will be stored with a `.fz' suf‐
3240 fix.
3241 log_unix_epoch=bool
3243 If set, fio will log Unix timestamps to the log files produced
3244 by enabling write_type_log for each log type, instead of the de‐
3245 fault zero-based timestamps.
3246 log_alternate_epoch=bool
3248 If set, fio will log timestamps based on the epoch used by the
3249 clock specified in the log_alternate_epoch_clock_id option, to
3250 the log files produced by enabling write_type_log for each log
3251 type, instead of the default zero-based timestamps.
3252 log_alternate_epoch_clock_id=int
3254 Specifies the clock_id to be used by clock_gettime to obtain the
3255 alternate epoch if either Blog_unix_epoch or log_alternate_epoch
3256 are true. Otherwise has no effect. Default value is 0, or
3257 CLOCK_REALTIME.
3258 block_error_percentiles=bool
3260 If set, record errors in trim block-sized units from writes and
3261 trims and output a histogram of how many trims it took to get to
3262 errors, and what kind of error was encountered.
3263 bwavgtime=int
3265 Average the calculated bandwidth over the given time. Value is
3266 specified in milliseconds. If the job also does bandwidth log‐
3267 ging through write_bw_log, then the minimum of this option and
3268 log_avg_msec will be used. Default: 500ms.
3269 iopsavgtime=int
3271 Average the calculated IOPS over the given time. Value is speci‐
3272 fied in milliseconds. If the job also does IOPS logging through
3273 write_iops_log, then the minimum of this option and log_avg_msec
3274 will be used. Default: 500ms.
3275 disk_util=bool
3277 Generate disk utilization statistics, if the platform supports
3278 it. Default: true.
3279 disable_lat=bool
3281 Disable measurements of total latency numbers. Useful only for
3282 cutting back the number of calls to gettimeofday(2), as that
3283 does impact performance at really high IOPS rates. Note that to
3284 really get rid of a large amount of these calls, this option
3285 must be used with disable_slat and disable_bw_measurement as
3286 well.
3287 disable_clat=bool
3289 Disable measurements of completion latency numbers. See dis‐
3290 able_lat.
3291 disable_slat=bool
3293 Disable measurements of submission latency numbers. See dis‐
3294 able_lat.
3295 disable_bw_measurement=bool, disable_bw=bool
3297 Disable measurements of throughput/bandwidth numbers. See dis‐
3298 able_lat.
3299 slat_percentiles=bool
3301 Report submission latency percentiles. Submission latency is not
3302 recorded for synchronous ioengines.
3303 clat_percentiles=bool
3305 Report completion latency percentiles.
3306 lat_percentiles=bool
3308 Report total latency percentiles. Total latency is the sum of
3309 submission latency and completion latency.
3310 percentile_list=float_list
3312 Overwrite the default list of percentiles for latencies and the
3313 block error histogram. Each number is a floating point number in
3314 the range (0,100], and the maximum length of the list is 20. Use
3315 ':' to separate the numbers. For example, `--per‐
3316 centile_list=99.5:99.9' will cause fio to report the latency du‐
3317 rations below which 99.5% and 99.9% of the observed latencies
3318 fell, respectively.
3319 significant_figures=int
3321 If using --output-format of `normal', set the significant fig‐
3322 ures to this value. Higher values will yield more precise IOPS
3323 and throughput units, while lower values will round. Requires a
3324 minimum value of 1 and a maximum value of 10. Defaults to 4.
3325 Error handling
3327 exitall_on_error
3328 When one job finishes in error, terminate the rest. The default
3329 is to wait for each job to finish.
3330 continue_on_error=str
3332 Normally fio will exit the job on the first observed failure. If
3333 this option is set, fio will continue the job when there is a
3334 'non-fatal error' (EIO or EILSEQ) until the runtime is exceeded
3335 or the I/O size specified is completed. If this option is used,
3336 there are two more stats that are appended, the total error
3337 count and the first error. The error field given in the stats is
3338 the first error that was hit during the run. The allowed values
3339 are:
3340 none Exit on any I/O or verify errors.
3342 read Continue on read errors, exit on all others.
3344 write Continue on write errors, exit on all others.
3346 io Continue on any I/O error, exit on all others.
3348 verify Continue on verify errors, exit on all others.
3350 all Continue on all errors.
3352 0 Backward-compatible alias for 'none'.
3354 1 Backward-compatible alias for 'all'.
3356 ignore_error=str
3358 Sometimes you want to ignore some errors during test in that
3359 case you can specify error list for each error type, instead of
3360 only being able to ignore the default 'non-fatal error' using
3361 continue_on_error. `ignore_er‐
3362 ror=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST' errors for
3363 given error type is separated with ':'. Error may be symbol
3364 ('ENOSPC', 'ENOMEM') or integer. Example:
3365 ignore_error=EAGAIN,ENOSPC:122
3367 This option will ignore EAGAIN from READ, and ENOSPC and
3369 122(EDQUOT) from WRITE. This option works by overriding con‐
3370 tinue_on_error with the list of errors for each error type if
3371 any.
3372 error_dump=bool
3374 If set dump every error even if it is non fatal, true by de‐
3375 fault. If disabled only fatal error will be dumped.
3376 Running predefined workloads
3378 Fio includes predefined profiles that mimic the I/O workloads generated
3379 by other tools.
3380 profile=str
3382 The predefined workload to run. Current profiles are:
3383 tiobench
3385 Threaded I/O bench (tiotest/tiobench) like work‐
3386 load.
3387 act Aerospike Certification Tool (ACT) like workload.
3389 To view a profile's additional options use --cmdhelp after specifying
3391 the profile. For example:
3392 $ fio --profile=act --cmdhelp
3394 Act profile options
3396 device-names=str
3397 Devices to use.
3398 load=int
3400 ACT load multiplier. Default: 1.
3401 test-duration=time
3403 How long the entire test takes to run. When the unit is omitted,
3404 the value is given in seconds. Default: 24h.
3405 threads-per-queue=int
3407 Number of read I/O threads per device. Default: 8.
3408 read-req-num-512-blocks=int
3410 Number of 512B blocks to read at the time. Default: 3.
3411 large-block-op-kbytes=int
3413 Size of large block ops in KiB (writes). Default: 131072.
3414 prep Set to run ACT prep phase.
3416 Tiobench profile options
3418 size=str
3419 Size in MiB.
3420 block=int
3422 Block size in bytes. Default: 4096.
3423 numruns=int
3425 Number of runs.
3426 dir=str
3428 Test directory.
3429 threads=int
3431 Number of threads.
3432
3434 Fio spits out a lot of output. While running, fio will display the sta‐
3435 tus of the jobs created. An example of that would be:
3436 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]
3438 The characters inside the first set of square brackets denote the cur‐
3440 rent status of each thread. The first character is the first job de‐
3441 fined in the job file, and so forth. The possible values (in typical
3442 life cycle order) are:
3443 P Thread setup, but not started.
3445 C Thread created.
3446 I Thread initialized, waiting or generating necessary data.
3447 p Thread running pre-reading file(s).
3448 / Thread is in ramp period.
3449 R Running, doing sequential reads.
3450 r Running, doing random reads.
3451 W Running, doing sequential writes.
3452 w Running, doing random writes.
3453 M Running, doing mixed sequential reads/writes.
3454 m Running, doing mixed random reads/writes.
3455 D Running, doing sequential trims.
3456 d Running, doing random trims.
3457 F Running, currently waiting for fsync(2).
3458 V Running, doing verification of written data.
3459 f Thread finishing.
3460 E Thread exited, not reaped by main thread yet.
3461 - Thread reaped.
3462 X Thread reaped, exited with an error.
3463 K Thread reaped, exited due to signal.
3464 Fio will condense the thread string as not to take up more space on the
3466 command line than needed. For instance, if you have 10 readers and 10
3467 writers running, the output would look like this:
3468 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]
3470 Note that the status string is displayed in order, so it's possible to
3472 tell which of the jobs are currently doing what. In the example above
3473 this means that jobs 1--10 are readers and 11--20 are writers.
3474 The other values are fairly self explanatory -- number of threads cur‐
3476 rently running and doing I/O, the number of currently open files (f=),
3477 the estimated completion percentage, the rate of I/O since last check
3478 (read speed listed first, then write speed and optionally trim speed)
3479 in terms of bandwidth and IOPS, and time to completion for the current
3480 running group. It's impossible to estimate runtime of the following
3481 groups (if any).
3482 When fio is done (or interrupted by Ctrl-C), it will show the data for
3484 each thread, group of threads, and disks in that order. For each over‐
3485 all thread (or group) the output looks like:
3486 Client1: (groupid=0, jobs=1): err= 0: pid=16109: Sat Jun 24 12:07:54 2017
3488 write: IOPS=88, BW=623KiB/s (638kB/s)(30.4MiB/50032msec)
3489 slat (nsec): min=500, max=145500, avg=8318.00, stdev=4781.50
3490 clat (usec): min=170, max=78367, avg=4019.02, stdev=8293.31
3491 lat (usec): min=174, max=78375, avg=4027.34, stdev=8291.79
3492 clat percentiles (usec):
3493 | 1.00th=[ 302], 5.00th=[ 326], 10.00th=[ 343], 20.00th=[ 363],
3494 | 30.00th=[ 392], 40.00th=[ 404], 50.00th=[ 416], 60.00th=[ 445],
3495 | 70.00th=[ 816], 80.00th=[ 6718], 90.00th=[12911], 95.00th=[21627],
3496 | 99.00th=[43779], 99.50th=[51643], 99.90th=[68682], 99.95th=[72877],
3497 | 99.99th=[78119]
3498 bw ( KiB/s): min= 532, max= 686, per=0.10%, avg=622.87, stdev=24.82, samples= 100
3499 iops : min= 76, max= 98, avg=88.98, stdev= 3.54, samples= 100
3500 lat (usec) : 250=0.04%, 500=64.11%, 750=4.81%, 1000=2.79%
3501 lat (msec) : 2=4.16%, 4=1.84%, 10=4.90%, 20=11.33%, 50=5.37%
3502 lat (msec) : 100=0.65%
3503 cpu : usr=0.27%, sys=0.18%, ctx=12072, majf=0, minf=21
3504 IO depths : 1=85.0%, 2=13.1%, 4=1.8%, 8=0.1%, 16=0.0%, 32=0.0%, >=64=0.0%
3505 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3506 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3507 issued rwt: total=0,4450,0, short=0,0,0, dropped=0,0,0
3508 latency : target=0, window=0, percentile=100.00%, depth=8
3509 The job name (or first job's name when using group_reporting) is
3511 printed, along with the group id, count of jobs being aggregated, last
3512 error id seen (which is 0 when there are no errors), pid/tid of that
3513 thread and the time the job/group completed. Below are the I/O statis‐
3514 tics for each data direction performed (showing writes in the example
3515 above). In the order listed, they denote:
3516 read/write/trim
3518 The string before the colon shows the I/O direction the
3519 statistics are for. IOPS is the average I/Os performed
3520 per second. BW is the average bandwidth rate shown as:
3521 value in power of 2 format (value in power of 10 format).
3522 The last two values show: (total I/O performed in power
3523 of 2 format / runtime of that thread).
3524 slat Submission latency (min being the minimum, max being the
3526 maximum, avg being the average, stdev being the standard
3527 deviation). This is the time it took to submit the I/O.
3528 For sync I/O this row is not displayed as the slat is re‐
3529 ally the completion latency (since queue/complete is one
3530 operation there). This value can be in nanoseconds, mi‐
3531 croseconds or milliseconds --- fio will choose the most
3532 appropriate base and print that (in the example above
3533 nanoseconds was the best scale). Note: in --minimal mode
3534 latencies are always expressed in microseconds.
3535 clat Completion latency. Same names as slat, this denotes the
3537 time from submission to completion of the I/O pieces. For
3538 sync I/O, clat will usually be equal (or very close) to
3539 0, as the time from submit to complete is basically just
3540 CPU time (I/O has already been done, see slat explana‐
3541 tion).
3542 lat Total latency. Same names as slat and clat, this denotes
3544 the time from when fio created the I/O unit to completion
3545 of the I/O operation.
3546 bw Bandwidth statistics based on samples. Same names as the
3548 xlat stats, but also includes the number of samples taken
3549 (samples) and an approximate percentage of total aggre‐
3550 gate bandwidth this thread received in its group (per).
3551 This last value is only really useful if the threads in
3552 this group are on the same disk, since they are then com‐
3553 peting for disk access.
3554 iops IOPS statistics based on samples. Same names as bw.
3556 lat (nsec/usec/msec)
3558 The distribution of I/O completion latencies. This is the
3559 time from when I/O leaves fio and when it gets completed.
3560 Unlike the separate read/write/trim sections above, the
3561 data here and in the remaining sections apply to all I/Os
3562 for the reporting group. 250=0.04% means that 0.04% of
3563 the I/Os completed in under 250us. 500=64.11% means that
3564 64.11% of the I/Os required 250 to 499us for completion.
3565 cpu CPU usage. User and system time, along with the number of
3567 context switches this thread went through, usage of sys‐
3568 tem and user time, and finally the number of major and
3569 minor page faults. The CPU utilization numbers are aver‐
3570 ages for the jobs in that reporting group, while the con‐
3571 text and fault counters are summed.
3572 IO depths
3574 The distribution of I/O depths over the job lifetime. The
3575 numbers are divided into powers of 2 and each entry cov‐
3576 ers depths from that value up to those that are lower
3577 than the next entry -- e.g., 16= covers depths from 16 to
3578 31. Note that the range covered by a depth distribution
3579 entry can be different to the range covered by the equiv‐
3580 alent submit/complete distribution entry.
3581 IO submit
3583 How many pieces of I/O were submitting in a single submit
3584 call. Each entry denotes that amount and below, until the
3585 previous entry -- e.g., 16=100% means that we submitted
3586 anywhere between 9 to 16 I/Os per submit call. Note that
3587 the range covered by a submit distribution entry can be
3588 different to the range covered by the equivalent depth
3589 distribution entry.
3590 IO complete
3592 Like the above submit number, but for completions in‐
3593 stead.
3594 IO issued rwt
3596 The number of read/write/trim requests issued, and how
3597 many of them were short or dropped.
3598 IO latency
3600 These values are for latency_target and related options.
3601 When these options are engaged, this section describes
3602 the I/O depth required to meet the specified latency tar‐
3603 get.
3604 After each client has been listed, the group statistics are printed.
3606 They will look like this:
3607 Run status group 0 (all jobs):
3609 READ: bw=20.9MiB/s (21.9MB/s), 10.4MiB/s-10.8MiB/s (10.9MB/s-11.3MB/s), io=64.0MiB (67.1MB), run=2973-3069msec
3610 WRITE: bw=1231KiB/s (1261kB/s), 616KiB/s-621KiB/s (630kB/s-636kB/s), io=64.0MiB (67.1MB), run=52747-53223msec
3611 For each data direction it prints:
3613 bw Aggregate bandwidth of threads in this group followed by
3615 the minimum and maximum bandwidth of all the threads in
3616 this group. Values outside of brackets are power-of-2
3617 format and those within are the equivalent value in a
3618 power-of-10 format.
3619 io Aggregate I/O performed of all threads in this group. The
3621 format is the same as bw.
3622 run The smallest and longest runtimes of the threads in this
3624 group.
3625 And finally, the disk statistics are printed. This is Linux specific.
3627 They will look like this:
3628 Disk stats (read/write):
3630 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
3631 Each value is printed for both reads and writes, with reads first. The
3633 numbers denote:
3634 ios Number of I/Os performed by all groups.
3636 merge Number of merges performed by the I/O scheduler.
3638 ticks Number of ticks we kept the disk busy.
3640 in_queue
3642 Total time spent in the disk queue.
3643 util The disk utilization. A value of 100% means we kept the
3645 disk busy constantly, 50% would be a disk idling half of
3646 the time.
3647 It is also possible to get fio to dump the current output while it is
3649 running, without terminating the job. To do that, send fio the USR1
3650 signal. You can also get regularly timed dumps by using the --sta‐
3651 tus-interval parameter, or by creating a file in `/tmp' named
3652 `fio-dump-status'. If fio sees this file, it will unlink it and dump
3653 the current output status.
3654
3656 For scripted usage where you typically want to generate tables or
3657 graphs of the results, fio can output the results in a semicolon sepa‐
3658 rated format. The format is one long line of values, such as:
3659 2;card0;0;0;7139336;121836;60004;1;10109;27.932460;116.933948;220;126861;3495.446807;1085.368601;226;126864;3523.635629;1089.012448;24063;99944;50.275485%;59818.274627;5540.657370;7155060;122104;60004;1;8338;29.086342;117.839068;388;128077;5032.488518;1234.785715;391;128085;5061.839412;1236.909129;23436;100928;50.287926%;59964.832030;5644.844189;14.595833%;19.394167%;123706;0;7313;0.1%;0.1%;0.1%;0.1%;0.1%;0.1%;100.0%;0.00%;0.00%;0.00%;0.00%;0.00%;0.00%;0.01%;0.02%;0.05%;0.16%;6.04%;40.40%;52.68%;0.64%;0.01%;0.00%;0.01%;0.00%;0.00%;0.00%;0.00%;0.00%
3661 A description of this job goes here.
3662 The job description (if provided) follows on a second line for terse
3664 v2. It appears on the same line for other terse versions.
3665 To enable terse output, use the --minimal or `--output-format=terse'
3667 command line options. The first value is the version of the terse out‐
3668 put format. If the output has to be changed for some reason, this num‐
3669 ber will be incremented by 1 to signify that change.
3670 Split up, the format is as follows (comments in brackets denote when a
3672 field was introduced or whether it's specific to some terse version):
3673 terse version, fio version [v3], jobname, groupid, error
3675 READ status:
3677 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3679 Submission latency: min, max, mean, stdev (usec)
3680 Completion latency: min, max, mean, stdev (usec)
3681 Completion latency percentiles: 20 fields (see below)
3682 Total latency: min, max, mean, stdev (usec)
3683 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3684 IOPS [v5]: min, max, mean, stdev, number of samples
3685 WRITE status:
3687 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3689 Submission latency: min, max, mean, stdev (usec)
3690 Completion latency: min, max, mean, stdev (usec)
3691 Completion latency percentiles: 20 fields (see below)
3692 Total latency: min, max, mean, stdev (usec)
3693 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3694 IOPS [v5]: min, max, mean, stdev, number of samples
3695 TRIM status [all but version 3]:
3697 Fields are similar to READ/WRITE status.
3699 CPU usage:
3701 user, system, context switches, major faults, minor faults
3703 I/O depths:
3705 <=1, 2, 4, 8, 16, 32, >=64
3707 I/O latencies microseconds:
3709 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
3711 I/O latencies milliseconds:
3713 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
3715 Disk utilization [v3]:
3717 disk name, read ios, write ios, read merges, write merges, read ticks, write ticks, time spent in queue, disk utilization percentage
3719 Additional Info (dependent on continue_on_error, default off):
3721 total # errors, first error code
3723 Additional Info (dependent on description being set):
3725 Text description
3727 Completion latency percentiles can be a grouping of up to 20 sets, so
3729 for the terse output fio writes all of them. Each field will look like
3730 this:
3731 1.00%=6112
3733 which is the Xth percentile, and the `usec' latency associated with it.
3735 For Disk utilization, all disks used by fio are shown. So for each disk
3737 there will be a disk utilization section.
3738 Below is a single line containing short names for each of the fields in
3740 the minimal output v3, separated by semicolons:
3741 terse_version_3;fio_version;jobname;groupid;error;read_kb;read_bandwidth_kb;read_iops;read_runtime_ms;read_slat_min_us;read_slat_max_us;read_slat_mean_us;read_slat_dev_us;read_clat_min_us;read_clat_max_us;read_clat_mean_us;read_clat_dev_us;read_clat_pct01;read_clat_pct02;read_clat_pct03;read_clat_pct04;read_clat_pct05;read_clat_pct06;read_clat_pct07;read_clat_pct08;read_clat_pct09;read_clat_pct10;read_clat_pct11;read_clat_pct12;read_clat_pct13;read_clat_pct14;read_clat_pct15;read_clat_pct16;read_clat_pct17;read_clat_pct18;read_clat_pct19;read_clat_pct20;read_tlat_min_us;read_lat_max_us;read_lat_mean_us;read_lat_dev_us;read_bw_min_kb;read_bw_max_kb;read_bw_agg_pct;read_bw_mean_kb;read_bw_dev_kb;write_kb;write_bandwidth_kb;write_iops;write_runtime_ms;write_slat_min_us;write_slat_max_us;write_slat_mean_us;write_slat_dev_us;write_clat_min_us;write_clat_max_us;write_clat_mean_us;write_clat_dev_us;write_clat_pct01;write_clat_pct02;write_clat_pct03;write_clat_pct04;write_clat_pct05;write_clat_pct06;write_clat_pct07;write_clat_pct08;write_clat_pct09;write_clat_pct10;write_clat_pct11;write_clat_pct12;write_clat_pct13;write_clat_pct14;write_clat_pct15;write_clat_pct16;write_clat_pct17;write_clat_pct18;write_clat_pct19;write_clat_pct20;write_tlat_min_us;write_lat_max_us;write_lat_mean_us;write_lat_dev_us;write_bw_min_kb;write_bw_max_kb;write_bw_agg_pct;write_bw_mean_kb;write_bw_dev_kb;cpu_user;cpu_sys;cpu_csw;cpu_mjf;cpu_minf;iodepth_1;iodepth_2;iodepth_4;iodepth_8;iodepth_16;iodepth_32;iodepth_64;lat_2us;lat_4us;lat_10us;lat_20us;lat_50us;lat_100us;lat_250us;lat_500us;lat_750us;lat_1000us;lat_2ms;lat_4ms;lat_10ms;lat_20ms;lat_50ms;lat_100ms;lat_250ms;lat_500ms;lat_750ms;lat_1000ms;lat_2000ms;lat_over_2000ms;disk_name;disk_read_iops;disk_write_iops;disk_read_merges;disk_write_merges;disk_read_ticks;write_ticks;disk_queue_time;disk_util
3743 In client/server mode terse output differs from what appears when jobs
3745 are run locally. Disk utilization data is omitted from the standard
3746 terse output and for v3 and later appears on its own separate line at
3747 the end of each terse reporting cycle.
3748
3750 The json output format is intended to be both human readable and conve‐
3751 nient for automated parsing. For the most part its sections mirror
3752 those of the normal output. The runtime value is reported in msec and
3753 the bw value is reported in 1024 bytes per second units.
3754
3756 The json+ output format is identical to the json output format except
3757 that it adds a full dump of the completion latency bins. Each bins ob‐
3758 ject contains a set of (key, value) pairs where keys are latency dura‐
3759 tions and values count how many I/Os had completion latencies of the
3760 corresponding duration. For example, consider:
3761 "bins" : { "87552" : 1, "89600" : 1, "94720" : 1, "96768" : 1,
3763 "97792" : 1, "99840" : 1, "100864" : 2, "103936" : 6, "104960" :
3764 534, "105984" : 5995, "107008" : 7529, ... }
3765 This data indicates that one I/O required 87,552ns to complete, two
3767 I/Os required 100,864ns to complete, and 7529 I/Os required 107,008ns
3768 to complete.
3769 Also included with fio is a Python script fio_jsonplus_clat2csv that
3771 takes json+ output and generates CSV-formatted latency data suitable
3772 for plotting.
3773 The latency durations actually represent the midpoints of latency in‐
3775 tervals. For details refer to `stat.h' in the fio source.
3776
3778 There are two trace file format that you can encounter. The older (v1)
3779 format is unsupported since version 1.20-rc3 (March 2008). It will
3780 still be described below in case that you get an old trace and want to
3781 understand it.
3782 In any case the trace is a simple text file with a single action per
3784 line.
3785 Trace file format v1
3787 Each line represents a single I/O action in the following for‐
3788 mat:
3789 rw, offset, length
3791 where `rw=0/1' for read/write, and the `offset' and `length' en‐
3793 tries being in bytes.
3794 This format is not supported in fio versions >= 1.20-rc3.
3796 Trace file format v2
3798 The second version of the trace file format was added in fio
3799 version 1.17. It allows to access more then one file per trace
3800 and has a bigger set of possible file actions.
3801 The first line of the trace file has to be:
3803 "fio version 2 iolog"
3805 Following this can be lines in two different formats, which are
3807 described below.
3808 The file management format:
3810 filename action
3811 The `filename' is given as an absolute path. The `action'
3813 can be one of these:
3814 add Add the given `filename' to the trace.
3816 open Open the file with the given `filename'.
3818 The `filename' has to have been added with
3819 the add action before.
3820 close Close the file with the given `filename'.
3822 The file has to have been opened before.
3823 The file I/O action format:
3825 filename action offset length
3826 The `filename' is given as an absolute path, and has to
3828 have been added and opened before it can be used with
3829 this format. The `offset' and `length' are given in
3830 bytes. The `action' can be one of these:
3831 wait Wait for `offset' microseconds. Everything
3833 below 100 is discarded. The time is rela‐
3834 tive to the previous `wait' statement.
3835 read Read `length' bytes beginning from `off‐
3837 set'.
3838 write Write `length' bytes beginning from `off‐
3840 set'.
3841 sync fsync(2) the file.
3843 datasync
3845 fdatasync(2) the file.
3846 trim Trim the given file from the given `offset'
3848 for `length' bytes.
3849
3851 Colocation is a common practice used to get the most out of a machine.
3852 Knowing which workloads play nicely with each other and which ones
3853 don't is a much harder task. While fio can replay workloads concur‐
3854 rently via multiple jobs, it leaves some variability up to the sched‐
3855 uler making results harder to reproduce. Merging is a way to make the
3856 order of events consistent.
3857 Merging is integrated into I/O replay and done when a merge_blk‐
3859 trace_file is specified. The list of files passed to read_iolog go
3860 through the merge process and output a single file stored to the speci‐
3861 fied file. The output file is passed on as if it were the only file
3862 passed to read_iolog. An example would look like:
3863 $ fio --read_iolog="<file1>:<file2>" --merge_blk‐
3865 trace_file="<output_file>"
3866 Creating only the merged file can be done by passing the command line
3868 argument merge-blktrace-only.
3869 Scaling traces can be done to see the relative impact of any particular
3871 trace being slowed down or sped up. merge_blktrace_scalars takes in a
3872 colon separated list of percentage scalars. It is index paired with the
3873 files passed to read_iolog.
3874 With scaling, it may be desirable to match the running time of all
3876 traces. This can be done with merge_blktrace_iters. It is index paired
3877 with read_iolog just like merge_blktrace_scalars.
3878 In an example, given two traces, A and B, each 60s long. If we want to
3880 see the impact of trace A issuing IOs twice as fast and repeat trace A
3881 over the runtime of trace B, the following can be done:
3882 $ fio --read_iolog="<trace_a>:"<trace_b>" --merge_blk‐
3884 trace_file"<output_file>" --merge_blktrace_scalars="50:100"
3885 --merge_blktrace_iters="2:1"
3886 This runs trace A at 2x the speed twice for approximately the same run‐
3888 time as a single run of trace B.
3889
3891 In some cases, we want to understand CPU overhead in a test. For exam‐
3892 ple, we test patches for the specific goodness of whether they reduce
3893 CPU usage. Fio implements a balloon approach to create a thread per
3894 CPU that runs at idle priority, meaning that it only runs when nobody
3895 else needs the cpu. By measuring the amount of work completed by the
3896 thread, idleness of each CPU can be derived accordingly.
3897 An unit work is defined as touching a full page of unsigned characters.
3899 Mean and standard deviation of time to complete an unit work is re‐
3900 ported in "unit work" section. Options can be chosen to report detailed
3901 percpu idleness or overall system idleness by aggregating percpu stats.
3902
3904 Fio is usually run in one of two ways, when data verification is done.
3905 The first is a normal write job of some sort with verify enabled. When
3906 the write phase has completed, fio switches to reads and verifies ev‐
3907 erything it wrote. The second model is running just the write phase,
3908 and then later on running the same job (but with reads instead of
3909 writes) to repeat the same I/O patterns and verify the contents. Both
3910 of these methods depend on the write phase being completed, as fio oth‐
3911 erwise has no idea how much data was written.
3912 With verification triggers, fio supports dumping the current write
3914 state to local files. Then a subsequent read verify workload can load
3915 this state and know exactly where to stop. This is useful for testing
3916 cases where power is cut to a server in a managed fashion, for in‐
3917 stance.
3918 A verification trigger consists of two things:
3920 1) Storing the write state of each job.
3922 2) Executing a trigger command.
3924 The write state is relatively small, on the order of hundreds of bytes
3926 to single kilobytes. It contains information on the number of comple‐
3927 tions done, the last X completions, etc.
3928 A trigger is invoked either through creation ('touch') of a specified
3930 file in the system, or through a timeout setting. If fio is run with
3931 `--trigger-file=/tmp/trigger-file', then it will continually check for
3932 the existence of `/tmp/trigger-file'. When it sees this file, it will
3933 fire off the trigger (thus saving state, and executing the trigger com‐
3934 mand).
3935 For client/server runs, there's both a local and remote trigger. If fio
3937 is running as a server backend, it will send the job states back to the
3938 client for safe storage, then execute the remote trigger, if specified.
3939 If a local trigger is specified, the server will still send back the
3940 write state, but the client will then execute the trigger.
3941 Verification trigger example
3943 Let's say we want to run a powercut test on the remote Linux ma‐
3944 chine 'server'. Our write workload is in `write-test.fio'. We
3945 want to cut power to 'server' at some point during the run, and
3946 we'll run this test from the safety or our local machine, 'lo‐
3947 calbox'. On the server, we'll start the fio backend normally:
3948 server# fio --server
3950 and on the client, we'll fire off the workload:
3952 localbox$ fio --client=server --trig‐
3954 ger-file=/tmp/my-trigger --trigger-remote="bash -c "echo
3955 b > /proc/sysrq-triger""
3956 We set `/tmp/my-trigger' as the trigger file, and we tell fio to
3958 execute:
3959 echo b > /proc/sysrq-trigger
3961 on the server once it has received the trigger and sent us the
3963 write state. This will work, but it's not really cutting power
3964 to the server, it's merely abruptly rebooting it. If we have a
3965 remote way of cutting power to the server through IPMI or simi‐
3966 lar, we could do that through a local trigger command instead.
3967 Let's assume we have a script that does IPMI reboot of a given
3968 hostname, ipmi-reboot. On localbox, we could then have run fio
3969 with a local trigger instead:
3970 localbox$ fio --client=server --trig‐
3972 ger-file=/tmp/my-trigger --trigger="ipmi-reboot server"
3973 For this case, fio would wait for the server to send us the
3975 write state, then execute `ipmi-reboot server' when that hap‐
3976 pened.
3977 Loading verify state
3979 To load stored write state, a read verification job file must
3980 contain the verify_state_load option. If that is set, fio will
3981 load the previously stored state. For a local fio run this is
3982 done by loading the files directly, and on a client/server run,
3983 the server backend will ask the client to send the files over
3984 and load them from there.
3985
3987 Fio supports a variety of log file formats, for logging latencies,
3988 bandwidth, and IOPS. The logs share a common format, which looks like
3989 this:
3990 time (msec), value, data direction, block size (bytes), offset
3992 (bytes), command priority
3993 `Time' for the log entry is always in milliseconds. The `value' logged
3995 depends on the type of log, it will be one of the following:
3996 Latency log
3998 Value is latency in nsecs
3999 Bandwidth log
4001 Value is in KiB/sec
4002 IOPS log
4004 Value is IOPS
4005 `Data direction' is one of the following:
4007 0 I/O is a READ
4009 1 I/O is a WRITE
4011 2 I/O is a TRIM
4013 The entry's `block size' is always in bytes. The `offset' is the posi‐
4015 tion in bytes from the start of the file for that particular I/O. The
4016 logging of the offset can be toggled with log_offset.
4017 If log_prio is not set, the entry's `Command priority` is 1 for an IO
4019 executed with the highest RT priority class (prioclass=1 or cmd‐
4020 prio_class=1) and 0 otherwise. This is controlled by the prioclass op‐
4021 tion and the ioengine specific cmdprio_percentage cmdprio_class op‐
4022 tions. If log_prio is set, the entry's `Command priority` is the prior‐
4023 ity set for the IO, as a 16-bits hexadecimal number with the lowest 13
4024 bits indicating the priority value (prio and cmdprio options) and the
4025 highest 3 bits indicating the IO priority class (prioclass and cmd‐
4026 prio_class options).
4027 Fio defaults to logging every individual I/O but when windowed logging
4029 is set through log_avg_msec, either the average (by default) or the
4030 maximum (log_max_value is set) `value' seen over the specified period
4031 of time is recorded. Each `data direction' seen within the window pe‐
4032 riod will aggregate its values in a separate row. Further, when using
4033 windowed logging the `block size' and `offset' entries will always con‐
4034 tain 0.
4035
4037 Normally fio is invoked as a stand-alone application on the machine
4038 where the I/O workload should be generated. However, the backend and
4039 frontend of fio can be run separately i.e., the fio server can generate
4040 an I/O workload on the "Device Under Test" while being controlled by a
4041 client on another machine.
4042 Start the server on the machine which has access to the storage DUT:
4044 $ fio --server=args
4046 where `args' defines what fio listens to. The arguments are of the form
4048 `type,hostname' or `IP,port'. `type' is either `ip' (or ip4) for TCP/IP
4049 v4, `ip6' for TCP/IP v6, or `sock' for a local unix domain socket.
4050 `hostname' is either a hostname or IP address, and `port' is the port
4051 to listen to (only valid for TCP/IP, not a local socket). Some exam‐
4052 ples:
4053 1) fio --server
4055 Start a fio server, listening on all interfaces on the
4056 default port (8765).
4057 2) fio --server=ip:hostname,4444
4059 Start a fio server, listening on IP belonging to hostname
4060 and on port 4444.
4061 3) fio --server=ip6:::1,4444
4063 Start a fio server, listening on IPv6 localhost ::1 and
4064 on port 4444.
4065 4) fio --server=,4444
4067 Start a fio server, listening on all interfaces on port
4068 4444.
4069 5) fio --server=1.2.3.4
4071 Start a fio server, listening on IP 1.2.3.4 on the de‐
4072 fault port.
4073 6) fio --server=sock:/tmp/fio.sock
4075 Start a fio server, listening on the local socket
4076 `/tmp/fio.sock'.
4077 Once a server is running, a "client" can connect to the fio server
4079 with:
4080 $ fio <local-args> --client=<server> <remote-args> <job file(s)>
4082 where `local-args' are arguments for the client where it is running,
4084 `server' is the connect string, and `remote-args' and `job file(s)' are
4085 sent to the server. The `server' string follows the same format as it
4086 does on the server side, to allow IP/hostname/socket and port strings.
4087 Fio can connect to multiple servers this way:
4089 $ fio --client=<server1> <job file(s)> --client=<server2> <job
4091 file(s)>
4092 If the job file is located on the fio server, then you can tell the
4094 server to load a local file as well. This is done by using --re‐
4095 mote-config:
4096 $ fio --client=server --remote-config /path/to/file.fio
4098 Then fio will open this local (to the server) job file instead of being
4100 passed one from the client.
4101 If you have many servers (example: 100 VMs/containers), you can input a
4103 pathname of a file containing host IPs/names as the parameter value for
4104 the --client option. For example, here is an example `host.list' file
4105 containing 2 hostnames:
4106 host1.your.dns.domain
4108 host2.your.dns.domain
4109 The fio command would then be:
4111 $ fio --client=host.list <job file(s)>
4113 In this mode, you cannot input server-specific parameters or job files
4115 -- all servers receive the same job file.
4116 In order to let `fio --client' runs use a shared filesystem from multi‐
4118 ple hosts, `fio --client' now prepends the IP address of the server to
4119 the filename. For example, if fio is using the directory `/mnt/nfs/fio'
4120 and is writing filename `fileio.tmp', with a --client `hostfile' con‐
4121 taining two hostnames `h1' and `h2' with IP addresses 192.168.10.120
4122 and 192.168.10.121, then fio will create two files:
4123 /mnt/nfs/fio/192.168.10.120.fileio.tmp
4125 /mnt/nfs/fio/192.168.10.121.fileio.tmp
4126 Terse output in client/server mode will differ slightly from what is
4128 produced when fio is run in stand-alone mode. See the terse output sec‐
4129 tion for details.
4130
4132 fio was written by Jens Axboe <axboe@kernel.dk>.
4133 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au>
4134 based on documentation by Jens Axboe.
4135 This man page was rewritten by Tomohiro Kusumi <tkusumi@tuxera.com>
4136 based on documentation by Jens Axboe.
4137
4139 Report bugs to the fio mailing list <fio@vger.kernel.org>.
4140 See REPORTING-BUGS.
4141 REPORTING-BUGS: http://git.kernel.dk/cgit/fio/plain/REPORTING-BUGS
4143
4145 For further documentation see HOWTO and README.
4146 Sample jobfiles are available in the `examples/' directory.
4147 These are typically located under `/usr/share/doc/fio'.
4148 HOWTO: http://git.kernel.dk/cgit/fio/plain/HOWTO
4150 README: http://git.kernel.dk/cgit/fio/plain/README
4151User Manual August 2017 fio(1)