4 The first step in getting fio to simulate a desired I/O workload, is writing a
5 job file describing that specific setup. A job file may contain any number of
6 threads and/or files -- the typical contents of the job file is a *global*
7 section defining shared parameters, and one or more job sections describing the
8 jobs involved. When run, fio parses this file and sets everything up as
9 described. If we break down a job from top to bottom, it contains the following
14 Defines the I/O pattern issued to the file(s). We may only be reading
15 sequentially from this file(s), or we may be writing randomly. Or even
16 mixing reads and writes, sequentially or randomly.
17 Should we be doing buffered I/O, or direct/raw I/O?
21 In how large chunks are we issuing I/O? This may be a single value,
22 or it may describe a range of block sizes.
26 How much data are we going to be reading/writing.
30 How do we issue I/O? We could be memory mapping the file, we could be
31 using regular read/write, we could be using splice, async I/O, or even
36 If the I/O engine is async, how large a queuing depth do we want to
42 How many files are we spreading the workload over.
44 `Threads, processes and job synchronization`_
46 How many threads or processes should we spread this workload over.
48 The above are the basic parameters defined for a workload, in addition there's a
49 multitude of parameters that modify other aspects of how this job behaves.
55 .. option:: --debug=type
57 Enable verbose tracing `type` of various fio actions. May be ``all`` for all types
58 or individual types separated by a comma (e.g. ``--debug=file,mem`` will
59 enable file and memory debugging). Currently, additional logging is
63 Dump info related to processes.
65 Dump info related to file actions.
67 Dump info related to I/O queuing.
69 Dump info related to memory allocations.
71 Dump info related to blktrace setup.
73 Dump info related to I/O verification.
75 Enable all debug options.
77 Dump info related to random offset generation.
79 Dump info related to option matching and parsing.
81 Dump info related to disk utilization updates.
83 Dump info only related to job number x.
85 Dump info only related to mutex up/down ops.
87 Dump info related to profile extensions.
89 Dump info related to internal time keeping.
91 Dump info related to networking connections.
93 Dump info related to I/O rate switching.
95 Dump info related to log compress/decompress.
97 Dump info related to steadystate detection.
99 Dump info related to the helper thread.
101 Dump info related to support for zoned block devices.
103 Show available debug options.
105 .. option:: --parse-only
107 Parse options only, don't start any I/O.
109 .. option:: --merge-blktrace-only
111 Merge blktraces only, don't start any I/O.
113 .. option:: --output=filename
115 Write output to file `filename`.
117 .. option:: --output-format=format
119 Set the reporting `format` to `normal`, `terse`, `json`, or `json+`. Multiple
120 formats can be selected, separated by a comma. `terse` is a CSV based
121 format. `json+` is like `json`, except it adds a full dump of the latency
124 .. option:: --bandwidth-log
126 Generate aggregate bandwidth logs.
128 .. option:: --minimal
130 Print statistics in a terse, semicolon-delimited format.
132 .. option:: --append-terse
134 Print statistics in selected mode AND terse, semicolon-delimited format.
135 **Deprecated**, use :option:`--output-format` instead to select multiple
138 .. option:: --terse-version=version
140 Set terse `version` output format (default 3, or 2 or 4 or 5).
142 .. option:: --version
144 Print version information and exit.
148 Print a summary of the command line options and exit.
150 .. option:: --cpuclock-test
152 Perform test and validation of internal CPU clock.
154 .. option:: --crctest=[test]
156 Test the speed of the built-in checksumming functions. If no argument is
157 given, all of them are tested. Alternatively, a comma separated list can
158 be passed, in which case the given ones are tested.
160 .. option:: --cmdhelp=command
162 Print help information for `command`. May be ``all`` for all commands.
164 .. option:: --enghelp=[ioengine[,command]]
166 List all commands defined by `ioengine`, or print help for `command`
167 defined by `ioengine`. If no `ioengine` is given, list all
170 .. option:: --showcmd=jobfile
172 Convert `jobfile` to a set of command-line options.
174 .. option:: --readonly
176 Turn on safety read-only checks, preventing writes and trims. The
177 ``--readonly`` option is an extra safety guard to prevent users from
178 accidentally starting a write or trim workload when that is not desired.
179 Fio will only modify the device under test if
180 `rw=write/randwrite/rw/randrw/trim/randtrim/trimwrite` is given. This
181 safety net can be used as an extra precaution.
183 .. option:: --eta=when
185 Specifies when real-time ETA estimate should be printed. `when` may be
186 `always`, `never` or `auto`. `auto` is the default, it prints ETA
187 when requested if the output is a TTY. `always` disregards the output
188 type, and prints ETA when requested. `never` never prints ETA.
190 .. option:: --eta-interval=time
192 By default, fio requests client ETA status roughly every second. With
193 this option, the interval is configurable. Fio imposes a minimum
194 allowed time to avoid flooding the console, less than 250 msec is
197 .. option:: --eta-newline=time
199 Force a new line for every `time` period passed. When the unit is omitted,
200 the value is interpreted in seconds.
202 .. option:: --status-interval=time
204 Force a full status dump of cumulative (from job start) values at `time`
205 intervals. This option does *not* provide per-period measurements. So
206 values such as bandwidth are running averages. When the time unit is omitted,
207 `time` is interpreted in seconds. Note that using this option with
208 ``--output-format=json`` will yield output that technically isn't valid
209 json, since the output will be collated sets of valid json. It will need
210 to be split into valid sets of json after the run.
212 .. option:: --section=name
214 Only run specified section `name` in job file. Multiple sections can be specified.
215 The ``--section`` option allows one to combine related jobs into one file.
216 E.g. one job file could define light, moderate, and heavy sections. Tell
217 fio to run only the "heavy" section by giving ``--section=heavy``
218 command line option. One can also specify the "write" operations in one
219 section and "verify" operation in another section. The ``--section`` option
220 only applies to job sections. The reserved *global* section is always
223 .. option:: --alloc-size=kb
225 Allocate additional internal smalloc pools of size `kb` in KiB. The
226 ``--alloc-size`` option increases shared memory set aside for use by fio.
227 If running large jobs with randommap enabled, fio can run out of memory.
228 Smalloc is an internal allocator for shared structures from a fixed size
229 memory pool and can grow to 16 pools. The pool size defaults to 16MiB.
231 NOTE: While running :file:`.fio_smalloc.*` backing store files are visible
234 .. option:: --warnings-fatal
236 All fio parser warnings are fatal, causing fio to exit with an
239 .. option:: --max-jobs=nr
241 Set the maximum number of threads/processes to support to `nr`.
242 NOTE: On Linux, it may be necessary to increase the shared-memory
243 limit (:file:`/proc/sys/kernel/shmmax`) if fio runs into errors while
246 .. option:: --server=args
248 Start a backend server, with `args` specifying what to listen to.
249 See `Client/Server`_ section.
251 .. option:: --daemonize=pidfile
253 Background a fio server, writing the pid to the given `pidfile` file.
255 .. option:: --client=hostname
257 Instead of running the jobs locally, send and run them on the given `hostname`
258 or set of `hostname`\s. See `Client/Server`_ section.
260 .. option:: --remote-config=file
262 Tell fio server to load this local `file`.
264 .. option:: --idle-prof=option
266 Report CPU idleness. `option` is one of the following:
269 Run unit work calibration only and exit.
272 Show aggregate system idleness and unit work.
275 As **system** but also show per CPU idleness.
277 .. option:: --inflate-log=log
279 Inflate and output compressed `log`.
281 .. option:: --trigger-file=file
283 Execute trigger command when `file` exists.
285 .. option:: --trigger-timeout=time
287 Execute trigger at this `time`.
289 .. option:: --trigger=command
291 Set this `command` as local trigger.
293 .. option:: --trigger-remote=command
295 Set this `command` as remote trigger.
297 .. option:: --aux-path=path
299 Use the directory specified by `path` for generated state files instead
300 of the current working directory.
302 Any parameters following the options will be assumed to be job files, unless
303 they match a job file parameter. Multiple job files can be listed and each job
304 file will be regarded as a separate group. Fio will :option:`stonewall`
305 execution between each group.
311 As previously described, fio accepts one or more job files describing what it is
312 supposed to do. The job file format is the classic ini file, where the names
313 enclosed in [] brackets define the job name. You are free to use any ASCII name
314 you want, except *global* which has special meaning. Following the job name is
315 a sequence of zero or more parameters, one per line, that define the behavior of
316 the job. If the first character in a line is a ';' or a '#', the entire line is
317 discarded as a comment.
319 A *global* section sets defaults for the jobs described in that file. A job may
320 override a *global* section parameter, and a job file may even have several
321 *global* sections if so desired. A job is only affected by a *global* section
324 The :option:`--cmdhelp` option also lists all options. If used with a `command`
325 argument, :option:`--cmdhelp` will detail the given `command`.
327 See the `examples/` directory for inspiration on how to write job files. Note
328 the copyright and license requirements currently apply to `examples/` files.
330 So let's look at a really simple job file that defines two processes, each
331 randomly reading from a 128MiB file:
335 ; -- start job file --
346 As you can see, the job file sections themselves are empty as all the described
347 parameters are shared. As no :option:`filename` option is given, fio makes up a
348 `filename` for each of the jobs as it sees fit. On the command line, this job
349 would look as follows::
351 $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
354 Let's look at an example that has a number of processes writing randomly to
359 ; -- start job file --
370 Here we have no *global* section, as we only have one job defined anyway. We
371 want to use async I/O here, with a depth of 4 for each file. We also increased
372 the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical
373 jobs. The result is 4 processes each randomly writing to their own 64MiB
374 file. Instead of using the above job file, you could have given the parameters
375 on the command line. For this case, you would specify::
377 $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
379 When fio is utilized as a basis of any reasonably large test suite, it might be
380 desirable to share a set of standardized settings across multiple job files.
381 Instead of copy/pasting such settings, any section may pull in an external
382 :file:`filename.fio` file with *include filename* directive, as in the following
385 ; -- start job file including.fio --
389 include glob-include.fio
396 include test-include.fio
397 ; -- end job file including.fio --
401 ; -- start job file glob-include.fio --
404 ; -- end job file glob-include.fio --
408 ; -- start job file test-include.fio --
411 ; -- end job file test-include.fio --
413 Settings pulled into a section apply to that section only (except *global*
414 section). Include directives may be nested in that any included file may contain
415 further include directive(s). Include files may not contain [] sections.
418 Environment variables
419 ~~~~~~~~~~~~~~~~~~~~~
421 Fio also supports environment variable expansion in job files. Any sub-string of
422 the form ``${VARNAME}`` as part of an option value (in other words, on the right
423 of the '='), will be expanded to the value of the environment variable called
424 `VARNAME`. If no such environment variable is defined, or `VARNAME` is the
425 empty string, the empty string will be substituted.
427 As an example, let's look at a sample fio invocation and job file::
429 $ SIZE=64m NUMJOBS=4 fio jobfile.fio
433 ; -- start job file --
440 This will expand to the following equivalent job file at runtime:
444 ; -- start job file --
451 Fio ships with a few example job files, you can also look there for inspiration.
456 Additionally, fio has a set of reserved keywords that will be replaced
457 internally with the appropriate value. Those keywords are:
461 The architecture page size of the running system.
465 Megabytes of total memory in the system.
469 Number of online available CPUs.
471 These can be used on the command line or in the job file, and will be
472 automatically substituted with the current system values when the job is
473 run. Simple math is also supported on these keywords, so you can perform actions
478 and get that properly expanded to 8 times the size of memory in the machine.
484 This section describes in details each parameter associated with a job. Some
485 parameters take an option of a given type, such as an integer or a
486 string. Anywhere a numeric value is required, an arithmetic expression may be
487 used, provided it is surrounded by parentheses. Supported operators are:
496 For time values in expressions, units are microseconds by default. This is
497 different than for time values not in expressions (not enclosed in
498 parentheses). The following types are used:
505 String: A sequence of alphanumeric characters.
508 Integer with possible time suffix. Without a unit value is interpreted as
509 seconds unless otherwise specified. Accepts a suffix of 'd' for days, 'h' for
510 hours, 'm' for minutes, 's' for seconds, 'ms' (or 'msec') for milliseconds and
511 'us' (or 'usec') for microseconds. For example, use 10m for 10 minutes.
516 Integer. A whole number value, which may contain an integer prefix
517 and an integer suffix:
519 [*integer prefix*] **number** [*integer suffix*]
521 The optional *integer prefix* specifies the number's base. The default
522 is decimal. *0x* specifies hexadecimal.
524 The optional *integer suffix* specifies the number's units, and includes an
525 optional unit prefix and an optional unit. For quantities of data, the
526 default unit is bytes. For quantities of time, the default unit is seconds
527 unless otherwise specified.
529 With :option:`kb_base`\=1000, fio follows international standards for unit
530 prefixes. To specify power-of-10 decimal values defined in the
531 International System of Units (SI):
533 * *K* -- means kilo (K) or 1000
534 * *M* -- means mega (M) or 1000**2
535 * *G* -- means giga (G) or 1000**3
536 * *T* -- means tera (T) or 1000**4
537 * *P* -- means peta (P) or 1000**5
539 To specify power-of-2 binary values defined in IEC 80000-13:
541 * *Ki* -- means kibi (Ki) or 1024
542 * *Mi* -- means mebi (Mi) or 1024**2
543 * *Gi* -- means gibi (Gi) or 1024**3
544 * *Ti* -- means tebi (Ti) or 1024**4
545 * *Pi* -- means pebi (Pi) or 1024**5
547 With :option:`kb_base`\=1024 (the default), the unit prefixes are opposite
548 from those specified in the SI and IEC 80000-13 standards to provide
549 compatibility with old scripts. For example, 4k means 4096.
551 For quantities of data, an optional unit of 'B' may be included
552 (e.g., 'kB' is the same as 'k').
554 The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega,
555 not milli). 'b' and 'B' both mean byte, not bit.
557 Examples with :option:`kb_base`\=1000:
559 * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB
560 * *1 MiB*: 1048576, 1mi, 1024ki
561 * *1 MB*: 1000000, 1m, 1000k
562 * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi
563 * *1 TB*: 1000000000, 1t, 1000m, 1000000k
565 Examples with :option:`kb_base`\=1024 (default):
567 * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
568 * *1 MiB*: 1048576, 1m, 1024k
569 * *1 MB*: 1000000, 1mi, 1000ki
570 * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m
571 * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki
573 To specify times (units are not case sensitive):
577 * *M* -- means minutes
578 * *s* -- or sec means seconds (default)
579 * *ms* -- or *msec* means milliseconds
580 * *us* -- or *usec* means microseconds
582 If the option accepts an upper and lower range, use a colon ':' or
583 minus '-' to separate such values. See :ref:`irange <irange>`.
584 If the lower value specified happens to be larger than the upper value
585 the two values are swapped.
590 Boolean. Usually parsed as an integer, however only defined for
591 true and false (1 and 0).
596 Integer range with suffix. Allows value range to be given, such as
597 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the
598 option allows two sets of ranges, they can be specified with a ',' or '/'
599 delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`.
602 A list of floating point numbers, separated by a ':' character.
604 With the above in mind, here follows the complete list of fio job parameters.
610 .. option:: kb_base=int
612 Select the interpretation of unit prefixes in input parameters.
615 Inputs comply with IEC 80000-13 and the International
616 System of Units (SI). Use:
618 - power-of-2 values with IEC prefixes (e.g., KiB)
619 - power-of-10 values with SI prefixes (e.g., kB)
622 Compatibility mode (default). To avoid breaking old scripts:
624 - power-of-2 values with SI prefixes
625 - power-of-10 values with IEC prefixes
627 See :option:`bs` for more details on input parameters.
629 Outputs always use correct prefixes. Most outputs include both
632 bw=2383.3kB/s (2327.4KiB/s)
634 If only one value is reported, then kb_base selects the one to use:
636 **1000** -- SI prefixes
638 **1024** -- IEC prefixes
640 .. option:: unit_base=int
642 Base unit for reporting. Allowed values are:
645 Use auto-detection (default).
657 ASCII name of the job. This may be used to override the name printed by fio
658 for this job. Otherwise the job name is used. On the command line this
659 parameter has the special purpose of also signaling the start of a new job.
661 .. option:: description=str
663 Text description of the job. Doesn't do anything except dump this text
664 description when this job is run. It's not parsed.
666 .. option:: loops=int
668 Run the specified number of iterations of this job. Used to repeat the same
669 workload a given number of times. Defaults to 1.
671 .. option:: numjobs=int
673 Create the specified number of clones of this job. Each clone of job
674 is spawned as an independent thread or process. May be used to setup a
675 larger number of threads/processes doing the same thing. Each thread is
676 reported separately; to see statistics for all clones as a whole, use
677 :option:`group_reporting` in conjunction with :option:`new_group`.
678 See :option:`--max-jobs`. Default: 1.
681 Time related parameters
682 ~~~~~~~~~~~~~~~~~~~~~~~
684 .. option:: runtime=time
686 Tell fio to terminate processing after the specified period of time. It
687 can be quite hard to determine for how long a specified job will run, so
688 this parameter is handy to cap the total runtime to a given time. When
689 the unit is omitted, the value is interpreted in seconds.
691 .. option:: time_based
693 If set, fio will run for the duration of the :option:`runtime` specified
694 even if the file(s) are completely read or written. It will simply loop over
695 the same workload as many times as the :option:`runtime` allows.
697 .. option:: startdelay=irange(time)
699 Delay the start of job for the specified amount of time. Can be a single
700 value or a range. When given as a range, each thread will choose a value
701 randomly from within the range. Value is in seconds if a unit is omitted.
703 .. option:: ramp_time=time
705 If set, fio will run the specified workload for this amount of time before
706 logging any performance numbers. Useful for letting performance settle
707 before logging results, thus minimizing the runtime required for stable
708 results. Note that the ``ramp_time`` is considered lead in time for a job,
709 thus it will increase the total runtime if a special timeout or
710 :option:`runtime` is specified. When the unit is omitted, the value is
713 .. option:: clocksource=str
715 Use the given clocksource as the base of timing. The supported options are:
718 :manpage:`gettimeofday(2)`
721 :manpage:`clock_gettime(2)`
724 Internal CPU clock source
726 cpu is the preferred clocksource if it is reliable, as it is very fast (and
727 fio is heavy on time calls). Fio will automatically use this clocksource if
728 it's supported and considered reliable on the system it is running on,
729 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
730 means supporting TSC Invariant.
732 .. option:: gtod_reduce=bool
734 Enable all of the :manpage:`gettimeofday(2)` reducing options
735 (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus
736 reduce precision of the timeout somewhat to really shrink the
737 :manpage:`gettimeofday(2)` call count. With this option enabled, we only do
738 about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all
739 time keeping was enabled.
741 .. option:: gtod_cpu=int
743 Sometimes it's cheaper to dedicate a single thread of execution to just
744 getting the current time. Fio (and databases, for instance) are very
745 intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set
746 one CPU aside for doing nothing but logging current time to a shared memory
747 location. Then the other threads/processes that run I/O workloads need only
748 copy that segment, instead of entering the kernel with a
749 :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time
750 calls will be excluded from other uses. Fio will manually clear it from the
751 CPU mask of other jobs.
757 .. option:: directory=str
759 Prefix filenames with this directory. Used to place files in a different
760 location than :file:`./`. You can specify a number of directories by
761 separating the names with a ':' character. These directories will be
762 assigned equally distributed to job clones created by :option:`numjobs` as
763 long as they are using generated filenames. If specific `filename(s)` are
764 set fio will use the first listed directory, and thereby matching the
765 `filename` semantic (which generates a file for each clone if not
766 specified, but lets all clones use the same file if set).
768 See the :option:`filename` option for information on how to escape "``:``"
769 characters within the directory path itself.
771 Note: To control the directory fio will use for internal state files
772 use :option:`--aux-path`.
774 .. option:: filename=str
776 Fio normally makes up a `filename` based on the job name, thread number, and
777 file number (see :option:`filename_format`). If you want to share files
778 between threads in a job or several
779 jobs with fixed file paths, specify a `filename` for each of them to override
780 the default. If the ioengine is file based, you can specify a number of files
781 by separating the names with a ':' colon. So if you wanted a job to open
782 :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use
783 ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is
784 specified, :option:`nrfiles` is ignored. The size of regular files specified
785 by this option will be :option:`size` divided by number of files unless an
786 explicit size is specified by :option:`filesize`.
788 Each colon in the wanted path must be escaped with a ``\``
789 character. For instance, if the path is :file:`/dev/dsk/foo@3,0:c` then you
790 would use ``filename=/dev/dsk/foo@3,0\:c`` and if the path is
791 :file:`F:\\filename` then you would use ``filename=F\:\filename``.
793 On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for
794 the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc.
795 Note: Windows and FreeBSD prevent write access to areas
796 of the disk containing in-use data (e.g. filesystems).
798 The filename "`-`" is a reserved name, meaning *stdin* or *stdout*. Which
799 of the two depends on the read/write direction set.
801 .. option:: filename_format=str
803 If sharing multiple files between jobs, it is usually necessary to have fio
804 generate the exact names that you want. By default, fio will name a file
805 based on the default file format specification of
806 :file:`jobname.jobnumber.filenumber`. With this option, that can be
807 customized. Fio will recognize and replace the following keywords in this
811 The name of the worker thread or process.
813 The incremental number of the worker thread or process.
815 The incremental number of the file for that worker thread or
818 To have dependent jobs share a set of files, this option can be set to have
819 fio generate filenames that are shared between the two. For instance, if
820 :file:`testfiles.$filenum` is specified, file number 4 for any job will be
821 named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum`
822 will be used if no other format specifier is given.
824 If you specify a path then the directories will be created up to the
825 main directory for the file. So for example if you specify
826 ``filename_format=a/b/c/$jobnum`` then the directories a/b/c will be
827 created before the file setup part of the job. If you specify
828 :option:`directory` then the path will be relative that directory,
829 otherwise it is treated as the absolute path.
831 .. option:: unique_filename=bool
833 To avoid collisions between networked clients, fio defaults to prefixing any
834 generated filenames (with a directory specified) with the source of the
835 client connecting. To disable this behavior, set this option to 0.
837 .. option:: opendir=str
839 Recursively open any files below directory `str`.
841 .. option:: lockfile=str
843 Fio defaults to not locking any files before it does I/O to them. If a file
844 or file descriptor is shared, fio can serialize I/O to that file to make the
845 end result consistent. This is usual for emulating real workloads that share
846 files. The lock modes are:
849 No locking. The default.
851 Only one thread or process may do I/O at a time, excluding all
854 Read-write locking on the file. Many readers may
855 access the file at the same time, but writes get exclusive access.
857 .. option:: nrfiles=int
859 Number of files to use for this job. Defaults to 1. The size of files
860 will be :option:`size` divided by this unless explicit size is specified by
861 :option:`filesize`. Files are created for each thread separately, and each
862 file will have a file number within its name by default, as explained in
863 :option:`filename` section.
866 .. option:: openfiles=int
868 Number of files to keep open at the same time. Defaults to the same as
869 :option:`nrfiles`, can be set smaller to limit the number simultaneous
872 .. option:: file_service_type=str
874 Defines how fio decides which file from a job to service next. The following
878 Choose a file at random.
881 Round robin over opened files. This is the default.
884 Finish one file before moving on to the next. Multiple files can
885 still be open depending on :option:`openfiles`.
888 Use a *Zipf* distribution to decide what file to access.
891 Use a *Pareto* distribution to decide what file to access.
894 Use a *Gaussian* (normal) distribution to decide what file to
900 For *random*, *roundrobin*, and *sequential*, a postfix can be appended to
901 tell fio how many I/Os to issue before switching to a new file. For example,
902 specifying ``file_service_type=random:8`` would cause fio to issue
903 8 I/Os before selecting a new file at random. For the non-uniform
904 distributions, a floating point postfix can be given to influence how the
905 distribution is skewed. See :option:`random_distribution` for a description
906 of how that would work.
908 .. option:: ioscheduler=str
910 Attempt to switch the device hosting the file to the specified I/O scheduler
913 .. option:: create_serialize=bool
915 If true, serialize the file creation for the jobs. This may be handy to
916 avoid interleaving of data files, which may greatly depend on the filesystem
917 used and even the number of processors in the system. Default: true.
919 .. option:: create_fsync=bool
921 :manpage:`fsync(2)` the data file after creation. This is the default.
923 .. option:: create_on_open=bool
925 If true, don't pre-create files but allow the job's open() to create a file
926 when it's time to do I/O. Default: false -- pre-create all necessary files
929 .. option:: create_only=bool
931 If true, fio will only run the setup phase of the job. If files need to be
932 laid out or updated on disk, only that will be done -- the actual job contents
933 are not executed. Default: false.
935 .. option:: allow_file_create=bool
937 If true, fio is permitted to create files as part of its workload. If this
938 option is false, then fio will error out if
939 the files it needs to use don't already exist. Default: true.
941 .. option:: allow_mounted_write=bool
943 If this isn't set, fio will abort jobs that are destructive (e.g. that write)
944 to what appears to be a mounted device or partition. This should help catch
945 creating inadvertently destructive tests, not realizing that the test will
946 destroy data on the mounted file system. Note that some platforms don't allow
947 writing against a mounted device regardless of this option. Default: false.
949 .. option:: pre_read=bool
951 If this is given, files will be pre-read into memory before starting the
952 given I/O operation. This will also clear the :option:`invalidate` flag,
953 since it is pointless to pre-read and then drop the cache. This will only
954 work for I/O engines that are seek-able, since they allow you to read the
955 same data multiple times. Thus it will not work on non-seekable I/O engines
956 (e.g. network, splice). Default: false.
958 .. option:: unlink=bool
960 Unlink the job files when done. Not the default, as repeated runs of that
961 job would then waste time recreating the file set again and again. Default:
964 .. option:: unlink_each_loop=bool
966 Unlink job files after each iteration or loop. Default: false.
968 .. option:: zonemode=str
973 The :option:`zonerange`, :option:`zonesize`,
974 :option `zonecapacity` and option:`zoneskip`
975 parameters are ignored.
977 I/O happens in a single zone until
978 :option:`zonesize` bytes have been transferred.
979 After that number of bytes has been
980 transferred processing of the next zone
981 starts. :option `zonecapacity` is ignored.
983 Zoned block device mode. I/O happens
984 sequentially in each zone, even if random I/O
985 has been selected. Random I/O happens across
986 all zones instead of being restricted to a
987 single zone. The :option:`zoneskip` parameter
988 is ignored. :option:`zonerange` and
989 :option:`zonesize` must be identical.
991 .. option:: zonerange=int
993 Size of a single zone. See also :option:`zonesize` and
996 .. option:: zonesize=int
998 For :option:`zonemode` =strided, this is the number of bytes to
999 transfer before skipping :option:`zoneskip` bytes. If this parameter
1000 is smaller than :option:`zonerange` then only a fraction of each zone
1001 with :option:`zonerange` bytes will be accessed. If this parameter is
1002 larger than :option:`zonerange` then each zone will be accessed
1003 multiple times before skipping to the next zone.
1005 For :option:`zonemode` =zbd, this is the size of a single zone. The
1006 :option:`zonerange` parameter is ignored in this mode.
1009 .. option:: zonecapacity=int
1011 For :option:`zonemode` =zbd, this defines the capacity of a single zone,
1012 which is the accessible area starting from the zone start address.
1013 This parameter only applies when using :option:`zonemode` =zbd in
1014 combination with regular block devices. If not specified it defaults to
1015 the zone size. If the target device is a zoned block device, the zone
1016 capacity is obtained from the device information and this option is
1019 .. option:: zoneskip=int
1021 For :option:`zonemode` =strided, the number of bytes to skip after
1022 :option:`zonesize` bytes of data have been transferred. This parameter
1023 must be zero for :option:`zonemode` =zbd.
1025 .. option:: read_beyond_wp=bool
1027 This parameter applies to :option:`zonemode` =zbd only.
1029 Zoned block devices are block devices that consist of multiple zones.
1030 Each zone has a type, e.g. conventional or sequential. A conventional
1031 zone can be written at any offset that is a multiple of the block
1032 size. Sequential zones must be written sequentially. The position at
1033 which a write must occur is called the write pointer. A zoned block
1034 device can be either drive managed, host managed or host aware. For
1035 host managed devices the host must ensure that writes happen
1036 sequentially. Fio recognizes host managed devices and serializes
1037 writes to sequential zones for these devices.
1039 If a read occurs in a sequential zone beyond the write pointer then
1040 the zoned block device will complete the read without reading any data
1041 from the storage medium. Since such reads lead to unrealistically high
1042 bandwidth and IOPS numbers fio only reads beyond the write pointer if
1043 explicitly told to do so. Default: false.
1045 .. option:: max_open_zones=int
1047 When running a random write test across an entire drive many more
1048 zones will be open than in a typical application workload. Hence this
1049 command line option that allows to limit the number of open zones. The
1050 number of open zones is defined as the number of zones to which write
1051 commands are issued.
1053 .. option:: zone_reset_threshold=float
1055 A number between zero and one that indicates the ratio of logical
1056 blocks with data to the total number of logical blocks in the test
1057 above which zones should be reset periodically.
1059 .. option:: zone_reset_frequency=float
1061 A number between zero and one that indicates how often a zone reset
1062 should be issued if the zone reset threshold has been exceeded. A zone
1063 reset is submitted after each (1 / zone_reset_frequency) write
1064 requests. This and the previous parameter can be used to simulate
1065 garbage collection activity.
1071 .. option:: direct=bool
1073 If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that
1074 OpenBSD and ZFS on Solaris don't support direct I/O. On Windows the synchronous
1075 ioengines don't support direct I/O. Default: false.
1077 .. option:: atomic=bool
1079 If value is true, attempt to use atomic direct I/O. Atomic writes are
1080 guaranteed to be stable once acknowledged by the operating system. Only
1081 Linux supports O_ATOMIC right now.
1083 .. option:: buffered=bool
1085 If value is true, use buffered I/O. This is the opposite of the
1086 :option:`direct` option. Defaults to true.
1088 .. option:: readwrite=str, rw=str
1090 Type of I/O pattern. Accepted values are:
1097 Sequential trims (Linux block devices and SCSI
1098 character devices only).
1104 Random trims (Linux block devices and SCSI
1105 character devices only).
1107 Sequential mixed reads and writes.
1109 Random mixed reads and writes.
1111 Sequential trim+write sequences. Blocks will be trimmed first,
1112 then the same blocks will be written to.
1114 Fio defaults to read if the option is not specified. For the mixed I/O
1115 types, the default is to split them 50/50. For certain types of I/O the
1116 result may still be skewed a bit, since the speed may be different.
1118 It is possible to specify the number of I/Os to do before getting a new
1119 offset by appending ``:<nr>`` to the end of the string given. For a
1120 random read, it would look like ``rw=randread:8`` for passing in an offset
1121 modifier with a value of 8. If the suffix is used with a sequential I/O
1122 pattern, then the *<nr>* value specified will be **added** to the generated
1123 offset for each I/O turning sequential I/O into sequential I/O with holes.
1124 For instance, using ``rw=write:4k`` will skip 4k for every write. Also see
1125 the :option:`rw_sequencer` option.
1127 .. option:: rw_sequencer=str
1129 If an offset modifier is given by appending a number to the ``rw=<str>``
1130 line, then this option controls how that number modifies the I/O offset
1131 being generated. Accepted values are:
1134 Generate sequential offset.
1136 Generate the same offset.
1138 ``sequential`` is only useful for random I/O, where fio would normally
1139 generate a new random offset for every I/O. If you append e.g. 8 to randread,
1140 you would get a new random offset for every 8 I/Os. The result would be a
1141 seek for only every 8 I/Os, instead of for every I/O. Use ``rw=randread:8``
1142 to specify that. As sequential I/O is already sequential, setting
1143 ``sequential`` for that would not result in any differences. ``identical``
1144 behaves in a similar fashion, except it sends the same offset 8 number of
1145 times before generating a new offset.
1147 .. option:: unified_rw_reporting=bool
1149 Fio normally reports statistics on a per data direction basis, meaning that
1150 reads, writes, and trims are accounted and reported separately. If this
1151 option is set fio sums the results and report them as "mixed" instead.
1153 .. option:: randrepeat=bool
1155 Seed the random number generator used for random I/O patterns in a
1156 predictable way so the pattern is repeatable across runs. Default: true.
1158 .. option:: allrandrepeat=bool
1160 Seed all random number generators in a predictable way so results are
1161 repeatable across runs. Default: false.
1163 .. option:: randseed=int
1165 Seed the random number generators based on this seed value, to be able to
1166 control what sequence of output is being generated. If not set, the random
1167 sequence depends on the :option:`randrepeat` setting.
1169 .. option:: fallocate=str
1171 Whether pre-allocation is performed when laying down files.
1172 Accepted values are:
1175 Do not pre-allocate space.
1178 Use a platform's native pre-allocation call but fall back to
1179 **none** behavior if it fails/is not implemented.
1182 Pre-allocate via :manpage:`posix_fallocate(3)`.
1185 Pre-allocate via :manpage:`fallocate(2)` with
1186 FALLOC_FL_KEEP_SIZE set.
1189 Extend file to final size via :manpage:`ftruncate(2)`
1190 instead of allocating.
1193 Backward-compatible alias for **none**.
1196 Backward-compatible alias for **posix**.
1198 May not be available on all supported platforms. **keep** is only available
1199 on Linux. If using ZFS on Solaris this cannot be set to **posix**
1200 because ZFS doesn't support pre-allocation. Default: **native** if any
1201 pre-allocation methods except **truncate** are available, **none** if not.
1203 Note that using **truncate** on Windows will interact surprisingly
1204 with non-sequential write patterns. When writing to a file that has
1205 been extended by setting the end-of-file information, Windows will
1206 backfill the unwritten portion of the file up to that offset with
1207 zeroes before issuing the new write. This means that a single small
1208 write to the end of an extended file will stall until the entire
1209 file has been filled with zeroes.
1211 .. option:: fadvise_hint=str
1213 Use :manpage:`posix_fadvise(2)` or :manpage:`posix_fadvise(2)` to
1214 advise the kernel on what I/O patterns are likely to be issued.
1215 Accepted values are:
1218 Backwards-compatible hint for "no hint".
1221 Backwards compatible hint for "advise with fio workload type". This
1222 uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL**
1223 for a sequential workload.
1226 Advise using **FADV_SEQUENTIAL**.
1229 Advise using **FADV_RANDOM**.
1231 .. option:: write_hint=str
1233 Use :manpage:`fcntl(2)` to advise the kernel what life time to expect
1234 from a write. Only supported on Linux, as of version 4.13. Accepted
1238 No particular life time associated with this file.
1241 Data written to this file has a short life time.
1244 Data written to this file has a medium life time.
1247 Data written to this file has a long life time.
1250 Data written to this file has a very long life time.
1252 The values are all relative to each other, and no absolute meaning
1253 should be associated with them.
1255 .. option:: offset=int
1257 Start I/O at the provided offset in the file, given as either a fixed size in
1258 bytes or a percentage. If a percentage is given, the generated offset will be
1259 aligned to the minimum ``blocksize`` or to the value of ``offset_align`` if
1260 provided. Data before the given offset will not be touched. This
1261 effectively caps the file size at `real_size - offset`. Can be combined with
1262 :option:`size` to constrain the start and end range of the I/O workload.
1263 A percentage can be specified by a number between 1 and 100 followed by '%',
1264 for example, ``offset=20%`` to specify 20%.
1266 .. option:: offset_align=int
1268 If set to non-zero value, the byte offset generated by a percentage ``offset``
1269 is aligned upwards to this value. Defaults to 0 meaning that a percentage
1270 offset is aligned to the minimum block size.
1272 .. option:: offset_increment=int
1274 If this is provided, then the real offset becomes `offset + offset_increment
1275 * thread_number`, where the thread number is a counter that starts at 0 and
1276 is incremented for each sub-job (i.e. when :option:`numjobs` option is
1277 specified). This option is useful if there are several jobs which are
1278 intended to operate on a file in parallel disjoint segments, with even
1279 spacing between the starting points. Percentages can be used for this option.
1280 If a percentage is given, the generated offset will be aligned to the minimum
1281 ``blocksize`` or to the value of ``offset_align`` if provided.
1283 .. option:: number_ios=int
1285 Fio will normally perform I/Os until it has exhausted the size of the region
1286 set by :option:`size`, or if it exhaust the allocated time (or hits an error
1287 condition). With this setting, the range/size can be set independently of
1288 the number of I/Os to perform. When fio reaches this number, it will exit
1289 normally and report status. Note that this does not extend the amount of I/O
1290 that will be done, it will only stop fio if this condition is met before
1291 other end-of-job criteria.
1293 .. option:: fsync=int
1295 If writing to a file, issue an :manpage:`fsync(2)` (or its equivalent) of
1296 the dirty data for every number of blocks given. For example, if you give 32
1297 as a parameter, fio will sync the file after every 32 writes issued. If fio is
1298 using non-buffered I/O, we may not sync the file. The exception is the sg
1299 I/O engine, which synchronizes the disk cache anyway. Defaults to 0, which
1300 means fio does not periodically issue and wait for a sync to complete. Also
1301 see :option:`end_fsync` and :option:`fsync_on_close`.
1303 .. option:: fdatasync=int
1305 Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and
1306 not metadata blocks. In Windows, FreeBSD, DragonFlyBSD or OSX there is no
1307 :manpage:`fdatasync(2)` so this falls back to using :manpage:`fsync(2)`.
1308 Defaults to 0, which means fio does not periodically issue and wait for a
1309 data-only sync to complete.
1311 .. option:: write_barrier=int
1313 Make every `N-th` write a barrier write.
1315 .. option:: sync_file_range=str:int
1317 Use :manpage:`sync_file_range(2)` for every `int` number of write
1318 operations. Fio will track range of writes that have happened since the last
1319 :manpage:`sync_file_range(2)` call. `str` can currently be one or more of:
1322 SYNC_FILE_RANGE_WAIT_BEFORE
1324 SYNC_FILE_RANGE_WRITE
1326 SYNC_FILE_RANGE_WAIT_AFTER
1328 So if you do ``sync_file_range=wait_before,write:8``, fio would use
1329 ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8
1330 writes. Also see the :manpage:`sync_file_range(2)` man page. This option is
1333 .. option:: overwrite=bool
1335 If true, writes to a file will always overwrite existing data. If the file
1336 doesn't already exist, it will be created before the write phase begins. If
1337 the file exists and is large enough for the specified write phase, nothing
1338 will be done. Default: false.
1340 .. option:: end_fsync=bool
1342 If true, :manpage:`fsync(2)` file contents when a write stage has completed.
1345 .. option:: fsync_on_close=bool
1347 If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs
1348 from :option:`end_fsync` in that it will happen on every file close, not
1349 just at the end of the job. Default: false.
1351 .. option:: rwmixread=int
1353 Percentage of a mixed workload that should be reads. Default: 50.
1355 .. option:: rwmixwrite=int
1357 Percentage of a mixed workload that should be writes. If both
1358 :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not
1359 add up to 100%, the latter of the two will be used to override the
1360 first. This may interfere with a given rate setting, if fio is asked to
1361 limit reads or writes to a certain rate. If that is the case, then the
1362 distribution may be skewed. Default: 50.
1364 .. option:: random_distribution=str:float[,str:float][,str:float]
1366 By default, fio will use a completely uniform random distribution when asked
1367 to perform random I/O. Sometimes it is useful to skew the distribution in
1368 specific ways, ensuring that some parts of the data is more hot than others.
1369 fio includes the following distribution models:
1372 Uniform random distribution
1381 Normal (Gaussian) distribution
1384 Zoned random distribution
1387 Zone absolute random distribution
1389 When using a **zipf** or **pareto** distribution, an input value is also
1390 needed to define the access pattern. For **zipf**, this is the `Zipf
1391 theta`. For **pareto**, it's the `Pareto power`. Fio includes a test
1392 program, :command:`fio-genzipf`, that can be used visualize what the given input
1393 values will yield in terms of hit rates. If you wanted to use **zipf** with
1394 a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the
1395 option. If a non-uniform model is used, fio will disable use of the random
1396 map. For the **normal** distribution, a normal (Gaussian) deviation is
1397 supplied as a value between 0 and 100.
1399 For a **zoned** distribution, fio supports specifying percentages of I/O
1400 access that should fall within what range of the file or device. For
1401 example, given a criteria of:
1403 * 60% of accesses should be to the first 10%
1404 * 30% of accesses should be to the next 20%
1405 * 8% of accesses should be to the next 30%
1406 * 2% of accesses should be to the next 40%
1408 we can define that through zoning of the random accesses. For the above
1409 example, the user would do::
1411 random_distribution=zoned:60/10:30/20:8/30:2/40
1413 A **zoned_abs** distribution works exactly like the **zoned**, except
1414 that it takes absolute sizes. For example, let's say you wanted to
1415 define access according to the following criteria:
1417 * 60% of accesses should be to the first 20G
1418 * 30% of accesses should be to the next 100G
1419 * 10% of accesses should be to the next 500G
1421 we can define an absolute zoning distribution with:
1423 random_distribution=zoned_abs=60/20G:30/100G:10/500g
1425 For both **zoned** and **zoned_abs**, fio supports defining up to
1428 Similarly to how :option:`bssplit` works for setting ranges and
1429 percentages of block sizes. Like :option:`bssplit`, it's possible to
1430 specify separate zones for reads, writes, and trims. If just one set
1431 is given, it'll apply to all of them. This goes for both **zoned**
1432 **zoned_abs** distributions.
1434 .. option:: percentage_random=int[,int][,int]
1436 For a random workload, set how big a percentage should be random. This
1437 defaults to 100%, in which case the workload is fully random. It can be set
1438 from anywhere from 0 to 100. Setting it to 0 would make the workload fully
1439 sequential. Any setting in between will result in a random mix of sequential
1440 and random I/O, at the given percentages. Comma-separated values may be
1441 specified for reads, writes, and trims as described in :option:`blocksize`.
1443 .. option:: norandommap
1445 Normally fio will cover every block of the file when doing random I/O. If
1446 this option is given, fio will just get a new random offset without looking
1447 at past I/O history. This means that some blocks may not be read or written,
1448 and that some blocks may be read/written more than once. If this option is
1449 used with :option:`verify` and multiple blocksizes (via :option:`bsrange`),
1450 only intact blocks are verified, i.e., partially-overwritten blocks are
1451 ignored. With an async I/O engine and an I/O depth > 1, it is possible for
1452 the same block to be overwritten, which can cause verification errors. Either
1453 do not use norandommap in this case, or also use the lfsr random generator.
1455 .. option:: softrandommap=bool
1457 See :option:`norandommap`. If fio runs with the random block map enabled and
1458 it fails to allocate the map, if this option is set it will continue without
1459 a random block map. As coverage will not be as complete as with random maps,
1460 this option is disabled by default.
1462 .. option:: random_generator=str
1464 Fio supports the following engines for generating I/O offsets for random I/O:
1467 Strong 2^88 cycle random number generator.
1469 Linear feedback shift register generator.
1471 Strong 64-bit 2^258 cycle random number generator.
1473 **tausworthe** is a strong random number generator, but it requires tracking
1474 on the side if we want to ensure that blocks are only read or written
1475 once. **lfsr** guarantees that we never generate the same offset twice, and
1476 it's also less computationally expensive. It's not a true random generator,
1477 however, though for I/O purposes it's typically good enough. **lfsr** only
1478 works with single block sizes, not with workloads that use multiple block
1479 sizes. If used with such a workload, fio may read or write some blocks
1480 multiple times. The default value is **tausworthe**, unless the required
1481 space exceeds 2^32 blocks. If it does, then **tausworthe64** is
1482 selected automatically.
1488 .. option:: blocksize=int[,int][,int], bs=int[,int][,int]
1490 The block size in bytes used for I/O units. Default: 4096. A single value
1491 applies to reads, writes, and trims. Comma-separated values may be
1492 specified for reads, writes, and trims. A value not terminated in a comma
1493 applies to subsequent types.
1498 means 256k for reads, writes and trims.
1501 means 8k for reads, 32k for writes and trims.
1504 means 8k for reads, 32k for writes, and default for trims.
1507 means default for reads, 8k for writes and trims.
1510 means default for reads, 8k for writes, and default for trims.
1512 .. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange]
1514 A range of block sizes in bytes for I/O units. The issued I/O unit will
1515 always be a multiple of the minimum size, unless
1516 :option:`blocksize_unaligned` is set.
1518 Comma-separated ranges may be specified for reads, writes, and trims as
1519 described in :option:`blocksize`.
1521 Example: ``bsrange=1k-4k,2k-8k``.
1523 .. option:: bssplit=str[,str][,str]
1525 Sometimes you want even finer grained control of the block sizes
1526 issued, not just an even split between them. This option allows you to
1527 weight various block sizes, so that you are able to define a specific
1528 amount of block sizes issued. The format for this option is::
1530 bssplit=blocksize/percentage:blocksize/percentage
1532 for as many block sizes as needed. So if you want to define a workload
1533 that has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would
1536 bssplit=4k/10:64k/50:32k/40
1538 Ordering does not matter. If the percentage is left blank, fio will
1539 fill in the remaining values evenly. So a bssplit option like this one::
1541 bssplit=4k/50:1k/:32k/
1543 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always
1544 add up to 100, if bssplit is given a range that adds up to more, it
1547 Comma-separated values may be specified for reads, writes, and trims as
1548 described in :option:`blocksize`.
1550 If you want a workload that has 50% 2k reads and 50% 4k reads, while
1551 having 90% 4k writes and 10% 8k writes, you would specify::
1553 bssplit=2k/50:4k/50,4k/90:8k/10
1555 Fio supports defining up to 64 different weights for each data
1558 .. option:: blocksize_unaligned, bs_unaligned
1560 If set, fio will issue I/O units with any size within
1561 :option:`blocksize_range`, not just multiples of the minimum size. This
1562 typically won't work with direct I/O, as that normally requires sector
1565 .. option:: bs_is_seq_rand=bool
1567 If this option is set, fio will use the normal read,write blocksize settings
1568 as sequential,random blocksize settings instead. Any random read or write
1569 will use the WRITE blocksize settings, and any sequential read or write will
1570 use the READ blocksize settings.
1572 .. option:: blockalign=int[,int][,int], ba=int[,int][,int]
1574 Boundary to which fio will align random I/O units. Default:
1575 :option:`blocksize`. Minimum alignment is typically 512b for using direct
1576 I/O, though it usually depends on the hardware block size. This option is
1577 mutually exclusive with using a random map for files, so it will turn off
1578 that option. Comma-separated values may be specified for reads, writes, and
1579 trims as described in :option:`blocksize`.
1585 .. option:: zero_buffers
1587 Initialize buffers with all zeros. Default: fill buffers with random data.
1589 .. option:: refill_buffers
1591 If this option is given, fio will refill the I/O buffers on every
1592 submit. Only makes sense if :option:`zero_buffers` isn't specified,
1593 naturally. Defaults to being unset i.e., the buffer is only filled at
1594 init time and the data in it is reused when possible but if any of
1595 :option:`verify`, :option:`buffer_compress_percentage` or
1596 :option:`dedupe_percentage` are enabled then `refill_buffers` is also
1597 automatically enabled.
1599 .. option:: scramble_buffers=bool
1601 If :option:`refill_buffers` is too costly and the target is using data
1602 deduplication, then setting this option will slightly modify the I/O buffer
1603 contents to defeat normal de-dupe attempts. This is not enough to defeat
1604 more clever block compression attempts, but it will stop naive dedupe of
1605 blocks. Default: true.
1607 .. option:: buffer_compress_percentage=int
1609 If this is set, then fio will attempt to provide I/O buffer content
1610 (on WRITEs) that compresses to the specified level. Fio does this by
1611 providing a mix of random data followed by fixed pattern data. The
1612 fixed pattern is either zeros, or the pattern specified by
1613 :option:`buffer_pattern`. If the `buffer_pattern` option is used, it
1614 might skew the compression ratio slightly. Setting
1615 `buffer_compress_percentage` to a value other than 100 will also
1616 enable :option:`refill_buffers` in order to reduce the likelihood that
1617 adjacent blocks are so similar that they over compress when seen
1618 together. See :option:`buffer_compress_chunk` for how to set a finer or
1619 coarser granularity for the random/fixed data region. Defaults to unset
1620 i.e., buffer data will not adhere to any compression level.
1622 .. option:: buffer_compress_chunk=int
1624 This setting allows fio to manage how big the random/fixed data region
1625 is when using :option:`buffer_compress_percentage`. When
1626 `buffer_compress_chunk` is set to some non-zero value smaller than the
1627 block size, fio can repeat the random/fixed region throughout the I/O
1628 buffer at the specified interval (which particularly useful when
1629 bigger block sizes are used for a job). When set to 0, fio will use a
1630 chunk size that matches the block size resulting in a single
1631 random/fixed region within the I/O buffer. Defaults to 512. When the
1632 unit is omitted, the value is interpreted in bytes.
1634 .. option:: buffer_pattern=str
1636 If set, fio will fill the I/O buffers with this pattern or with the contents
1637 of a file. If not set, the contents of I/O buffers are defined by the other
1638 options related to buffer contents. The setting can be any pattern of bytes,
1639 and can be prefixed with 0x for hex values. It may also be a string, where
1640 the string must then be wrapped with ``""``. Or it may also be a filename,
1641 where the filename must be wrapped with ``''`` in which case the file is
1642 opened and read. Note that not all the file contents will be read if that
1643 would cause the buffers to overflow. So, for example::
1645 buffer_pattern='filename'
1649 buffer_pattern="abcd"
1657 buffer_pattern=0xdeadface
1659 Also you can combine everything together in any order::
1661 buffer_pattern=0xdeadface"abcd"-12'filename'
1663 .. option:: dedupe_percentage=int
1665 If set, fio will generate this percentage of identical buffers when
1666 writing. These buffers will be naturally dedupable. The contents of the
1667 buffers depend on what other buffer compression settings have been set. It's
1668 possible to have the individual buffers either fully compressible, or not at
1669 all -- this option only controls the distribution of unique buffers. Setting
1670 this option will also enable :option:`refill_buffers` to prevent every buffer
1673 .. option:: invalidate=bool
1675 Invalidate the buffer/page cache parts of the files to be used prior to
1676 starting I/O if the platform and file type support it. Defaults to true.
1677 This will be ignored if :option:`pre_read` is also specified for the
1680 .. option:: sync=bool
1682 Use synchronous I/O for buffered writes. For the majority of I/O engines,
1683 this means using O_SYNC. Default: false.
1685 .. option:: iomem=str, mem=str
1687 Fio can use various types of memory as the I/O unit buffer. The allowed
1691 Use memory from :manpage:`malloc(3)` as the buffers. Default memory
1695 Use shared memory as the buffers. Allocated through
1696 :manpage:`shmget(2)`.
1699 Same as shm, but use huge pages as backing.
1702 Use :manpage:`mmap(2)` to allocate buffers. May either be anonymous memory, or can
1703 be file backed if a filename is given after the option. The format
1704 is `mem=mmap:/path/to/file`.
1707 Use a memory mapped huge file as the buffer backing. Append filename
1708 after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`.
1711 Same as mmap, but use a MMAP_SHARED mapping.
1714 Use GPU memory as the buffers for GPUDirect RDMA benchmark.
1715 The :option:`ioengine` must be `rdma`.
1717 The area allocated is a function of the maximum allowed bs size for the job,
1718 multiplied by the I/O depth given. Note that for **shmhuge** and
1719 **mmaphuge** to work, the system must have free huge pages allocated. This
1720 can normally be checked and set by reading/writing
1721 :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page
1722 is 4MiB in size. So to calculate the number of huge pages you need for a
1723 given job file, add up the I/O depth of all jobs (normally one unless
1724 :option:`iodepth` is used) and multiply by the maximum bs set. Then divide
1725 that number by the huge page size. You can see the size of the huge pages in
1726 :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero
1727 number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also
1728 see :option:`hugepage-size`.
1730 **mmaphuge** also needs to have hugetlbfs mounted and the file location
1731 should point there. So if it's mounted in :file:`/huge`, you would use
1732 `mem=mmaphuge:/huge/somefile`.
1734 .. option:: iomem_align=int, mem_align=int
1736 This indicates the memory alignment of the I/O memory buffers. Note that
1737 the given alignment is applied to the first I/O unit buffer, if using
1738 :option:`iodepth` the alignment of the following buffers are given by the
1739 :option:`bs` used. In other words, if using a :option:`bs` that is a
1740 multiple of the page sized in the system, all buffers will be aligned to
1741 this value. If using a :option:`bs` that is not page aligned, the alignment
1742 of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and
1745 .. option:: hugepage-size=int
1747 Defines the size of a huge page. Must at least be equal to the system
1748 setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably
1749 always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the
1750 preferred way to set this to avoid setting a non-pow-2 bad value.
1752 .. option:: lockmem=int
1754 Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to
1755 simulate a smaller amount of memory. The amount specified is per worker.
1761 .. option:: size=int
1763 The total size of file I/O for each thread of this job. Fio will run until
1764 this many bytes has been transferred, unless runtime is limited by other options
1765 (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`).
1766 Fio will divide this size between the available files determined by options
1767 such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is
1768 specified by the job. If the result of division happens to be 0, the size is
1769 set to the physical size of the given files or devices if they exist.
1770 If this option is not specified, fio will use the full size of the given
1771 files or devices. If the files do not exist, size must be given. It is also
1772 possible to give size as a percentage between 1 and 100. If ``size=20%`` is
1773 given, fio will use 20% of the full size of the given files or devices.
1774 Can be combined with :option:`offset` to constrain the start and end range
1775 that I/O will be done within.
1777 .. option:: io_size=int, io_limit=int
1779 Normally fio operates within the region set by :option:`size`, which means
1780 that the :option:`size` option sets both the region and size of I/O to be
1781 performed. Sometimes that is not what you want. With this option, it is
1782 possible to define just the amount of I/O that fio should do. For instance,
1783 if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio
1784 will perform I/O within the first 20GiB but exit when 5GiB have been
1785 done. The opposite is also possible -- if :option:`size` is set to 20GiB,
1786 and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within
1787 the 0..20GiB region.
1789 .. option:: filesize=irange(int)
1791 Individual file sizes. May be a range, in which case fio will select sizes
1792 for files at random within the given range and limited to :option:`size` in
1793 total (if that is given). If not given, each created file is the same size.
1794 This option overrides :option:`size` in terms of file size, which means
1795 this value is used as a fixed size or possible range of each file.
1797 .. option:: file_append=bool
1799 Perform I/O after the end of the file. Normally fio will operate within the
1800 size of a file. If this option is set, then fio will append to the file
1801 instead. This has identical behavior to setting :option:`offset` to the size
1802 of a file. This option is ignored on non-regular files.
1804 .. option:: fill_device=bool, fill_fs=bool
1806 Sets size to something really large and waits for ENOSPC (no space left on
1807 device) as the terminating condition. Only makes sense with sequential
1808 write. For a read workload, the mount point will be filled first then I/O
1809 started on the result. This option doesn't make sense if operating on a raw
1810 device node, since the size of that is already known by the file system.
1811 Additionally, writing beyond end-of-device will not return ENOSPC there.
1817 .. option:: ioengine=str
1819 Defines how the job issues I/O to the file. The following types are defined:
1822 Basic :manpage:`read(2)` or :manpage:`write(2)`
1823 I/O. :manpage:`lseek(2)` is used to position the I/O location.
1824 See :option:`fsync` and :option:`fdatasync` for syncing write I/Os.
1827 Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on
1828 all supported operating systems except for Windows.
1831 Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate
1832 queuing by coalescing adjacent I/Os into a single submission.
1835 Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O.
1838 Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O.
1841 Fast Linux native asynchronous I/O. Supports async IO
1842 for both direct and buffered IO.
1843 This engine defines engine specific options.
1846 Linux native asynchronous I/O. Note that Linux may only support
1847 queued behavior with non-buffered I/O (set ``direct=1`` or
1849 This engine defines engine specific options.
1852 POSIX asynchronous I/O using :manpage:`aio_read(3)` and
1853 :manpage:`aio_write(3)`.
1856 Solaris native asynchronous I/O.
1859 Windows native asynchronous I/O. Default on Windows.
1862 File is memory mapped with :manpage:`mmap(2)` and data copied
1863 to/from using :manpage:`memcpy(3)`.
1866 :manpage:`splice(2)` is used to transfer the data and
1867 :manpage:`vmsplice(2)` to transfer data from user space to the
1871 SCSI generic sg v3 I/O. May either be synchronous using the SG_IO
1872 ioctl, or if the target is an sg character device we use
1873 :manpage:`read(2)` and :manpage:`write(2)` for asynchronous
1874 I/O. Requires :option:`filename` option to specify either block or
1875 character devices. This engine supports trim operations.
1876 The sg engine includes engine specific options.
1879 Doesn't transfer any data, just pretends to. This is mainly used to
1880 exercise fio itself and for debugging/testing purposes.
1883 Transfer over the network to given ``host:port``. Depending on the
1884 :option:`protocol` used, the :option:`hostname`, :option:`port`,
1885 :option:`listen` and :option:`filename` options are used to specify
1886 what sort of connection to make, while the :option:`protocol` option
1887 determines which protocol will be used. This engine defines engine
1891 Like **net**, but uses :manpage:`splice(2)` and
1892 :manpage:`vmsplice(2)` to map data and send/receive.
1893 This engine defines engine specific options.
1896 Doesn't transfer any data, but burns CPU cycles according to the
1897 :option:`cpuload` and :option:`cpuchunks` options. Setting
1898 :option:`cpuload`\=85 will cause that job to do nothing but burn 85%
1899 of the CPU. In case of SMP machines, use :option:`numjobs`\=<nr_of_cpu>
1900 to get desired CPU usage, as the cpuload only loads a
1901 single CPU at the desired rate. A job never finishes unless there is
1902 at least one non-cpuio job.
1905 The RDMA I/O engine supports both RDMA memory semantics
1906 (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the
1907 InfiniBand, RoCE and iWARP protocols. This engine defines engine
1911 I/O engine that does regular fallocate to simulate data transfer as
1915 does fallocate(,mode = FALLOC_FL_KEEP_SIZE,).
1918 does fallocate(,mode = 0).
1921 does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE).
1924 I/O engine that sends :manpage:`ftruncate(2)` operations in response
1925 to write (DDIR_WRITE) events. Each ftruncate issued sets the file's
1926 size to the current block offset. :option:`blocksize` is ignored.
1929 I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate
1930 defragment activity in request to DDIR_WRITE event.
1933 I/O engine supporting direct access to Ceph Reliable Autonomic
1934 Distributed Object Store (RADOS) via librados. This ioengine
1935 defines engine specific options.
1938 I/O engine supporting direct access to Ceph Rados Block Devices
1939 (RBD) via librbd without the need to use the kernel rbd driver. This
1940 ioengine defines engine specific options.
1943 I/O engine supporting GET/PUT requests over HTTP(S) with libcurl to
1944 a WebDAV or S3 endpoint. This ioengine defines engine specific options.
1946 This engine only supports direct IO of iodepth=1; you need to scale this
1947 via numjobs. blocksize defines the size of the objects to be created.
1949 TRIM is translated to object deletion.
1952 Using GlusterFS libgfapi sync interface to direct access to
1953 GlusterFS volumes without having to go through FUSE. This ioengine
1954 defines engine specific options.
1957 Using GlusterFS libgfapi async interface to direct access to
1958 GlusterFS volumes without having to go through FUSE. This ioengine
1959 defines engine specific options.
1962 Read and write through Hadoop (HDFS). The :option:`filename` option
1963 is used to specify host,port of the hdfs name-node to connect. This
1964 engine interprets offsets a little differently. In HDFS, files once
1965 created cannot be modified so random writes are not possible. To
1966 imitate this the libhdfs engine expects a bunch of small files to be
1967 created over HDFS and will randomly pick a file from them
1968 based on the offset generated by fio backend (see the example
1969 job file to create such files, use ``rw=write`` option). Please
1970 note, it may be necessary to set environment variables to work
1971 with HDFS/libhdfs properly. Each job uses its own connection to
1975 Read, write and erase an MTD character device (e.g.,
1976 :file:`/dev/mtd0`). Discards are treated as erases. Depending on the
1977 underlying device type, the I/O may have to go in a certain pattern,
1978 e.g., on NAND, writing sequentially to erase blocks and discarding
1979 before overwriting. The `trimwrite` mode works well for this
1983 Read and write using filesystem DAX to a file on a filesystem
1984 mounted with DAX on a persistent memory device through the PMDK
1988 Read and write using device DAX to a persistent memory device (e.g.,
1989 /dev/dax0.0) through the PMDK libpmem library.
1992 Prefix to specify loading an external I/O engine object file. Append
1993 the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load
1994 ioengine :file:`foo.o` in :file:`/tmp`. The path can be either
1995 absolute or relative. See :file:`engines/skeleton_external.c` for
1996 details of writing an external I/O engine.
1999 Simply create the files and do no I/O to them. You still need to
2000 set `filesize` so that all the accounting still occurs, but no
2001 actual I/O will be done other than creating the file.
2004 Simply do stat() and do no I/O to the file. You need to set 'filesize'
2005 and 'nrfiles', so that files will be created.
2006 This engine is to measure file lookup and meta data access.
2009 Read and write using mmap I/O to a file on a filesystem
2010 mounted with DAX on a persistent memory device through the PMDK
2014 Synchronous read and write using DDN's Infinite Memory Engine (IME).
2015 This engine is very basic and issues calls to IME whenever an IO is
2019 Synchronous read and write using DDN's Infinite Memory Engine (IME).
2020 This engine uses iovecs and will try to stack as much IOs as possible
2021 (if the IOs are "contiguous" and the IO depth is not exceeded)
2022 before issuing a call to IME.
2025 Asynchronous read and write using DDN's Infinite Memory Engine (IME).
2026 This engine will try to stack as much IOs as possible by creating
2027 requests for IME. FIO will then decide when to commit these requests.
2029 Read and write iscsi lun with libiscsi.
2031 Read and write a Network Block Device (NBD).
2033 I/O engine specific parameters
2034 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2036 In addition, there are some parameters which are only valid when a specific
2037 :option:`ioengine` is in use. These are used identically to normal parameters,
2038 with the caveat that when used on the command line, they must come after the
2039 :option:`ioengine` that defines them is selected.
2041 .. option:: cmdprio_percentage=int : [io_uring] [libaio]
2043 Set the percentage of I/O that will be issued with higher priority by setting
2044 the priority bit. Non-read I/O is likely unaffected by ``cmdprio_percentage``.
2045 This option cannot be used with the `prio` or `prioclass` options. For this
2046 option to set the priority bit properly, NCQ priority must be supported and
2047 enabled and :option:`direct`\=1 option must be used. fio must also be run as
2050 .. option:: fixedbufs : [io_uring]
2052 If fio is asked to do direct IO, then Linux will map pages for each
2053 IO call, and release them when IO is done. If this option is set, the
2054 pages are pre-mapped before IO is started. This eliminates the need to
2055 map and release for each IO. This is more efficient, and reduces the
2058 .. option:: hipri : [io_uring]
2060 If this option is set, fio will attempt to use polled IO completions.
2061 Normal IO completions generate interrupts to signal the completion of
2062 IO, polled completions do not. Hence they are require active reaping
2063 by the application. The benefits are more efficient IO for high IOPS
2064 scenarios, and lower latencies for low queue depth IO.
2066 .. option:: registerfiles : [io_uring]
2068 With this option, fio registers the set of files being used with the
2069 kernel. This avoids the overhead of managing file counts in the kernel,
2070 making the submission and completion part more lightweight. Required
2071 for the below :option:`sqthread_poll` option.
2073 .. option:: sqthread_poll : [io_uring]
2075 Normally fio will submit IO by issuing a system call to notify the
2076 kernel of available items in the SQ ring. If this option is set, the
2077 act of submitting IO will be done by a polling thread in the kernel.
2078 This frees up cycles for fio, at the cost of using more CPU in the
2081 .. option:: sqthread_poll_cpu : [io_uring]
2083 When :option:`sqthread_poll` is set, this option provides a way to
2084 define which CPU should be used for the polling thread.
2086 .. option:: userspace_reap : [libaio]
2088 Normally, with the libaio engine in use, fio will use the
2089 :manpage:`io_getevents(2)` system call to reap newly returned events. With
2090 this flag turned on, the AIO ring will be read directly from user-space to
2091 reap events. The reaping mode is only enabled when polling for a minimum of
2092 0 events (e.g. when :option:`iodepth_batch_complete` `=0`).
2094 .. option:: hipri : [pvsync2]
2096 Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority
2099 .. option:: hipri_percentage : [pvsync2]
2101 When hipri is set this determines the probability of a pvsync2 I/O being high
2102 priority. The default is 100%.
2104 .. option:: nowait : [pvsync2] [libaio] [io_uring]
2106 By default if a request cannot be executed immediately (e.g. resource starvation,
2107 waiting on locks) it is queued and the initiating process will be blocked until
2108 the required resource becomes free.
2110 This option sets the RWF_NOWAIT flag (supported from the 4.14 Linux kernel) and
2111 the call will return instantly with EAGAIN or a partial result rather than waiting.
2113 It is useful to also use ignore_error=EAGAIN when using this option.
2115 Note: glibc 2.27, 2.28 have a bug in syscall wrappers preadv2, pwritev2.
2116 They return EOPNOTSUP instead of EAGAIN.
2118 For cached I/O, using this option usually means a request operates only with
2119 cached data. Currently the RWF_NOWAIT flag does not supported for cached write.
2121 For direct I/O, requests will only succeed if cache invalidation isn't required,
2122 file blocks are fully allocated and the disk request could be issued immediately.
2124 .. option:: cpuload=int : [cpuio]
2126 Attempt to use the specified percentage of CPU cycles. This is a mandatory
2127 option when using cpuio I/O engine.
2129 .. option:: cpuchunks=int : [cpuio]
2131 Split the load into cycles of the given time. In microseconds.
2133 .. option:: exit_on_io_done=bool : [cpuio]
2135 Detect when I/O threads are done, then exit.
2137 .. option:: namenode=str : [libhdfs]
2139 The hostname or IP address of a HDFS cluster namenode to contact.
2141 .. option:: port=int
2145 The listening port of the HFDS cluster namenode.
2149 The TCP or UDP port to bind to or connect to. If this is used with
2150 :option:`numjobs` to spawn multiple instances of the same job type, then
2151 this will be the starting port number since fio will use a range of
2156 The port to use for RDMA-CM communication. This should be the same value
2157 on the client and the server side.
2159 .. option:: hostname=str : [netsplice] [net] [rdma]
2161 The hostname or IP address to use for TCP, UDP or RDMA-CM based I/O. If the job
2162 is a TCP listener or UDP reader, the hostname is not used and must be omitted
2163 unless it is a valid UDP multicast address.
2165 .. option:: interface=str : [netsplice] [net]
2167 The IP address of the network interface used to send or receive UDP
2170 .. option:: ttl=int : [netsplice] [net]
2172 Time-to-live value for outgoing UDP multicast packets. Default: 1.
2174 .. option:: nodelay=bool : [netsplice] [net]
2176 Set TCP_NODELAY on TCP connections.
2178 .. option:: protocol=str, proto=str : [netsplice] [net]
2180 The network protocol to use. Accepted values are:
2183 Transmission control protocol.
2185 Transmission control protocol V6.
2187 User datagram protocol.
2189 User datagram protocol V6.
2193 When the protocol is TCP or UDP, the port must also be given, as well as the
2194 hostname if the job is a TCP listener or UDP reader. For unix sockets, the
2195 normal :option:`filename` option should be used and the port is invalid.
2197 .. option:: listen : [netsplice] [net]
2199 For TCP network connections, tell fio to listen for incoming connections
2200 rather than initiating an outgoing connection. The :option:`hostname` must
2201 be omitted if this option is used.
2203 .. option:: pingpong : [netsplice] [net]
2205 Normally a network writer will just continue writing data, and a network
2206 reader will just consume packages. If ``pingpong=1`` is set, a writer will
2207 send its normal payload to the reader, then wait for the reader to send the
2208 same payload back. This allows fio to measure network latencies. The
2209 submission and completion latencies then measure local time spent sending or
2210 receiving, and the completion latency measures how long it took for the
2211 other end to receive and send back. For UDP multicast traffic
2212 ``pingpong=1`` should only be set for a single reader when multiple readers
2213 are listening to the same address.
2215 .. option:: window_size : [netsplice] [net]
2217 Set the desired socket buffer size for the connection.
2219 .. option:: mss : [netsplice] [net]
2221 Set the TCP maximum segment size (TCP_MAXSEG).
2223 .. option:: donorname=str : [e4defrag]
2225 File will be used as a block donor (swap extents between files).
2227 .. option:: inplace=int : [e4defrag]
2229 Configure donor file blocks allocation strategy:
2232 Default. Preallocate donor's file on init.
2234 Allocate space immediately inside defragment event, and free right
2237 .. option:: clustername=str : [rbd,rados]
2239 Specifies the name of the Ceph cluster.
2241 .. option:: rbdname=str : [rbd]
2243 Specifies the name of the RBD.
2245 .. option:: pool=str : [rbd,rados]
2247 Specifies the name of the Ceph pool containing RBD or RADOS data.
2249 .. option:: clientname=str : [rbd,rados]
2251 Specifies the username (without the 'client.' prefix) used to access the
2252 Ceph cluster. If the *clustername* is specified, the *clientname* shall be
2253 the full *type.id* string. If no type. prefix is given, fio will add
2254 'client.' by default.
2256 .. option:: busy_poll=bool : [rbd,rados]
2258 Poll store instead of waiting for completion. Usually this provides better
2259 throughput at cost of higher(up to 100%) CPU utilization.
2261 .. option:: skip_bad=bool : [mtd]
2263 Skip operations against known bad blocks.
2265 .. option:: hdfsdirectory : [libhdfs]
2267 libhdfs will create chunk in this HDFS directory.
2269 .. option:: chunk_size : [libhdfs]
2271 The size of the chunk to use for each file.
2273 .. option:: verb=str : [rdma]
2275 The RDMA verb to use on this side of the RDMA ioengine connection. Valid
2276 values are write, read, send and recv. These correspond to the equivalent
2277 RDMA verbs (e.g. write = rdma_write etc.). Note that this only needs to be
2278 specified on the client side of the connection. See the examples folder.
2280 .. option:: bindname=str : [rdma]
2282 The name to use to bind the local RDMA-CM connection to a local RDMA device.
2283 This could be a hostname or an IPv4 or IPv6 address. On the server side this
2284 will be passed into the rdma_bind_addr() function and on the client site it
2285 will be used in the rdma_resolve_add() function. This can be useful when
2286 multiple paths exist between the client and the server or in certain loopback
2289 .. option:: stat_type=str : [filestat]
2291 Specify stat system call type to measure lookup/getattr performance.
2292 Default is **stat** for :manpage:`stat(2)`.
2294 .. option:: readfua=bool : [sg]
2296 With readfua option set to 1, read operations include
2297 the force unit access (fua) flag. Default is 0.
2299 .. option:: writefua=bool : [sg]
2301 With writefua option set to 1, write operations include
2302 the force unit access (fua) flag. Default is 0.
2304 .. option:: sg_write_mode=str : [sg]
2306 Specify the type of write commands to issue. This option can take three values:
2309 This is the default where write opcodes are issued as usual.
2311 Issue WRITE AND VERIFY commands. The BYTCHK bit is set to 0. This
2312 directs the device to carry out a medium verification with no data
2313 comparison. The writefua option is ignored with this selection.
2315 Issue WRITE SAME commands. This transfers a single block to the device
2316 and writes this same block of data to a contiguous sequence of LBAs
2317 beginning at the specified offset. fio's block size parameter specifies
2318 the amount of data written with each command. However, the amount of data
2319 actually transferred to the device is equal to the device's block
2320 (sector) size. For a device with 512 byte sectors, blocksize=8k will
2321 write 16 sectors with each command. fio will still generate 8k of data
2322 for each command but only the first 512 bytes will be used and
2323 transferred to the device. The writefua option is ignored with this
2326 .. option:: http_host=str : [http]
2328 Hostname to connect to. For S3, this could be the bucket hostname.
2329 Default is **localhost**
2331 .. option:: http_user=str : [http]
2333 Username for HTTP authentication.
2335 .. option:: http_pass=str : [http]
2337 Password for HTTP authentication.
2339 .. option:: https=str : [http]
2341 Enable HTTPS instead of http. *on* enables HTTPS; *insecure*
2342 will enable HTTPS, but disable SSL peer verification (use with
2343 caution!). Default is **off**
2345 .. option:: http_mode=str : [http]
2347 Which HTTP access mode to use: *webdav*, *swift*, or *s3*.
2348 Default is **webdav**
2350 .. option:: http_s3_region=str : [http]
2352 The S3 region/zone string.
2353 Default is **us-east-1**
2355 .. option:: http_s3_key=str : [http]
2359 .. option:: http_s3_keyid=str : [http]
2361 The S3 key/access id.
2363 .. option:: http_swift_auth_token=str : [http]
2365 The Swift auth token. See the example configuration file on how
2368 .. option:: http_verbose=int : [http]
2370 Enable verbose requests from libcurl. Useful for debugging. 1
2371 turns on verbose logging from libcurl, 2 additionally enables
2372 HTTP IO tracing. Default is **0**
2374 .. option:: uri=str : [nbd]
2376 Specify the NBD URI of the server to test. The string
2377 is a standard NBD URI
2378 (see https://github.com/NetworkBlockDevice/nbd/tree/master/doc).
2379 Example URIs: nbd://localhost:10809
2380 nbd+unix:///?socket=/tmp/socket
2381 nbds://tlshost/exportname
2386 .. option:: iodepth=int
2388 Number of I/O units to keep in flight against the file. Note that
2389 increasing *iodepth* beyond 1 will not affect synchronous ioengines (except
2390 for small degrees when :option:`verify_async` is in use). Even async
2391 engines may impose OS restrictions causing the desired depth not to be
2392 achieved. This may happen on Linux when using libaio and not setting
2393 :option:`direct`\=1, since buffered I/O is not async on that OS. Keep an
2394 eye on the I/O depth distribution in the fio output to verify that the
2395 achieved depth is as expected. Default: 1.
2397 .. option:: iodepth_batch_submit=int, iodepth_batch=int
2399 This defines how many pieces of I/O to submit at once. It defaults to 1
2400 which means that we submit each I/O as soon as it is available, but can be
2401 raised to submit bigger batches of I/O at the time. If it is set to 0 the
2402 :option:`iodepth` value will be used.
2404 .. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int
2406 This defines how many pieces of I/O to retrieve at once. It defaults to 1
2407 which means that we'll ask for a minimum of 1 I/O in the retrieval process
2408 from the kernel. The I/O retrieval will go on until we hit the limit set by
2409 :option:`iodepth_low`. If this variable is set to 0, then fio will always
2410 check for completed events before queuing more I/O. This helps reduce I/O
2411 latency, at the cost of more retrieval system calls.
2413 .. option:: iodepth_batch_complete_max=int
2415 This defines maximum pieces of I/O to retrieve at once. This variable should
2416 be used along with :option:`iodepth_batch_complete_min`\=int variable,
2417 specifying the range of min and max amount of I/O which should be
2418 retrieved. By default it is equal to the :option:`iodepth_batch_complete_min`
2423 iodepth_batch_complete_min=1
2424 iodepth_batch_complete_max=<iodepth>
2426 which means that we will retrieve at least 1 I/O and up to the whole
2427 submitted queue depth. If none of I/O has been completed yet, we will wait.
2431 iodepth_batch_complete_min=0
2432 iodepth_batch_complete_max=<iodepth>
2434 which means that we can retrieve up to the whole submitted queue depth, but
2435 if none of I/O has been completed yet, we will NOT wait and immediately exit
2436 the system call. In this example we simply do polling.
2438 .. option:: iodepth_low=int
2440 The low water mark indicating when to start filling the queue
2441 again. Defaults to the same as :option:`iodepth`, meaning that fio will
2442 attempt to keep the queue full at all times. If :option:`iodepth` is set to
2443 e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of
2444 16 requests, it will let the depth drain down to 4 before starting to fill
2447 .. option:: serialize_overlap=bool
2449 Serialize in-flight I/Os that might otherwise cause or suffer from data races.
2450 When two or more I/Os are submitted simultaneously, there is no guarantee that
2451 the I/Os will be processed or completed in the submitted order. Further, if
2452 two or more of those I/Os are writes, any overlapping region between them can
2453 become indeterminate/undefined on certain storage. These issues can cause
2454 verification to fail erratically when at least one of the racing I/Os is
2455 changing data and the overlapping region has a non-zero size. Setting
2456 ``serialize_overlap`` tells fio to avoid provoking this behavior by explicitly
2457 serializing in-flight I/Os that have a non-zero overlap. Note that setting
2458 this option can reduce both performance and the :option:`iodepth` achieved.
2460 This option only applies to I/Os issued for a single job except when it is
2461 enabled along with :option:`io_submit_mode`\=offload. In offload mode, fio
2462 will check for overlap among all I/Os submitted by offload jobs with :option:`serialize_overlap`
2467 .. option:: io_submit_mode=str
2469 This option controls how fio submits the I/O to the I/O engine. The default
2470 is `inline`, which means that the fio job threads submit and reap I/O
2471 directly. If set to `offload`, the job threads will offload I/O submission
2472 to a dedicated pool of I/O threads. This requires some coordination and thus
2473 has a bit of extra overhead, especially for lower queue depth I/O where it
2474 can increase latencies. The benefit is that fio can manage submission rates
2475 independently of the device completion rates. This avoids skewed latency
2476 reporting if I/O gets backed up on the device side (the coordinated omission
2483 .. option:: thinktime=time
2485 Stall the job for the specified period of time after an I/O has completed before issuing the
2486 next. May be used to simulate processing being done by an application.
2487 When the unit is omitted, the value is interpreted in microseconds. See
2488 :option:`thinktime_blocks` and :option:`thinktime_spin`.
2490 .. option:: thinktime_spin=time
2492 Only valid if :option:`thinktime` is set - pretend to spend CPU time doing
2493 something with the data received, before falling back to sleeping for the
2494 rest of the period specified by :option:`thinktime`. When the unit is
2495 omitted, the value is interpreted in microseconds.
2497 .. option:: thinktime_blocks=int
2499 Only valid if :option:`thinktime` is set - control how many blocks to issue,
2500 before waiting :option:`thinktime` usecs. If not set, defaults to 1 which will make
2501 fio wait :option:`thinktime` usecs after every block. This effectively makes any
2502 queue depth setting redundant, since no more than 1 I/O will be queued
2503 before we have to complete it and do our :option:`thinktime`. In other words, this
2504 setting effectively caps the queue depth if the latter is larger.
2506 .. option:: rate=int[,int][,int]
2508 Cap the bandwidth used by this job. The number is in bytes/sec, the normal
2509 suffix rules apply. Comma-separated values may be specified for reads,
2510 writes, and trims as described in :option:`blocksize`.
2512 For example, using `rate=1m,500k` would limit reads to 1MiB/sec and writes to
2513 500KiB/sec. Capping only reads or writes can be done with `rate=,500k` or
2514 `rate=500k,` where the former will only limit writes (to 500KiB/sec) and the
2515 latter will only limit reads.
2517 .. option:: rate_min=int[,int][,int]
2519 Tell fio to do whatever it can to maintain at least this bandwidth. Failing
2520 to meet this requirement will cause the job to exit. Comma-separated values
2521 may be specified for reads, writes, and trims as described in
2522 :option:`blocksize`.
2524 .. option:: rate_iops=int[,int][,int]
2526 Cap the bandwidth to this number of IOPS. Basically the same as
2527 :option:`rate`, just specified independently of bandwidth. If the job is
2528 given a block size range instead of a fixed value, the smallest block size
2529 is used as the metric. Comma-separated values may be specified for reads,
2530 writes, and trims as described in :option:`blocksize`.
2532 .. option:: rate_iops_min=int[,int][,int]
2534 If fio doesn't meet this rate of I/O, it will cause the job to exit.
2535 Comma-separated values may be specified for reads, writes, and trims as
2536 described in :option:`blocksize`.
2538 .. option:: rate_process=str
2540 This option controls how fio manages rated I/O submissions. The default is
2541 `linear`, which submits I/O in a linear fashion with fixed delays between
2542 I/Os that gets adjusted based on I/O completion rates. If this is set to
2543 `poisson`, fio will submit I/O based on a more real world random request
2544 flow, known as the Poisson process
2545 (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be
2546 10^6 / IOPS for the given workload.
2548 .. option:: rate_ignore_thinktime=bool
2550 By default, fio will attempt to catch up to the specified rate setting,
2551 if any kind of thinktime setting was used. If this option is set, then
2552 fio will ignore the thinktime and continue doing IO at the specified
2553 rate, instead of entering a catch-up mode after thinktime is done.
2559 .. option:: latency_target=time
2561 If set, fio will attempt to find the max performance point that the given
2562 workload will run at while maintaining a latency below this target. When
2563 the unit is omitted, the value is interpreted in microseconds. See
2564 :option:`latency_window` and :option:`latency_percentile`.
2566 .. option:: latency_window=time
2568 Used with :option:`latency_target` to specify the sample window that the job
2569 is run at varying queue depths to test the performance. When the unit is
2570 omitted, the value is interpreted in microseconds.
2572 .. option:: latency_percentile=float
2574 The percentage of I/Os that must fall within the criteria specified by
2575 :option:`latency_target` and :option:`latency_window`. If not set, this
2576 defaults to 100.0, meaning that all I/Os must be equal or below to the value
2577 set by :option:`latency_target`.
2579 .. option:: latency_run=bool
2581 Used with :option:`latency_target`. If false (default), fio will find
2582 the highest queue depth that meets :option:`latency_target` and exit. If
2583 true, fio will continue running and try to meet :option:`latency_target`
2584 by adjusting queue depth.
2586 .. option:: max_latency=time
2588 If set, fio will exit the job with an ETIMEDOUT error if it exceeds this
2589 maximum latency. When the unit is omitted, the value is interpreted in
2592 .. option:: rate_cycle=int
2594 Average bandwidth for :option:`rate` and :option:`rate_min` over this number
2595 of milliseconds. Defaults to 1000.
2601 .. option:: write_iolog=str
2603 Write the issued I/O patterns to the specified file. See
2604 :option:`read_iolog`. Specify a separate file for each job, otherwise the
2605 iologs will be interspersed and the file may be corrupt.
2607 .. option:: read_iolog=str
2609 Open an iolog with the specified filename and replay the I/O patterns it
2610 contains. This can be used to store a workload and replay it sometime
2611 later. The iolog given may also be a blktrace binary file, which allows fio
2612 to replay a workload captured by :command:`blktrace`. See
2613 :manpage:`blktrace(8)` for how to capture such logging data. For blktrace
2614 replay, the file needs to be turned into a blkparse binary data file first
2615 (``blkparse <device> -o /dev/null -d file_for_fio.bin``).
2616 You can specify a number of files by separating the names with a ':'
2617 character. See the :option:`filename` option for information on how to
2618 escape ':' characters within the file names. These files will
2619 be sequentially assigned to job clones created by :option:`numjobs`.
2620 '-' is a reserved name, meaning read from stdin, notably if
2621 :option:`filename` is set to '-' which means stdin as well, then
2622 this flag can't be set to '-'.
2624 .. option:: read_iolog_chunked=bool
2626 Determines how iolog is read. If false(default) entire :option:`read_iolog`
2627 will be read at once. If selected true, input from iolog will be read
2628 gradually. Useful when iolog is very large, or it is generated.
2630 .. option:: merge_blktrace_file=str
2632 When specified, rather than replaying the logs passed to :option:`read_iolog`,
2633 the logs go through a merge phase which aggregates them into a single
2634 blktrace. The resulting file is then passed on as the :option:`read_iolog`
2635 parameter. The intention here is to make the order of events consistent.
2636 This limits the influence of the scheduler compared to replaying multiple
2637 blktraces via concurrent jobs.
2639 .. option:: merge_blktrace_scalars=float_list
2641 This is a percentage based option that is index paired with the list of
2642 files passed to :option:`read_iolog`. When merging is performed, scale
2643 the time of each event by the corresponding amount. For example,
2644 ``--merge_blktrace_scalars="50:100"`` runs the first trace in halftime
2645 and the second trace in realtime. This knob is separately tunable from
2646 :option:`replay_time_scale` which scales the trace during runtime and
2647 does not change the output of the merge unlike this option.
2649 .. option:: merge_blktrace_iters=float_list
2651 This is a whole number option that is index paired with the list of files
2652 passed to :option:`read_iolog`. When merging is performed, run each trace
2653 for the specified number of iterations. For example,
2654 ``--merge_blktrace_iters="2:1"`` runs the first trace for two iterations
2655 and the second trace for one iteration.
2657 .. option:: replay_no_stall=bool
2659 When replaying I/O with :option:`read_iolog` the default behavior is to
2660 attempt to respect the timestamps within the log and replay them with the
2661 appropriate delay between IOPS. By setting this variable fio will not
2662 respect the timestamps and attempt to replay them as fast as possible while
2663 still respecting ordering. The result is the same I/O pattern to a given
2664 device, but different timings.
2666 .. option:: replay_time_scale=int
2668 When replaying I/O with :option:`read_iolog`, fio will honor the
2669 original timing in the trace. With this option, it's possible to scale
2670 the time. It's a percentage option, if set to 50 it means run at 50%
2671 the original IO rate in the trace. If set to 200, run at twice the
2672 original IO rate. Defaults to 100.
2674 .. option:: replay_redirect=str
2676 While replaying I/O patterns using :option:`read_iolog` the default behavior
2677 is to replay the IOPS onto the major/minor device that each IOP was recorded
2678 from. This is sometimes undesirable because on a different machine those
2679 major/minor numbers can map to a different device. Changing hardware on the
2680 same system can also result in a different major/minor mapping.
2681 ``replay_redirect`` causes all I/Os to be replayed onto the single specified
2682 device regardless of the device it was recorded
2683 from. i.e. :option:`replay_redirect`\= :file:`/dev/sdc` would cause all I/O
2684 in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means
2685 multiple devices will be replayed onto a single device, if the trace
2686 contains multiple devices. If you want multiple devices to be replayed
2687 concurrently to multiple redirected devices you must blkparse your trace
2688 into separate traces and replay them with independent fio invocations.
2689 Unfortunately this also breaks the strict time ordering between multiple
2692 .. option:: replay_align=int
2694 Force alignment of the byte offsets in a trace to this value. The value
2695 must be a power of 2.
2697 .. option:: replay_scale=int
2699 Scale byte offsets down by this factor when replaying traces. Should most
2700 likely use :option:`replay_align` as well.
2702 .. option:: replay_skip=str
2704 Sometimes it's useful to skip certain IO types in a replay trace.
2705 This could be, for instance, eliminating the writes in the trace.
2706 Or not replaying the trims/discards, if you are redirecting to
2707 a device that doesn't support them. This option takes a comma
2708 separated list of read, write, trim, sync.
2711 Threads, processes and job synchronization
2712 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2716 Fio defaults to creating jobs by using fork, however if this option is
2717 given, fio will create jobs by using POSIX Threads' function
2718 :manpage:`pthread_create(3)` to create threads instead.
2720 .. option:: wait_for=str
2722 If set, the current job won't be started until all workers of the specified
2723 waitee job are done.
2725 ``wait_for`` operates on the job name basis, so there are a few
2726 limitations. First, the waitee must be defined prior to the waiter job
2727 (meaning no forward references). Second, if a job is being referenced as a
2728 waitee, it must have a unique name (no duplicate waitees).
2730 .. option:: nice=int
2732 Run the job with the given nice value. See man :manpage:`nice(2)`.
2734 On Windows, values less than -15 set the process class to "High"; -1 through
2735 -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle"
2738 .. option:: prio=int
2740 Set the I/O priority value of this job. Linux limits us to a positive value
2741 between 0 and 7, with 0 being the highest. See man
2742 :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating
2743 systems since meaning of priority may differ. For per-command priority
2744 setting, see I/O engine specific `cmdprio_percentage` and `hipri_percentage`
2747 .. option:: prioclass=int
2749 Set the I/O priority class. See man :manpage:`ionice(1)`. For per-command
2750 priority setting, see I/O engine specific `cmdprio_percentage` and
2751 `hipri_percentage` options.
2753 .. option:: cpus_allowed=str
2755 Controls the same options as :option:`cpumask`, but accepts a textual
2756 specification of the permitted CPUs instead and CPUs are indexed from 0. So
2757 to use CPUs 0 and 5 you would specify ``cpus_allowed=0,5``. This option also
2758 allows a range of CPUs to be specified -- say you wanted a binding to CPUs
2759 0, 5, and 8 to 15, you would set ``cpus_allowed=0,5,8-15``.
2761 On Windows, when ``cpus_allowed`` is unset only CPUs from fio's current
2762 processor group will be used and affinity settings are inherited from the
2763 system. An fio build configured to target Windows 7 makes options that set
2764 CPUs processor group aware and values will set both the processor group
2765 and a CPU from within that group. For example, on a system where processor
2766 group 0 has 40 CPUs and processor group 1 has 32 CPUs, ``cpus_allowed``
2767 values between 0 and 39 will bind CPUs from processor group 0 and
2768 ``cpus_allowed`` values between 40 and 71 will bind CPUs from processor
2769 group 1. When using ``cpus_allowed_policy=shared`` all CPUs specified by a
2770 single ``cpus_allowed`` option must be from the same processor group. For
2771 Windows fio builds not built for Windows 7, CPUs will only be selected from
2772 (and be relative to) whatever processor group fio happens to be running in
2773 and CPUs from other processor groups cannot be used.
2775 .. option:: cpus_allowed_policy=str
2777 Set the policy of how fio distributes the CPUs specified by
2778 :option:`cpus_allowed` or :option:`cpumask`. Two policies are supported:
2781 All jobs will share the CPU set specified.
2783 Each job will get a unique CPU from the CPU set.
2785 **shared** is the default behavior, if the option isn't specified. If
2786 **split** is specified, then fio will assign one cpu per job. If not
2787 enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs
2790 .. option:: cpumask=int
2792 Set the CPU affinity of this job. The parameter given is a bit mask of
2793 allowed CPUs the job may run on. So if you want the allowed CPUs to be 1
2794 and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man
2795 :manpage:`sched_setaffinity(2)`. This may not work on all supported
2796 operating systems or kernel versions. This option doesn't work well for a
2797 higher CPU count than what you can store in an integer mask, so it can only
2798 control cpus 1-32. For boxes with larger CPU counts, use
2799 :option:`cpus_allowed`.
2801 .. option:: numa_cpu_nodes=str
2803 Set this job running on specified NUMA nodes' CPUs. The arguments allow
2804 comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable
2805 NUMA options support, fio must be built on a system with libnuma-dev(el)
2808 .. option:: numa_mem_policy=str
2810 Set this job's memory policy and corresponding NUMA nodes. Format of the
2815 ``mode`` is one of the following memory policies: ``default``, ``prefer``,
2816 ``bind``, ``interleave`` or ``local``. For ``default`` and ``local`` memory
2817 policies, no node needs to be specified. For ``prefer``, only one node is
2818 allowed. For ``bind`` and ``interleave`` the ``nodelist`` may be as
2819 follows: a comma delimited list of numbers, A-B ranges, or `all`.
2821 .. option:: cgroup=str
2823 Add job to this control group. If it doesn't exist, it will be created. The
2824 system must have a mounted cgroup blkio mount point for this to work. If
2825 your system doesn't have it mounted, you can do so with::
2827 # mount -t cgroup -o blkio none /cgroup
2829 .. option:: cgroup_weight=int
2831 Set the weight of the cgroup to this value. See the documentation that comes
2832 with the kernel, allowed values are in the range of 100..1000.
2834 .. option:: cgroup_nodelete=bool
2836 Normally fio will delete the cgroups it has created after the job
2837 completion. To override this behavior and to leave cgroups around after the
2838 job completion, set ``cgroup_nodelete=1``. This can be useful if one wants
2839 to inspect various cgroup files after job completion. Default: false.
2841 .. option:: flow_id=int
2843 The ID of the flow. If not specified, it defaults to being a global
2844 flow. See :option:`flow`.
2846 .. option:: flow=int
2848 Weight in token-based flow control. If this value is used, then there is a
2849 'flow counter' which is used to regulate the proportion of activity between
2850 two or more jobs. Fio attempts to keep this flow counter near zero. The
2851 ``flow`` parameter stands for how much should be added or subtracted to the
2852 flow counter on each iteration of the main I/O loop. That is, if one job has
2853 ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8
2854 ratio in how much one runs vs the other.
2856 .. option:: flow_sleep=int
2858 The period of time, in microseconds, to wait after the flow counter
2859 has exceeded its proportion before retrying operations.
2861 .. option:: stonewall, wait_for_previous
2863 Wait for preceding jobs in the job file to exit, before starting this
2864 one. Can be used to insert serialization points in the job file. A stone
2865 wall also implies starting a new reporting group, see
2866 :option:`group_reporting`.
2870 By default, fio will continue running all other jobs when one job finishes.
2871 Sometimes this is not the desired action. Setting ``exitall`` will instead
2872 make fio terminate all jobs in the same group, as soon as one job of that
2875 .. option:: exit_what
2877 By default, fio will continue running all other jobs when one job finishes.
2878 Sometimes this is not the desired action. Setting ``exit_all`` will
2879 instead make fio terminate all jobs in the same group. The option
2880 ``exit_what`` allows to control which jobs get terminated when ``exitall`` is
2881 enabled. The default is ``group`` and does not change the behaviour of
2882 ``exitall``. The setting ``all`` terminates all jobs. The setting ``stonewall``
2883 terminates all currently running jobs across all groups and continues execution
2884 with the next stonewalled group.
2886 .. option:: exec_prerun=str
2888 Before running this job, issue the command specified through
2889 :manpage:`system(3)`. Output is redirected in a file called
2890 :file:`jobname.prerun.txt`.
2892 .. option:: exec_postrun=str
2894 After the job completes, issue the command specified though
2895 :manpage:`system(3)`. Output is redirected in a file called
2896 :file:`jobname.postrun.txt`.
2900 Instead of running as the invoking user, set the user ID to this value
2901 before the thread/process does any work.
2905 Set group ID, see :option:`uid`.
2911 .. option:: verify_only
2913 Do not perform specified workload, only verify data still matches previous
2914 invocation of this workload. This option allows one to check data multiple
2915 times at a later date without overwriting it. This option makes sense only
2916 for workloads that write data, and does not support workloads with the
2917 :option:`time_based` option set.
2919 .. option:: do_verify=bool
2921 Run the verify phase after a write phase. Only valid if :option:`verify` is
2924 .. option:: verify=str
2926 If writing to a file, fio can verify the file contents after each iteration
2927 of the job. Each verification method also implies verification of special
2928 header, which is written to the beginning of each block. This header also
2929 includes meta information, like offset of the block, block number, timestamp
2930 when block was written, etc. :option:`verify` can be combined with
2931 :option:`verify_pattern` option. The allowed values are:
2934 Use an md5 sum of the data area and store it in the header of
2938 Use an experimental crc64 sum of the data area and store it in the
2939 header of each block.
2942 Use a crc32c sum of the data area and store it in the header of
2943 each block. This will automatically use hardware acceleration
2944 (e.g. SSE4.2 on an x86 or CRC crypto extensions on ARM64) but will
2945 fall back to software crc32c if none is found. Generally the
2946 fastest checksum fio supports when hardware accelerated.
2952 Use a crc32 sum of the data area and store it in the header of each
2956 Use a crc16 sum of the data area and store it in the header of each
2960 Use a crc7 sum of the data area and store it in the header of each
2964 Use xxhash as the checksum function. Generally the fastest software
2965 checksum that fio supports.
2968 Use sha512 as the checksum function.
2971 Use sha256 as the checksum function.
2974 Use optimized sha1 as the checksum function.
2977 Use optimized sha3-224 as the checksum function.
2980 Use optimized sha3-256 as the checksum function.
2983 Use optimized sha3-384 as the checksum function.
2986 Use optimized sha3-512 as the checksum function.
2989 This option is deprecated, since now meta information is included in
2990 generic verification header and meta verification happens by
2991 default. For detailed information see the description of the
2992 :option:`verify` setting. This option is kept because of
2993 compatibility's sake with old configurations. Do not use it.
2996 Verify a strict pattern. Normally fio includes a header with some
2997 basic information and checksumming, but if this option is set, only
2998 the specific pattern set with :option:`verify_pattern` is verified.
3001 Only pretend to verify. Useful for testing internals with
3002 :option:`ioengine`\=null, not for much else.
3004 This option can be used for repeated burn-in tests of a system to make sure
3005 that the written data is also correctly read back. If the data direction
3006 given is a read or random read, fio will assume that it should verify a
3007 previously written file. If the data direction includes any form of write,
3008 the verify will be of the newly written data.
3010 To avoid false verification errors, do not use the norandommap option when
3011 verifying data with async I/O engines and I/O depths > 1. Or use the
3012 norandommap and the lfsr random generator together to avoid writing to the
3013 same offset with muliple outstanding I/Os.
3015 .. option:: verify_offset=int
3017 Swap the verification header with data somewhere else in the block before
3018 writing. It is swapped back before verifying.
3020 .. option:: verify_interval=int
3022 Write the verification header at a finer granularity than the
3023 :option:`blocksize`. It will be written for chunks the size of
3024 ``verify_interval``. :option:`blocksize` should divide this evenly.
3026 .. option:: verify_pattern=str
3028 If set, fio will fill the I/O buffers with this pattern. Fio defaults to
3029 filling with totally random bytes, but sometimes it's interesting to fill
3030 with a known pattern for I/O verification purposes. Depending on the width
3031 of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time (it can
3032 be either a decimal or a hex number). The ``verify_pattern`` if larger than
3033 a 32-bit quantity has to be a hex number that starts with either "0x" or
3034 "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o
3035 format, which means that for each block offset will be written and then
3036 verified back, e.g.::
3040 Or use combination of everything::
3042 verify_pattern=0xff%o"abcd"-12
3044 .. option:: verify_fatal=bool
3046 Normally fio will keep checking the entire contents before quitting on a
3047 block verification failure. If this option is set, fio will exit the job on
3048 the first observed failure. Default: false.
3050 .. option:: verify_dump=bool
3052 If set, dump the contents of both the original data block and the data block
3053 we read off disk to files. This allows later analysis to inspect just what
3054 kind of data corruption occurred. Off by default.
3056 .. option:: verify_async=int
3058 Fio will normally verify I/O inline from the submitting thread. This option
3059 takes an integer describing how many async offload threads to create for I/O
3060 verification instead, causing fio to offload the duty of verifying I/O
3061 contents to one or more separate threads. If using this offload option, even
3062 sync I/O engines can benefit from using an :option:`iodepth` setting higher
3063 than 1, as it allows them to have I/O in flight while verifies are running.
3064 Defaults to 0 async threads, i.e. verification is not asynchronous.
3066 .. option:: verify_async_cpus=str
3068 Tell fio to set the given CPU affinity on the async I/O verification
3069 threads. See :option:`cpus_allowed` for the format used.
3071 .. option:: verify_backlog=int
3073 Fio will normally verify the written contents of a job that utilizes verify
3074 once that job has completed. In other words, everything is written then
3075 everything is read back and verified. You may want to verify continually
3076 instead for a variety of reasons. Fio stores the meta data associated with
3077 an I/O block in memory, so for large verify workloads, quite a bit of memory
3078 would be used up holding this meta data. If this option is enabled, fio will
3079 write only N blocks before verifying these blocks.
3081 .. option:: verify_backlog_batch=int
3083 Control how many blocks fio will verify if :option:`verify_backlog` is
3084 set. If not set, will default to the value of :option:`verify_backlog`
3085 (meaning the entire queue is read back and verified). If
3086 ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all
3087 blocks will be verified, if ``verify_backlog_batch`` is larger than
3088 :option:`verify_backlog`, some blocks will be verified more than once.
3090 .. option:: verify_state_save=bool
3092 When a job exits during the write phase of a verify workload, save its
3093 current state. This allows fio to replay up until that point, if the verify
3094 state is loaded for the verify read phase. The format of the filename is,
3097 <type>-<jobname>-<jobindex>-verify.state.
3099 <type> is "local" for a local run, "sock" for a client/server socket
3100 connection, and "ip" (192.168.0.1, for instance) for a networked
3101 client/server connection. Defaults to true.
3103 .. option:: verify_state_load=bool
3105 If a verify termination trigger was used, fio stores the current write state
3106 of each thread. This can be used at verification time so that fio knows how
3107 far it should verify. Without this information, fio will run a full
3108 verification pass, according to the settings in the job file used. Default
3111 .. option:: trim_percentage=int
3113 Number of verify blocks to discard/trim.
3115 .. option:: trim_verify_zero=bool
3117 Verify that trim/discarded blocks are returned as zeros.
3119 .. option:: trim_backlog=int
3121 Trim after this number of blocks are written.
3123 .. option:: trim_backlog_batch=int
3125 Trim this number of I/O blocks.
3127 .. option:: experimental_verify=bool
3129 Enable experimental verification.
3134 .. option:: steadystate=str:float, ss=str:float
3136 Define the criterion and limit for assessing steady state performance. The
3137 first parameter designates the criterion whereas the second parameter sets
3138 the threshold. When the criterion falls below the threshold for the
3139 specified duration, the job will stop. For example, `iops_slope:0.1%` will
3140 direct fio to terminate the job when the least squares regression slope
3141 falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled
3142 this will apply to all jobs in the group. Below is the list of available
3143 steady state assessment criteria. All assessments are carried out using only
3144 data from the rolling collection window. Threshold limits can be expressed
3145 as a fixed value or as a percentage of the mean in the collection window.
3147 When using this feature, most jobs should include the :option:`time_based`
3148 and :option:`runtime` options or the :option:`loops` option so that fio does not
3149 stop running after it has covered the full size of the specified file(s) or device(s).
3152 Collect IOPS data. Stop the job if all individual IOPS measurements
3153 are within the specified limit of the mean IOPS (e.g., ``iops:2``
3154 means that all individual IOPS values must be within 2 of the mean,
3155 whereas ``iops:0.2%`` means that all individual IOPS values must be
3156 within 0.2% of the mean IOPS to terminate the job).
3159 Collect IOPS data and calculate the least squares regression
3160 slope. Stop the job if the slope falls below the specified limit.
3163 Collect bandwidth data. Stop the job if all individual bandwidth
3164 measurements are within the specified limit of the mean bandwidth.
3167 Collect bandwidth data and calculate the least squares regression
3168 slope. Stop the job if the slope falls below the specified limit.
3170 .. option:: steadystate_duration=time, ss_dur=time
3172 A rolling window of this duration will be used to judge whether steady state
3173 has been reached. Data will be collected once per second. The default is 0
3174 which disables steady state detection. When the unit is omitted, the
3175 value is interpreted in seconds.
3177 .. option:: steadystate_ramp_time=time, ss_ramp=time
3179 Allow the job to run for the specified duration before beginning data
3180 collection for checking the steady state job termination criterion. The
3181 default is 0. When the unit is omitted, the value is interpreted in seconds.
3184 Measurements and reporting
3185 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3187 .. option:: per_job_logs=bool
3189 If set, this generates bw/clat/iops log with per file private filenames. If
3190 not set, jobs with identical names will share the log filename. Default:
3193 .. option:: group_reporting
3195 It may sometimes be interesting to display statistics for groups of jobs as
3196 a whole instead of for each individual job. This is especially true if
3197 :option:`numjobs` is used; looking at individual thread/process output
3198 quickly becomes unwieldy. To see the final report per-group instead of
3199 per-job, use :option:`group_reporting`. Jobs in a file will be part of the
3200 same reporting group, unless if separated by a :option:`stonewall`, or by
3201 using :option:`new_group`.
3203 .. option:: new_group
3205 Start a new reporting group. See: :option:`group_reporting`. If not given,
3206 all jobs in a file will be part of the same reporting group, unless
3207 separated by a :option:`stonewall`.
3209 .. option:: stats=bool
3211 By default, fio collects and shows final output results for all jobs
3212 that run. If this option is set to 0, then fio will ignore it in
3213 the final stat output.
3215 .. option:: write_bw_log=str
3217 If given, write a bandwidth log for this job. Can be used to store data of
3218 the bandwidth of the jobs in their lifetime.
3220 If no str argument is given, the default filename of
3221 :file:`jobname_type.x.log` is used. Even when the argument is given, fio
3222 will still append the type of log. So if one specifies::
3226 The actual log name will be :file:`foo_bw.x.log` where `x` is the index
3227 of the job (`1..N`, where `N` is the number of jobs). If
3228 :option:`per_job_logs` is false, then the filename will not include the
3231 The included :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these
3232 text files into nice graphs. See `Log File Formats`_ for how data is
3233 structured within the file.
3235 .. option:: write_lat_log=str
3237 Same as :option:`write_bw_log`, except this option creates I/O
3238 submission (e.g., :file:`name_slat.x.log`), completion (e.g.,
3239 :file:`name_clat.x.log`), and total (e.g., :file:`name_lat.x.log`)
3240 latency files instead. See :option:`write_bw_log` for details about
3241 the filename format and `Log File Formats`_ for how data is structured
3244 .. option:: write_hist_log=str
3246 Same as :option:`write_bw_log` but writes an I/O completion latency
3247 histogram file (e.g., :file:`name_hist.x.log`) instead. Note that this
3248 file will be empty unless :option:`log_hist_msec` has also been set.
3249 See :option:`write_bw_log` for details about the filename format and
3250 `Log File Formats`_ for how data is structured within the file.
3252 .. option:: write_iops_log=str
3254 Same as :option:`write_bw_log`, but writes an IOPS file (e.g.
3255 :file:`name_iops.x.log`) instead. Because fio defaults to individual
3256 I/O logging, the value entry in the IOPS log will be 1 unless windowed
3257 logging (see :option:`log_avg_msec`) has been enabled. See
3258 :option:`write_bw_log` for details about the filename format and `Log
3259 File Formats`_ for how data is structured within the file.
3261 .. option:: log_avg_msec=int
3263 By default, fio will log an entry in the iops, latency, or bw log for every
3264 I/O that completes. When writing to the disk log, that can quickly grow to a
3265 very large size. Setting this option makes fio average the each log entry
3266 over the specified period of time, reducing the resolution of the log. See
3267 :option:`log_max_value` as well. Defaults to 0, logging all entries.
3268 Also see `Log File Formats`_.
3270 .. option:: log_hist_msec=int
3272 Same as :option:`log_avg_msec`, but logs entries for completion latency
3273 histograms. Computing latency percentiles from averages of intervals using
3274 :option:`log_avg_msec` is inaccurate. Setting this option makes fio log
3275 histogram entries over the specified period of time, reducing log sizes for
3276 high IOPS devices while retaining percentile accuracy. See
3277 :option:`log_hist_coarseness` and :option:`write_hist_log` as well.
3278 Defaults to 0, meaning histogram logging is disabled.
3280 .. option:: log_hist_coarseness=int
3282 Integer ranging from 0 to 6, defining the coarseness of the resolution of
3283 the histogram logs enabled with :option:`log_hist_msec`. For each increment
3284 in coarseness, fio outputs half as many bins. Defaults to 0, for which
3285 histogram logs contain 1216 latency bins. See :option:`write_hist_log`
3286 and `Log File Formats`_.
3288 .. option:: log_max_value=bool
3290 If :option:`log_avg_msec` is set, fio logs the average over that window. If
3291 you instead want to log the maximum value, set this option to 1. Defaults to
3292 0, meaning that averaged values are logged.
3294 .. option:: log_offset=bool
3296 If this is set, the iolog options will include the byte offset for the I/O
3297 entry as well as the other data values. Defaults to 0 meaning that
3298 offsets are not present in logs. Also see `Log File Formats`_.
3300 .. option:: log_compression=int
3302 If this is set, fio will compress the I/O logs as it goes, to keep the
3303 memory footprint lower. When a log reaches the specified size, that chunk is
3304 removed and compressed in the background. Given that I/O logs are fairly
3305 highly compressible, this yields a nice memory savings for longer runs. The
3306 downside is that the compression will consume some background CPU cycles, so
3307 it may impact the run. This, however, is also true if the logging ends up
3308 consuming most of the system memory. So pick your poison. The I/O logs are
3309 saved normally at the end of a run, by decompressing the chunks and storing
3310 them in the specified log file. This feature depends on the availability of
3313 .. option:: log_compression_cpus=str
3315 Define the set of CPUs that are allowed to handle online log compression for
3316 the I/O jobs. This can provide better isolation between performance
3317 sensitive jobs, and background compression work. See
3318 :option:`cpus_allowed` for the format used.
3320 .. option:: log_store_compressed=bool
3322 If set, fio will store the log files in a compressed format. They can be
3323 decompressed with fio, using the :option:`--inflate-log` command line
3324 parameter. The files will be stored with a :file:`.fz` suffix.
3326 .. option:: log_unix_epoch=bool
3328 If set, fio will log Unix timestamps to the log files produced by enabling
3329 write_type_log for each log type, instead of the default zero-based
3332 .. option:: block_error_percentiles=bool
3334 If set, record errors in trim block-sized units from writes and trims and
3335 output a histogram of how many trims it took to get to errors, and what kind
3336 of error was encountered.
3338 .. option:: bwavgtime=int
3340 Average the calculated bandwidth over the given time. Value is specified in
3341 milliseconds. If the job also does bandwidth logging through
3342 :option:`write_bw_log`, then the minimum of this option and
3343 :option:`log_avg_msec` will be used. Default: 500ms.
3345 .. option:: iopsavgtime=int
3347 Average the calculated IOPS over the given time. Value is specified in
3348 milliseconds. If the job also does IOPS logging through
3349 :option:`write_iops_log`, then the minimum of this option and
3350 :option:`log_avg_msec` will be used. Default: 500ms.
3352 .. option:: disk_util=bool
3354 Generate disk utilization statistics, if the platform supports it.
3357 .. option:: disable_lat=bool
3359 Disable measurements of total latency numbers. Useful only for cutting back
3360 the number of calls to :manpage:`gettimeofday(2)`, as that does impact
3361 performance at really high IOPS rates. Note that to really get rid of a
3362 large amount of these calls, this option must be used with
3363 :option:`disable_slat` and :option:`disable_bw_measurement` as well.
3365 .. option:: disable_clat=bool
3367 Disable measurements of completion latency numbers. See
3368 :option:`disable_lat`.
3370 .. option:: disable_slat=bool
3372 Disable measurements of submission latency numbers. See
3373 :option:`disable_lat`.
3375 .. option:: disable_bw_measurement=bool, disable_bw=bool
3377 Disable measurements of throughput/bandwidth numbers. See
3378 :option:`disable_lat`.
3380 .. option:: slat_percentiles=bool
3382 Report submission latency percentiles. Submission latency is not recorded
3383 for synchronous ioengines.
3385 .. option:: clat_percentiles=bool
3387 Report completion latency percentiles.
3389 .. option:: lat_percentiles=bool
3391 Report total latency percentiles. Total latency is the sum of submission
3392 latency and completion latency.
3394 .. option:: percentile_list=float_list
3396 Overwrite the default list of percentiles for latencies and the block error
3397 histogram. Each number is a floating point number in the range (0,100], and
3398 the maximum length of the list is 20. Use ``:`` to separate the numbers. For
3399 example, ``--percentile_list=99.5:99.9`` will cause fio to report the
3400 latency durations below which 99.5% and 99.9% of the observed latencies fell,
3403 .. option:: significant_figures=int
3405 If using :option:`--output-format` of `normal`, set the significant
3406 figures to this value. Higher values will yield more precise IOPS and
3407 throughput units, while lower values will round. Requires a minimum
3408 value of 1 and a maximum value of 10. Defaults to 4.
3414 .. option:: exitall_on_error
3416 When one job finishes in error, terminate the rest. The default is to wait
3417 for each job to finish.
3419 .. option:: continue_on_error=str
3421 Normally fio will exit the job on the first observed failure. If this option
3422 is set, fio will continue the job when there is a 'non-fatal error' (EIO or
3423 EILSEQ) until the runtime is exceeded or the I/O size specified is
3424 completed. If this option is used, there are two more stats that are
3425 appended, the total error count and the first error. The error field given
3426 in the stats is the first error that was hit during the run.
3428 The allowed values are:
3431 Exit on any I/O or verify errors.
3434 Continue on read errors, exit on all others.
3437 Continue on write errors, exit on all others.
3440 Continue on any I/O error, exit on all others.
3443 Continue on verify errors, exit on all others.
3446 Continue on all errors.
3449 Backward-compatible alias for 'none'.
3452 Backward-compatible alias for 'all'.
3454 .. option:: ignore_error=str
3456 Sometimes you want to ignore some errors during test in that case you can
3457 specify error list for each error type, instead of only being able to
3458 ignore the default 'non-fatal error' using :option:`continue_on_error`.
3459 ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for
3460 given error type is separated with ':'. Error may be symbol ('ENOSPC',
3461 'ENOMEM') or integer. Example::
3463 ignore_error=EAGAIN,ENOSPC:122
3465 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from
3466 WRITE. This option works by overriding :option:`continue_on_error` with
3467 the list of errors for each error type if any.
3469 .. option:: error_dump=bool
3471 If set dump every error even if it is non fatal, true by default. If
3472 disabled only fatal error will be dumped.
3474 Running predefined workloads
3475 ----------------------------
3477 Fio includes predefined profiles that mimic the I/O workloads generated by
3480 .. option:: profile=str
3482 The predefined workload to run. Current profiles are:
3485 Threaded I/O bench (tiotest/tiobench) like workload.
3488 Aerospike Certification Tool (ACT) like workload.
3490 To view a profile's additional options use :option:`--cmdhelp` after specifying
3491 the profile. For example::
3493 $ fio --profile=act --cmdhelp
3498 .. option:: device-names=str
3503 .. option:: load=int
3506 ACT load multiplier. Default: 1.
3508 .. option:: test-duration=time
3511 How long the entire test takes to run. When the unit is omitted, the value
3512 is given in seconds. Default: 24h.
3514 .. option:: threads-per-queue=int
3517 Number of read I/O threads per device. Default: 8.
3519 .. option:: read-req-num-512-blocks=int
3522 Number of 512B blocks to read at the time. Default: 3.
3524 .. option:: large-block-op-kbytes=int
3527 Size of large block ops in KiB (writes). Default: 131072.
3532 Set to run ACT prep phase.
3534 Tiobench profile options
3535 ~~~~~~~~~~~~~~~~~~~~~~~~
3537 .. option:: size=str
3542 .. option:: block=int
3545 Block size in bytes. Default: 4096.
3547 .. option:: numruns=int
3557 .. option:: threads=int
3562 Interpreting the output
3563 -----------------------
3566 Example output was based on the following:
3567 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --time_based \
3568 --rate=1256k --bs=14K --name=quick --runtime=1s --name=mixed \
3569 --runtime=2m --rw=rw
3571 Fio spits out a lot of output. While running, fio will display the status of the
3572 jobs created. An example of that would be::
3574 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]
3576 The characters inside the first set of square brackets denote the current status of
3577 each thread. The first character is the first job defined in the job file, and so
3578 forth. The possible values (in typical life cycle order) are:
3580 +------+-----+-----------------------------------------------------------+
3582 +======+=====+===========================================================+
3583 | P | | Thread setup, but not started. |
3584 +------+-----+-----------------------------------------------------------+
3585 | C | | Thread created. |
3586 +------+-----+-----------------------------------------------------------+
3587 | I | | Thread initialized, waiting or generating necessary data. |
3588 +------+-----+-----------------------------------------------------------+
3589 | | p | Thread running pre-reading file(s). |
3590 +------+-----+-----------------------------------------------------------+
3591 | | / | Thread is in ramp period. |
3592 +------+-----+-----------------------------------------------------------+
3593 | | R | Running, doing sequential reads. |
3594 +------+-----+-----------------------------------------------------------+
3595 | | r | Running, doing random reads. |
3596 +------+-----+-----------------------------------------------------------+
3597 | | W | Running, doing sequential writes. |
3598 +------+-----+-----------------------------------------------------------+
3599 | | w | Running, doing random writes. |
3600 +------+-----+-----------------------------------------------------------+
3601 | | M | Running, doing mixed sequential reads/writes. |
3602 +------+-----+-----------------------------------------------------------+
3603 | | m | Running, doing mixed random reads/writes. |
3604 +------+-----+-----------------------------------------------------------+
3605 | | D | Running, doing sequential trims. |
3606 +------+-----+-----------------------------------------------------------+
3607 | | d | Running, doing random trims. |
3608 +------+-----+-----------------------------------------------------------+
3609 | | F | Running, currently waiting for :manpage:`fsync(2)`. |
3610 +------+-----+-----------------------------------------------------------+
3611 | | V | Running, doing verification of written data. |
3612 +------+-----+-----------------------------------------------------------+
3613 | f | | Thread finishing. |
3614 +------+-----+-----------------------------------------------------------+
3615 | E | | Thread exited, not reaped by main thread yet. |
3616 +------+-----+-----------------------------------------------------------+
3617 | _ | | Thread reaped. |
3618 +------+-----+-----------------------------------------------------------+
3619 | X | | Thread reaped, exited with an error. |
3620 +------+-----+-----------------------------------------------------------+
3621 | K | | Thread reaped, exited due to signal. |
3622 +------+-----+-----------------------------------------------------------+
3625 Example output was based on the following:
3626 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --runtime=58m \
3627 --time_based --rate=2512k --bs=256K --numjobs=10 \
3628 --name=readers --rw=read --name=writers --rw=write
3630 Fio will condense the thread string as not to take up more space on the command
3631 line than needed. For instance, if you have 10 readers and 10 writers running,
3632 the output would look like this::
3634 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]
3636 Note that the status string is displayed in order, so it's possible to tell which of
3637 the jobs are currently doing what. In the example above this means that jobs 1--10
3638 are readers and 11--20 are writers.
3640 The other values are fairly self explanatory -- number of threads currently
3641 running and doing I/O, the number of currently open files (f=), the estimated
3642 completion percentage, the rate of I/O since last check (read speed listed first,
3643 then write speed and optionally trim speed) in terms of bandwidth and IOPS,
3644 and time to completion for the current running group. It's impossible to estimate
3645 runtime of the following groups (if any).
3648 Example output was based on the following:
3649 TZ=UTC fio --iodepth=16 --ioengine=posixaio --filename=/tmp/fiofile \
3650 --direct=1 --size=100M --time_based --runtime=50s --rate_iops=89 \
3651 --bs=7K --name=Client1 --rw=write
3653 When fio is done (or interrupted by :kbd:`Ctrl-C`), it will show the data for
3654 each thread, group of threads, and disks in that order. For each overall thread (or
3655 group) the output looks like::
3657 Client1: (groupid=0, jobs=1): err= 0: pid=16109: Sat Jun 24 12:07:54 2017
3658 write: IOPS=88, BW=623KiB/s (638kB/s)(30.4MiB/50032msec)
3659 slat (nsec): min=500, max=145500, avg=8318.00, stdev=4781.50
3660 clat (usec): min=170, max=78367, avg=4019.02, stdev=8293.31
3661 lat (usec): min=174, max=78375, avg=4027.34, stdev=8291.79
3662 clat percentiles (usec):
3663 | 1.00th=[ 302], 5.00th=[ 326], 10.00th=[ 343], 20.00th=[ 363],
3664 | 30.00th=[ 392], 40.00th=[ 404], 50.00th=[ 416], 60.00th=[ 445],
3665 | 70.00th=[ 816], 80.00th=[ 6718], 90.00th=[12911], 95.00th=[21627],
3666 | 99.00th=[43779], 99.50th=[51643], 99.90th=[68682], 99.95th=[72877],
3668 bw ( KiB/s): min= 532, max= 686, per=0.10%, avg=622.87, stdev=24.82, samples= 100
3669 iops : min= 76, max= 98, avg=88.98, stdev= 3.54, samples= 100
3670 lat (usec) : 250=0.04%, 500=64.11%, 750=4.81%, 1000=2.79%
3671 lat (msec) : 2=4.16%, 4=1.84%, 10=4.90%, 20=11.33%, 50=5.37%
3672 lat (msec) : 100=0.65%
3673 cpu : usr=0.27%, sys=0.18%, ctx=12072, majf=0, minf=21
3674 IO depths : 1=85.0%, 2=13.1%, 4=1.8%, 8=0.1%, 16=0.0%, 32=0.0%, >=64=0.0%
3675 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3676 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3677 issued rwt: total=0,4450,0, short=0,0,0, dropped=0,0,0
3678 latency : target=0, window=0, percentile=100.00%, depth=8
3680 The job name (or first job's name when using :option:`group_reporting`) is printed,
3681 along with the group id, count of jobs being aggregated, last error id seen (which
3682 is 0 when there are no errors), pid/tid of that thread and the time the job/group
3683 completed. Below are the I/O statistics for each data direction performed (showing
3684 writes in the example above). In the order listed, they denote:
3687 The string before the colon shows the I/O direction the statistics
3688 are for. **IOPS** is the average I/Os performed per second. **BW**
3689 is the average bandwidth rate shown as: value in power of 2 format
3690 (value in power of 10 format). The last two values show: (**total
3691 I/O performed** in power of 2 format / **runtime** of that thread).
3694 Submission latency (**min** being the minimum, **max** being the
3695 maximum, **avg** being the average, **stdev** being the standard
3696 deviation). This is the time it took to submit the I/O. For
3697 sync I/O this row is not displayed as the slat is really the
3698 completion latency (since queue/complete is one operation there).
3699 This value can be in nanoseconds, microseconds or milliseconds ---
3700 fio will choose the most appropriate base and print that (in the
3701 example above nanoseconds was the best scale). Note: in :option:`--minimal` mode
3702 latencies are always expressed in microseconds.
3705 Completion latency. Same names as slat, this denotes the time from
3706 submission to completion of the I/O pieces. For sync I/O, clat will
3707 usually be equal (or very close) to 0, as the time from submit to
3708 complete is basically just CPU time (I/O has already been done, see slat
3712 Total latency. Same names as slat and clat, this denotes the time from
3713 when fio created the I/O unit to completion of the I/O operation.
3716 Bandwidth statistics based on samples. Same names as the xlat stats,
3717 but also includes the number of samples taken (**samples**) and an
3718 approximate percentage of total aggregate bandwidth this thread
3719 received in its group (**per**). This last value is only really
3720 useful if the threads in this group are on the same disk, since they
3721 are then competing for disk access.
3724 IOPS statistics based on samples. Same names as bw.
3726 **lat (nsec/usec/msec)**
3727 The distribution of I/O completion latencies. This is the time from when
3728 I/O leaves fio and when it gets completed. Unlike the separate
3729 read/write/trim sections above, the data here and in the remaining
3730 sections apply to all I/Os for the reporting group. 250=0.04% means that
3731 0.04% of the I/Os completed in under 250us. 500=64.11% means that 64.11%
3732 of the I/Os required 250 to 499us for completion.
3735 CPU usage. User and system time, along with the number of context
3736 switches this thread went through, usage of system and user time, and
3737 finally the number of major and minor page faults. The CPU utilization
3738 numbers are averages for the jobs in that reporting group, while the
3739 context and fault counters are summed.
3742 The distribution of I/O depths over the job lifetime. The numbers are
3743 divided into powers of 2 and each entry covers depths from that value
3744 up to those that are lower than the next entry -- e.g., 16= covers
3745 depths from 16 to 31. Note that the range covered by a depth
3746 distribution entry can be different to the range covered by the
3747 equivalent submit/complete distribution entry.
3750 How many pieces of I/O were submitting in a single submit call. Each
3751 entry denotes that amount and below, until the previous entry -- e.g.,
3752 16=100% means that we submitted anywhere between 9 to 16 I/Os per submit
3753 call. Note that the range covered by a submit distribution entry can
3754 be different to the range covered by the equivalent depth distribution
3758 Like the above submit number, but for completions instead.
3761 The number of read/write/trim requests issued, and how many of them were
3765 These values are for :option:`latency_target` and related options. When
3766 these options are engaged, this section describes the I/O depth required
3767 to meet the specified latency target.
3770 Example output was based on the following:
3771 TZ=UTC fio --ioengine=null --iodepth=2 --size=100M --numjobs=2 \
3772 --rate_process=poisson --io_limit=32M --name=read --bs=128k \
3773 --rate=11M --name=write --rw=write --bs=2k --rate=700k
3775 After each client has been listed, the group statistics are printed. They
3776 will look like this::
3778 Run status group 0 (all jobs):
3779 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
3780 WRITE: bw=1231KiB/s (1261kB/s), 616KiB/s-621KiB/s (630kB/s-636kB/s), io=64.0MiB (67.1MB), run=52747-53223msec
3782 For each data direction it prints:
3785 Aggregate bandwidth of threads in this group followed by the
3786 minimum and maximum bandwidth of all the threads in this group.
3787 Values outside of brackets are power-of-2 format and those
3788 within are the equivalent value in a power-of-10 format.
3790 Aggregate I/O performed of all threads in this group. The
3791 format is the same as bw.
3793 The smallest and longest runtimes of the threads in this group.
3795 And finally, the disk statistics are printed. This is Linux specific. They will look like this::
3797 Disk stats (read/write):
3798 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
3800 Each value is printed for both reads and writes, with reads first. The
3804 Number of I/Os performed by all groups.
3806 Number of merges performed by the I/O scheduler.
3808 Number of ticks we kept the disk busy.
3810 Total time spent in the disk queue.
3812 The disk utilization. A value of 100% means we kept the disk
3813 busy constantly, 50% would be a disk idling half of the time.
3815 It is also possible to get fio to dump the current output while it is running,
3816 without terminating the job. To do that, send fio the **USR1** signal. You can
3817 also get regularly timed dumps by using the :option:`--status-interval`
3818 parameter, or by creating a file in :file:`/tmp` named
3819 :file:`fio-dump-status`. If fio sees this file, it will unlink it and dump the
3820 current output status.
3826 For scripted usage where you typically want to generate tables or graphs of the
3827 results, fio can output the results in a semicolon separated format. The format
3828 is one long line of values, such as::
3830 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%
3831 A description of this job goes here.
3833 The job description (if provided) follows on a second line for terse v2.
3834 It appears on the same line for other terse versions.
3836 To enable terse output, use the :option:`--minimal` or
3837 :option:`--output-format`\=terse command line options. The
3838 first value is the version of the terse output format. If the output has to be
3839 changed for some reason, this number will be incremented by 1 to signify that
3842 Split up, the format is as follows (comments in brackets denote when a
3843 field was introduced or whether it's specific to some terse version):
3847 terse version, fio version [v3], jobname, groupid, error
3851 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3852 Submission latency: min, max, mean, stdev (usec)
3853 Completion latency: min, max, mean, stdev (usec)
3854 Completion latency percentiles: 20 fields (see below)
3855 Total latency: min, max, mean, stdev (usec)
3856 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3857 IOPS [v5]: min, max, mean, stdev, number of samples
3863 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3864 Submission latency: min, max, mean, stdev (usec)
3865 Completion latency: min, max, mean, stdev (usec)
3866 Completion latency percentiles: 20 fields (see below)
3867 Total latency: min, max, mean, stdev (usec)
3868 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3869 IOPS [v5]: min, max, mean, stdev, number of samples
3871 TRIM status [all but version 3]:
3873 Fields are similar to READ/WRITE status.
3877 user, system, context switches, major faults, minor faults
3881 <=1, 2, 4, 8, 16, 32, >=64
3883 I/O latencies microseconds::
3885 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
3887 I/O latencies milliseconds::
3889 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
3891 Disk utilization [v3]::
3893 disk name, read ios, write ios, read merges, write merges, read ticks, write ticks,
3894 time spent in queue, disk utilization percentage
3896 Additional Info (dependent on continue_on_error, default off)::
3898 total # errors, first error code
3900 Additional Info (dependent on description being set)::
3904 Completion latency percentiles can be a grouping of up to 20 sets, so for the
3905 terse output fio writes all of them. Each field will look like this::
3909 which is the Xth percentile, and the `usec` latency associated with it.
3911 For `Disk utilization`, all disks used by fio are shown. So for each disk there
3912 will be a disk utilization section.
3914 Below is a single line containing short names for each of the fields in the
3915 minimal output v3, separated by semicolons::
3917 terse_version_3;fio_version;jobname;groupid;error;read_kb;read_bandwidth;read_iops;read_runtime_ms;read_slat_min;read_slat_max;read_slat_mean;read_slat_dev;read_clat_min;read_clat_max;read_clat_mean;read_clat_dev;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;read_lat_max;read_lat_mean;read_lat_dev;read_bw_min;read_bw_max;read_bw_agg_pct;read_bw_mean;read_bw_dev;write_kb;write_bandwidth;write_iops;write_runtime_ms;write_slat_min;write_slat_max;write_slat_mean;write_slat_dev;write_clat_min;write_clat_max;write_clat_mean;write_clat_dev;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;write_lat_max;write_lat_mean;write_lat_dev;write_bw_min;write_bw_max;write_bw_agg_pct;write_bw_mean;write_bw_dev;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
3919 In client/server mode terse output differs from what appears when jobs are run
3920 locally. Disk utilization data is omitted from the standard terse output and
3921 for v3 and later appears on its own separate line at the end of each terse
3928 The `json` output format is intended to be both human readable and convenient
3929 for automated parsing. For the most part its sections mirror those of the
3930 `normal` output. The `runtime` value is reported in msec and the `bw` value is
3931 reported in 1024 bytes per second units.
3937 The `json+` output format is identical to the `json` output format except that it
3938 adds a full dump of the completion latency bins. Each `bins` object contains a
3939 set of (key, value) pairs where keys are latency durations and values count how
3940 many I/Os had completion latencies of the corresponding duration. For example,
3943 "bins" : { "87552" : 1, "89600" : 1, "94720" : 1, "96768" : 1, "97792" : 1, "99840" : 1, "100864" : 2, "103936" : 6, "104960" : 534, "105984" : 5995, "107008" : 7529, ... }
3945 This data indicates that one I/O required 87,552ns to complete, two I/Os required
3946 100,864ns to complete, and 7529 I/Os required 107,008ns to complete.
3948 Also included with fio is a Python script `fio_jsonplus_clat2csv` that takes
3949 json+ output and generates CSV-formatted latency data suitable for plotting.
3951 The latency durations actually represent the midpoints of latency intervals.
3952 For details refer to :file:`stat.h`.
3958 There are two trace file format that you can encounter. The older (v1) format is
3959 unsupported since version 1.20-rc3 (March 2008). It will still be described
3960 below in case that you get an old trace and want to understand it.
3962 In any case the trace is a simple text file with a single action per line.
3965 Trace file format v1
3966 ~~~~~~~~~~~~~~~~~~~~
3968 Each line represents a single I/O action in the following format::
3972 where `rw=0/1` for read/write, and the `offset` and `length` entries being in bytes.
3974 This format is not supported in fio versions >= 1.20-rc3.
3977 Trace file format v2
3978 ~~~~~~~~~~~~~~~~~~~~
3980 The second version of the trace file format was added in fio version 1.17. It
3981 allows to access more then one file per trace and has a bigger set of possible
3984 The first line of the trace file has to be::
3988 Following this can be lines in two different formats, which are described below.
3990 The file management format::
3994 The `filename` is given as an absolute path. The `action` can be one of these:
3997 Add the given `filename` to the trace.
3999 Open the file with the given `filename`. The `filename` has to have
4000 been added with the **add** action before.
4002 Close the file with the given `filename`. The file has to have been
4006 The file I/O action format::
4008 filename action offset length
4010 The `filename` is given as an absolute path, and has to have been added and
4011 opened before it can be used with this format. The `offset` and `length` are
4012 given in bytes. The `action` can be one of these:
4015 Wait for `offset` microseconds. Everything below 100 is discarded.
4016 The time is relative to the previous `wait` statement.
4018 Read `length` bytes beginning from `offset`.
4020 Write `length` bytes beginning from `offset`.
4022 :manpage:`fsync(2)` the file.
4024 :manpage:`fdatasync(2)` the file.
4026 Trim the given file from the given `offset` for `length` bytes.
4029 I/O Replay - Merging Traces
4030 ---------------------------
4032 Colocation is a common practice used to get the most out of a machine.
4033 Knowing which workloads play nicely with each other and which ones don't is
4034 a much harder task. While fio can replay workloads concurrently via multiple
4035 jobs, it leaves some variability up to the scheduler making results harder to
4036 reproduce. Merging is a way to make the order of events consistent.
4038 Merging is integrated into I/O replay and done when a
4039 :option:`merge_blktrace_file` is specified. The list of files passed to
4040 :option:`read_iolog` go through the merge process and output a single file
4041 stored to the specified file. The output file is passed on as if it were the
4042 only file passed to :option:`read_iolog`. An example would look like::
4044 $ fio --read_iolog="<file1>:<file2>" --merge_blktrace_file="<output_file>"
4046 Creating only the merged file can be done by passing the command line argument
4047 :option:`--merge-blktrace-only`.
4049 Scaling traces can be done to see the relative impact of any particular trace
4050 being slowed down or sped up. :option:`merge_blktrace_scalars` takes in a colon
4051 separated list of percentage scalars. It is index paired with the files passed
4052 to :option:`read_iolog`.
4054 With scaling, it may be desirable to match the running time of all traces.
4055 This can be done with :option:`merge_blktrace_iters`. It is index paired with
4056 :option:`read_iolog` just like :option:`merge_blktrace_scalars`.
4058 In an example, given two traces, A and B, each 60s long. If we want to see
4059 the impact of trace A issuing IOs twice as fast and repeat trace A over the
4060 runtime of trace B, the following can be done::
4062 $ fio --read_iolog="<trace_a>:"<trace_b>" --merge_blktrace_file"<output_file>" --merge_blktrace_scalars="50:100" --merge_blktrace_iters="2:1"
4064 This runs trace A at 2x the speed twice for approximately the same runtime as
4065 a single run of trace B.
4068 CPU idleness profiling
4069 ----------------------
4071 In some cases, we want to understand CPU overhead in a test. For example, we
4072 test patches for the specific goodness of whether they reduce CPU usage.
4073 Fio implements a balloon approach to create a thread per CPU that runs at idle
4074 priority, meaning that it only runs when nobody else needs the cpu.
4075 By measuring the amount of work completed by the thread, idleness of each CPU
4076 can be derived accordingly.
4078 An unit work is defined as touching a full page of unsigned characters. Mean and
4079 standard deviation of time to complete an unit work is reported in "unit work"
4080 section. Options can be chosen to report detailed percpu idleness or overall
4081 system idleness by aggregating percpu stats.
4084 Verification and triggers
4085 -------------------------
4087 Fio is usually run in one of two ways, when data verification is done. The first
4088 is a normal write job of some sort with verify enabled. When the write phase has
4089 completed, fio switches to reads and verifies everything it wrote. The second
4090 model is running just the write phase, and then later on running the same job
4091 (but with reads instead of writes) to repeat the same I/O patterns and verify
4092 the contents. Both of these methods depend on the write phase being completed,
4093 as fio otherwise has no idea how much data was written.
4095 With verification triggers, fio supports dumping the current write state to
4096 local files. Then a subsequent read verify workload can load this state and know
4097 exactly where to stop. This is useful for testing cases where power is cut to a
4098 server in a managed fashion, for instance.
4100 A verification trigger consists of two things:
4102 1) Storing the write state of each job.
4103 2) Executing a trigger command.
4105 The write state is relatively small, on the order of hundreds of bytes to single
4106 kilobytes. It contains information on the number of completions done, the last X
4109 A trigger is invoked either through creation ('touch') of a specified file in
4110 the system, or through a timeout setting. If fio is run with
4111 :option:`--trigger-file`\= :file:`/tmp/trigger-file`, then it will continually
4112 check for the existence of :file:`/tmp/trigger-file`. When it sees this file, it
4113 will fire off the trigger (thus saving state, and executing the trigger
4116 For client/server runs, there's both a local and remote trigger. If fio is
4117 running as a server backend, it will send the job states back to the client for
4118 safe storage, then execute the remote trigger, if specified. If a local trigger
4119 is specified, the server will still send back the write state, but the client
4120 will then execute the trigger.
4122 Verification trigger example
4123 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4125 Let's say we want to run a powercut test on the remote Linux machine 'server'.
4126 Our write workload is in :file:`write-test.fio`. We want to cut power to 'server' at
4127 some point during the run, and we'll run this test from the safety or our local
4128 machine, 'localbox'. On the server, we'll start the fio backend normally::
4130 server# fio --server
4132 and on the client, we'll fire off the workload::
4134 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger-remote="bash -c \"echo b > /proc/sysrq-triger\""
4136 We set :file:`/tmp/my-trigger` as the trigger file, and we tell fio to execute::
4138 echo b > /proc/sysrq-trigger
4140 on the server once it has received the trigger and sent us the write state. This
4141 will work, but it's not **really** cutting power to the server, it's merely
4142 abruptly rebooting it. If we have a remote way of cutting power to the server
4143 through IPMI or similar, we could do that through a local trigger command
4144 instead. Let's assume we have a script that does IPMI reboot of a given hostname,
4145 ipmi-reboot. On localbox, we could then have run fio with a local trigger
4148 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger="ipmi-reboot server"
4150 For this case, fio would wait for the server to send us the write state, then
4151 execute ``ipmi-reboot server`` when that happened.
4153 Loading verify state
4154 ~~~~~~~~~~~~~~~~~~~~
4156 To load stored write state, a read verification job file must contain the
4157 :option:`verify_state_load` option. If that is set, fio will load the previously
4158 stored state. For a local fio run this is done by loading the files directly,
4159 and on a client/server run, the server backend will ask the client to send the
4160 files over and load them from there.
4166 Fio supports a variety of log file formats, for logging latencies, bandwidth,
4167 and IOPS. The logs share a common format, which looks like this:
4169 *time* (`msec`), *value*, *data direction*, *block size* (`bytes`),
4170 *offset* (`bytes`), *command priority*
4172 *Time* for the log entry is always in milliseconds. The *value* logged depends
4173 on the type of log, it will be one of the following:
4176 Value is latency in nsecs
4182 *Data direction* is one of the following:
4191 The entry's *block size* is always in bytes. The *offset* is the position in bytes
4192 from the start of the file for that particular I/O. The logging of the offset can be
4193 toggled with :option:`log_offset`.
4195 *Command priority* is 0 for normal priority and 1 for high priority. This is controlled
4196 by the ioengine specific :option:`cmdprio_percentage`.
4198 Fio defaults to logging every individual I/O but when windowed logging is set
4199 through :option:`log_avg_msec`, either the average (by default) or the maximum
4200 (:option:`log_max_value` is set) *value* seen over the specified period of time
4201 is recorded. Each *data direction* seen within the window period will aggregate
4202 its values in a separate row. Further, when using windowed logging the *block
4203 size* and *offset* entries will always contain 0.
4209 Normally fio is invoked as a stand-alone application on the machine where the
4210 I/O workload should be generated. However, the backend and frontend of fio can
4211 be run separately i.e., the fio server can generate an I/O workload on the "Device
4212 Under Test" while being controlled by a client on another machine.
4214 Start the server on the machine which has access to the storage DUT::
4218 where `args` defines what fio listens to. The arguments are of the form
4219 ``type,hostname`` or ``IP,port``. *type* is either ``ip`` (or ip4) for TCP/IP
4220 v4, ``ip6`` for TCP/IP v6, or ``sock`` for a local unix domain socket.
4221 *hostname* is either a hostname or IP address, and *port* is the port to listen
4222 to (only valid for TCP/IP, not a local socket). Some examples:
4226 Start a fio server, listening on all interfaces on the default port (8765).
4228 2) ``fio --server=ip:hostname,4444``
4230 Start a fio server, listening on IP belonging to hostname and on port 4444.
4232 3) ``fio --server=ip6:::1,4444``
4234 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
4236 4) ``fio --server=,4444``
4238 Start a fio server, listening on all interfaces on port 4444.
4240 5) ``fio --server=1.2.3.4``
4242 Start a fio server, listening on IP 1.2.3.4 on the default port.
4244 6) ``fio --server=sock:/tmp/fio.sock``
4246 Start a fio server, listening on the local socket :file:`/tmp/fio.sock`.
4248 Once a server is running, a "client" can connect to the fio server with::
4250 fio <local-args> --client=<server> <remote-args> <job file(s)>
4252 where `local-args` are arguments for the client where it is running, `server`
4253 is the connect string, and `remote-args` and `job file(s)` are sent to the
4254 server. The `server` string follows the same format as it does on the server
4255 side, to allow IP/hostname/socket and port strings.
4257 Fio can connect to multiple servers this way::
4259 fio --client=<server1> <job file(s)> --client=<server2> <job file(s)>
4261 If the job file is located on the fio server, then you can tell the server to
4262 load a local file as well. This is done by using :option:`--remote-config` ::
4264 fio --client=server --remote-config /path/to/file.fio
4266 Then fio will open this local (to the server) job file instead of being passed
4267 one from the client.
4269 If you have many servers (example: 100 VMs/containers), you can input a pathname
4270 of a file containing host IPs/names as the parameter value for the
4271 :option:`--client` option. For example, here is an example :file:`host.list`
4272 file containing 2 hostnames::
4274 host1.your.dns.domain
4275 host2.your.dns.domain
4277 The fio command would then be::
4279 fio --client=host.list <job file(s)>
4281 In this mode, you cannot input server-specific parameters or job files -- all
4282 servers receive the same job file.
4284 In order to let ``fio --client`` runs use a shared filesystem from multiple
4285 hosts, ``fio --client`` now prepends the IP address of the server to the
4286 filename. For example, if fio is using the directory :file:`/mnt/nfs/fio` and is
4287 writing filename :file:`fileio.tmp`, with a :option:`--client` `hostfile`
4288 containing two hostnames ``h1`` and ``h2`` with IP addresses 192.168.10.120 and
4289 192.168.10.121, then fio will create two files::
4291 /mnt/nfs/fio/192.168.10.120.fileio.tmp
4292 /mnt/nfs/fio/192.168.10.121.fileio.tmp
4294 Terse output in client/server mode will differ slightly from what is produced
4295 when fio is run in stand-alone mode. See the terse output section for details.