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 Show available debug options.
99 .. option:: --parse-only
101 Parse options only, don't start any I/O.
103 .. option:: --output=filename
105 Write output to file `filename`.
107 .. option:: --output-format=format
109 Set the reporting `format` to `normal`, `terse`, `json`, or `json+`. Multiple
110 formats can be selected, separated by a comma. `terse` is a CSV based
111 format. `json+` is like `json`, except it adds a full dump of the latency
114 .. option:: --bandwidth-log
116 Generate aggregate bandwidth logs.
118 .. option:: --minimal
120 Print statistics in a terse, semicolon-delimited format.
122 .. option:: --append-terse
124 Print statistics in selected mode AND terse, semicolon-delimited format.
125 **Deprecated**, use :option:`--output-format` instead to select multiple
128 .. option:: --terse-version=version
130 Set terse `version` output format (default 3, or 2 or 4 or 5).
132 .. option:: --version
134 Print version information and exit.
138 Print a summary of the command line options and exit.
140 .. option:: --cpuclock-test
142 Perform test and validation of internal CPU clock.
144 .. option:: --crctest=[test]
146 Test the speed of the built-in checksumming functions. If no argument is
147 given, all of them are tested. Alternatively, a comma separated list can
148 be passed, in which case the given ones are tested.
150 .. option:: --cmdhelp=command
152 Print help information for `command`. May be ``all`` for all commands.
154 .. option:: --enghelp=[ioengine[,command]]
156 List all commands defined by `ioengine`, or print help for `command`
157 defined by `ioengine`. If no `ioengine` is given, list all
160 .. option:: --showcmd=jobfile
162 Convert `jobfile` to a set of command-line options.
164 .. option:: --readonly
166 Turn on safety read-only checks, preventing writes. The ``--readonly``
167 option is an extra safety guard to prevent users from accidentally starting
168 a write workload when that is not desired. Fio will only write if
169 `rw=write/randwrite/rw/randrw` is given. This extra safety net can be used
170 as an extra precaution as ``--readonly`` will also enable a write check in
171 the I/O engine core to prevent writes due to unknown user space bug(s).
173 .. option:: --eta=when
175 Specifies when real-time ETA estimate should be printed. `when` may be
176 `always`, `never` or `auto`.
178 .. option:: --eta-newline=time
180 Force a new line for every `time` period passed. When the unit is omitted,
181 the value is interpreted in seconds.
183 .. option:: --status-interval=time
185 Force a full status dump of cumulative (from job start) values at `time`
186 intervals. This option does *not* provide per-period measurements. So
187 values such as bandwidth are running averages. When the time unit is omitted,
188 `time` is interpreted in seconds.
190 .. option:: --section=name
192 Only run specified section `name` in job file. Multiple sections can be specified.
193 The ``--section`` option allows one to combine related jobs into one file.
194 E.g. one job file could define light, moderate, and heavy sections. Tell
195 fio to run only the "heavy" section by giving ``--section=heavy``
196 command line option. One can also specify the "write" operations in one
197 section and "verify" operation in another section. The ``--section`` option
198 only applies to job sections. The reserved *global* section is always
201 .. option:: --alloc-size=kb
203 Set the internal smalloc pool size to `kb` in KiB. The
204 ``--alloc-size`` switch allows one to use a larger pool size for smalloc.
205 If running large jobs with randommap enabled, fio can run out of memory.
206 Smalloc is an internal allocator for shared structures from a fixed size
207 memory pool and can grow to 16 pools. The pool size defaults to 16MiB.
209 NOTE: While running :file:`.fio_smalloc.*` backing store files are visible
212 .. option:: --warnings-fatal
214 All fio parser warnings are fatal, causing fio to exit with an
217 .. option:: --max-jobs=nr
219 Set the maximum number of threads/processes to support to `nr`.
220 NOTE: On Linux, it may be necessary to increase the shared-memory
221 limit (:file:`/proc/sys/kernel/shmmax`) if fio runs into errors while
224 .. option:: --server=args
226 Start a backend server, with `args` specifying what to listen to.
227 See `Client/Server`_ section.
229 .. option:: --daemonize=pidfile
231 Background a fio server, writing the pid to the given `pidfile` file.
233 .. option:: --client=hostname
235 Instead of running the jobs locally, send and run them on the given `hostname`
236 or set of `hostname`\s. See `Client/Server`_ section.
238 .. option:: --remote-config=file
240 Tell fio server to load this local `file`.
242 .. option:: --idle-prof=option
244 Report CPU idleness. `option` is one of the following:
247 Run unit work calibration only and exit.
250 Show aggregate system idleness and unit work.
253 As **system** but also show per CPU idleness.
255 .. option:: --inflate-log=log
257 Inflate and output compressed `log`.
259 .. option:: --trigger-file=file
261 Execute trigger command when `file` exists.
263 .. option:: --trigger-timeout=time
265 Execute trigger at this `time`.
267 .. option:: --trigger=command
269 Set this `command` as local trigger.
271 .. option:: --trigger-remote=command
273 Set this `command` as remote trigger.
275 .. option:: --aux-path=path
277 Use this `path` for fio state generated files.
279 Any parameters following the options will be assumed to be job files, unless
280 they match a job file parameter. Multiple job files can be listed and each job
281 file will be regarded as a separate group. Fio will :option:`stonewall`
282 execution between each group.
288 As previously described, fio accepts one or more job files describing what it is
289 supposed to do. The job file format is the classic ini file, where the names
290 enclosed in [] brackets define the job name. You are free to use any ASCII name
291 you want, except *global* which has special meaning. Following the job name is
292 a sequence of zero or more parameters, one per line, that define the behavior of
293 the job. If the first character in a line is a ';' or a '#', the entire line is
294 discarded as a comment.
296 A *global* section sets defaults for the jobs described in that file. A job may
297 override a *global* section parameter, and a job file may even have several
298 *global* sections if so desired. A job is only affected by a *global* section
301 The :option:`--cmdhelp` option also lists all options. If used with a `command`
302 argument, :option:`--cmdhelp` will detail the given `command`.
304 See the `examples/` directory for inspiration on how to write job files. Note
305 the copyright and license requirements currently apply to `examples/` files.
307 So let's look at a really simple job file that defines two processes, each
308 randomly reading from a 128MiB file:
312 ; -- start job file --
323 As you can see, the job file sections themselves are empty as all the described
324 parameters are shared. As no :option:`filename` option is given, fio makes up a
325 `filename` for each of the jobs as it sees fit. On the command line, this job
326 would look as follows::
328 $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
331 Let's look at an example that has a number of processes writing randomly to
336 ; -- start job file --
347 Here we have no *global* section, as we only have one job defined anyway. We
348 want to use async I/O here, with a depth of 4 for each file. We also increased
349 the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical
350 jobs. The result is 4 processes each randomly writing to their own 64MiB
351 file. Instead of using the above job file, you could have given the parameters
352 on the command line. For this case, you would specify::
354 $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
356 When fio is utilized as a basis of any reasonably large test suite, it might be
357 desirable to share a set of standardized settings across multiple job files.
358 Instead of copy/pasting such settings, any section may pull in an external
359 :file:`filename.fio` file with *include filename* directive, as in the following
362 ; -- start job file including.fio --
366 include glob-include.fio
373 include test-include.fio
374 ; -- end job file including.fio --
378 ; -- start job file glob-include.fio --
381 ; -- end job file glob-include.fio --
385 ; -- start job file test-include.fio --
388 ; -- end job file test-include.fio --
390 Settings pulled into a section apply to that section only (except *global*
391 section). Include directives may be nested in that any included file may contain
392 further include directive(s). Include files may not contain [] sections.
395 Environment variables
396 ~~~~~~~~~~~~~~~~~~~~~
398 Fio also supports environment variable expansion in job files. Any sub-string of
399 the form ``${VARNAME}`` as part of an option value (in other words, on the right
400 of the '='), will be expanded to the value of the environment variable called
401 `VARNAME`. If no such environment variable is defined, or `VARNAME` is the
402 empty string, the empty string will be substituted.
404 As an example, let's look at a sample fio invocation and job file::
406 $ SIZE=64m NUMJOBS=4 fio jobfile.fio
410 ; -- start job file --
417 This will expand to the following equivalent job file at runtime:
421 ; -- start job file --
428 Fio ships with a few example job files, you can also look there for inspiration.
433 Additionally, fio has a set of reserved keywords that will be replaced
434 internally with the appropriate value. Those keywords are:
438 The architecture page size of the running system.
442 Megabytes of total memory in the system.
446 Number of online available CPUs.
448 These can be used on the command line or in the job file, and will be
449 automatically substituted with the current system values when the job is
450 run. Simple math is also supported on these keywords, so you can perform actions
455 and get that properly expanded to 8 times the size of memory in the machine.
461 This section describes in details each parameter associated with a job. Some
462 parameters take an option of a given type, such as an integer or a
463 string. Anywhere a numeric value is required, an arithmetic expression may be
464 used, provided it is surrounded by parentheses. Supported operators are:
473 For time values in expressions, units are microseconds by default. This is
474 different than for time values not in expressions (not enclosed in
475 parentheses). The following types are used:
482 String: A sequence of alphanumeric characters.
485 Integer with possible time suffix. Without a unit value is interpreted as
486 seconds unless otherwise specified. Accepts a suffix of 'd' for days, 'h' for
487 hours, 'm' for minutes, 's' for seconds, 'ms' (or 'msec') for milliseconds and
488 'us' (or 'usec') for microseconds. For example, use 10m for 10 minutes.
493 Integer. A whole number value, which may contain an integer prefix
494 and an integer suffix:
496 [*integer prefix*] **number** [*integer suffix*]
498 The optional *integer prefix* specifies the number's base. The default
499 is decimal. *0x* specifies hexadecimal.
501 The optional *integer suffix* specifies the number's units, and includes an
502 optional unit prefix and an optional unit. For quantities of data, the
503 default unit is bytes. For quantities of time, the default unit is seconds
504 unless otherwise specified.
506 With :option:`kb_base`\=1000, fio follows international standards for unit
507 prefixes. To specify power-of-10 decimal values defined in the
508 International System of Units (SI):
510 * *K* -- means kilo (K) or 1000
511 * *M* -- means mega (M) or 1000**2
512 * *G* -- means giga (G) or 1000**3
513 * *T* -- means tera (T) or 1000**4
514 * *P* -- means peta (P) or 1000**5
516 To specify power-of-2 binary values defined in IEC 80000-13:
518 * *Ki* -- means kibi (Ki) or 1024
519 * *Mi* -- means mebi (Mi) or 1024**2
520 * *Gi* -- means gibi (Gi) or 1024**3
521 * *Ti* -- means tebi (Ti) or 1024**4
522 * *Pi* -- means pebi (Pi) or 1024**5
524 With :option:`kb_base`\=1024 (the default), the unit prefixes are opposite
525 from those specified in the SI and IEC 80000-13 standards to provide
526 compatibility with old scripts. For example, 4k means 4096.
528 For quantities of data, an optional unit of 'B' may be included
529 (e.g., 'kB' is the same as 'k').
531 The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega,
532 not milli). 'b' and 'B' both mean byte, not bit.
534 Examples with :option:`kb_base`\=1000:
536 * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB
537 * *1 MiB*: 1048576, 1mi, 1024ki
538 * *1 MB*: 1000000, 1m, 1000k
539 * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi
540 * *1 TB*: 1000000000, 1t, 1000m, 1000000k
542 Examples with :option:`kb_base`\=1024 (default):
544 * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
545 * *1 MiB*: 1048576, 1m, 1024k
546 * *1 MB*: 1000000, 1mi, 1000ki
547 * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m
548 * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki
550 To specify times (units are not case sensitive):
554 * *M* -- means minutes
555 * *s* -- or sec means seconds (default)
556 * *ms* -- or *msec* means milliseconds
557 * *us* -- or *usec* means microseconds
559 If the option accepts an upper and lower range, use a colon ':' or
560 minus '-' to separate such values. See :ref:`irange <irange>`.
561 If the lower value specified happens to be larger than the upper value
562 the two values are swapped.
567 Boolean. Usually parsed as an integer, however only defined for
568 true and false (1 and 0).
573 Integer range with suffix. Allows value range to be given, such as
574 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the
575 option allows two sets of ranges, they can be specified with a ',' or '/'
576 delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`.
579 A list of floating point numbers, separated by a ':' character.
581 With the above in mind, here follows the complete list of fio job parameters.
587 .. option:: kb_base=int
589 Select the interpretation of unit prefixes in input parameters.
592 Inputs comply with IEC 80000-13 and the International
593 System of Units (SI). Use:
595 - power-of-2 values with IEC prefixes (e.g., KiB)
596 - power-of-10 values with SI prefixes (e.g., kB)
599 Compatibility mode (default). To avoid breaking old scripts:
601 - power-of-2 values with SI prefixes
602 - power-of-10 values with IEC prefixes
604 See :option:`bs` for more details on input parameters.
606 Outputs always use correct prefixes. Most outputs include both
609 bw=2383.3kB/s (2327.4KiB/s)
611 If only one value is reported, then kb_base selects the one to use:
613 **1000** -- SI prefixes
615 **1024** -- IEC prefixes
617 .. option:: unit_base=int
619 Base unit for reporting. Allowed values are:
622 Use auto-detection (default).
634 ASCII name of the job. This may be used to override the name printed by fio
635 for this job. Otherwise the job name is used. On the command line this
636 parameter has the special purpose of also signaling the start of a new job.
638 .. option:: description=str
640 Text description of the job. Doesn't do anything except dump this text
641 description when this job is run. It's not parsed.
643 .. option:: loops=int
645 Run the specified number of iterations of this job. Used to repeat the same
646 workload a given number of times. Defaults to 1.
648 .. option:: numjobs=int
650 Create the specified number of clones of this job. Each clone of job
651 is spawned as an independent thread or process. May be used to setup a
652 larger number of threads/processes doing the same thing. Each thread is
653 reported separately; to see statistics for all clones as a whole, use
654 :option:`group_reporting` in conjunction with :option:`new_group`.
655 See :option:`--max-jobs`. Default: 1.
658 Time related parameters
659 ~~~~~~~~~~~~~~~~~~~~~~~
661 .. option:: runtime=time
663 Tell fio to terminate processing after the specified period of time. It
664 can be quite hard to determine for how long a specified job will run, so
665 this parameter is handy to cap the total runtime to a given time. When
666 the unit is omitted, the value is intepreted in seconds.
668 .. option:: time_based
670 If set, fio will run for the duration of the :option:`runtime` specified
671 even if the file(s) are completely read or written. It will simply loop over
672 the same workload as many times as the :option:`runtime` allows.
674 .. option:: startdelay=irange(time)
676 Delay the start of job for the specified amount of time. Can be a single
677 value or a range. When given as a range, each thread will choose a value
678 randomly from within the range. Value is in seconds if a unit is omitted.
680 .. option:: ramp_time=time
682 If set, fio will run the specified workload for this amount of time before
683 logging any performance numbers. Useful for letting performance settle
684 before logging results, thus minimizing the runtime required for stable
685 results. Note that the ``ramp_time`` is considered lead in time for a job,
686 thus it will increase the total runtime if a special timeout or
687 :option:`runtime` is specified. When the unit is omitted, the value is
690 .. option:: clocksource=str
692 Use the given clocksource as the base of timing. The supported options are:
695 :manpage:`gettimeofday(2)`
698 :manpage:`clock_gettime(2)`
701 Internal CPU clock source
703 cpu is the preferred clocksource if it is reliable, as it is very fast (and
704 fio is heavy on time calls). Fio will automatically use this clocksource if
705 it's supported and considered reliable on the system it is running on,
706 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
707 means supporting TSC Invariant.
709 .. option:: gtod_reduce=bool
711 Enable all of the :manpage:`gettimeofday(2)` reducing options
712 (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus
713 reduce precision of the timeout somewhat to really shrink the
714 :manpage:`gettimeofday(2)` call count. With this option enabled, we only do
715 about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all
716 time keeping was enabled.
718 .. option:: gtod_cpu=int
720 Sometimes it's cheaper to dedicate a single thread of execution to just
721 getting the current time. Fio (and databases, for instance) are very
722 intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set
723 one CPU aside for doing nothing but logging current time to a shared memory
724 location. Then the other threads/processes that run I/O workloads need only
725 copy that segment, instead of entering the kernel with a
726 :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time
727 calls will be excluded from other uses. Fio will manually clear it from the
728 CPU mask of other jobs.
734 .. option:: directory=str
736 Prefix filenames with this directory. Used to place files in a different
737 location than :file:`./`. You can specify a number of directories by
738 separating the names with a ':' character. These directories will be
739 assigned equally distributed to job clones created by :option:`numjobs` as
740 long as they are using generated filenames. If specific `filename(s)` are
741 set fio will use the first listed directory, and thereby matching the
742 `filename` semantic which generates a file each clone if not specified, but
743 let all clones use the same if set.
745 See the :option:`filename` option for information on how to escape "``:``" and
746 "``\``" characters within the directory path itself.
748 .. option:: filename=str
750 Fio normally makes up a `filename` based on the job name, thread number, and
751 file number (see :option:`filename_format`). If you want to share files
752 between threads in a job or several
753 jobs with fixed file paths, specify a `filename` for each of them to override
754 the default. If the ioengine is file based, you can specify a number of files
755 by separating the names with a ':' colon. So if you wanted a job to open
756 :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use
757 ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is
758 specified, :option:`nrfiles` is ignored. The size of regular files specified
759 by this option will be :option:`size` divided by number of files unless an
760 explicit size is specified by :option:`filesize`.
762 Each colon and backslash in the wanted path must be escaped with a ``\``
763 character. For instance, if the path is :file:`/dev/dsk/foo@3,0:c` then you
764 would use ``filename=/dev/dsk/foo@3,0\:c`` and if the path is
765 :file:`F:\\filename` then you would use ``filename=F\:\\filename``.
767 On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for
768 the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc.
769 Note: Windows and FreeBSD prevent write access to areas
770 of the disk containing in-use data (e.g. filesystems).
772 The filename "`-`" is a reserved name, meaning *stdin* or *stdout*. Which
773 of the two depends on the read/write direction set.
775 .. option:: filename_format=str
777 If sharing multiple files between jobs, it is usually necessary to have fio
778 generate the exact names that you want. By default, fio will name a file
779 based on the default file format specification of
780 :file:`jobname.jobnumber.filenumber`. With this option, that can be
781 customized. Fio will recognize and replace the following keywords in this
785 The name of the worker thread or process.
787 The incremental number of the worker thread or process.
789 The incremental number of the file for that worker thread or
792 To have dependent jobs share a set of files, this option can be set to have
793 fio generate filenames that are shared between the two. For instance, if
794 :file:`testfiles.$filenum` is specified, file number 4 for any job will be
795 named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum`
796 will be used if no other format specifier is given.
798 If you specify a path then the directories will be created up to the
799 main directory for the file. So for example if you specify
800 ``filename_format=a/b/c/$jobnum`` then the directories a/b/c will be
801 created before the file setup part of the job. If you specify
802 :option:`directory` then the path will be relative that directory,
803 otherwise it is treated as the absolute path.
805 .. option:: unique_filename=bool
807 To avoid collisions between networked clients, fio defaults to prefixing any
808 generated filenames (with a directory specified) with the source of the
809 client connecting. To disable this behavior, set this option to 0.
811 .. option:: opendir=str
813 Recursively open any files below directory `str`.
815 .. option:: lockfile=str
817 Fio defaults to not locking any files before it does I/O to them. If a file
818 or file descriptor is shared, fio can serialize I/O to that file to make the
819 end result consistent. This is usual for emulating real workloads that share
820 files. The lock modes are:
823 No locking. The default.
825 Only one thread or process may do I/O at a time, excluding all
828 Read-write locking on the file. Many readers may
829 access the file at the same time, but writes get exclusive access.
831 .. option:: nrfiles=int
833 Number of files to use for this job. Defaults to 1. The size of files
834 will be :option:`size` divided by this unless explicit size is specified by
835 :option:`filesize`. Files are created for each thread separately, and each
836 file will have a file number within its name by default, as explained in
837 :option:`filename` section.
840 .. option:: openfiles=int
842 Number of files to keep open at the same time. Defaults to the same as
843 :option:`nrfiles`, can be set smaller to limit the number simultaneous
846 .. option:: file_service_type=str
848 Defines how fio decides which file from a job to service next. The following
852 Choose a file at random.
855 Round robin over opened files. This is the default.
858 Finish one file before moving on to the next. Multiple files can
859 still be open depending on :option:`openfiles`.
862 Use a *Zipf* distribution to decide what file to access.
865 Use a *Pareto* distribution to decide what file to access.
868 Use a *Gaussian* (normal) distribution to decide what file to
874 For *random*, *roundrobin*, and *sequential*, a postfix can be appended to
875 tell fio how many I/Os to issue before switching to a new file. For example,
876 specifying ``file_service_type=random:8`` would cause fio to issue
877 8 I/Os before selecting a new file at random. For the non-uniform
878 distributions, a floating point postfix can be given to influence how the
879 distribution is skewed. See :option:`random_distribution` for a description
880 of how that would work.
882 .. option:: ioscheduler=str
884 Attempt to switch the device hosting the file to the specified I/O scheduler
887 .. option:: create_serialize=bool
889 If true, serialize the file creation for the jobs. This may be handy to
890 avoid interleaving of data files, which may greatly depend on the filesystem
891 used and even the number of processors in the system. Default: true.
893 .. option:: create_fsync=bool
895 :manpage:`fsync(2)` the data file after creation. This is the default.
897 .. option:: create_on_open=bool
899 If true, don't pre-create files but allow the job's open() to create a file
900 when it's time to do I/O. Default: false -- pre-create all necessary files
903 .. option:: create_only=bool
905 If true, fio will only run the setup phase of the job. If files need to be
906 laid out or updated on disk, only that will be done -- the actual job contents
907 are not executed. Default: false.
909 .. option:: allow_file_create=bool
911 If true, fio is permitted to create files as part of its workload. If this
912 option is false, then fio will error out if
913 the files it needs to use don't already exist. Default: true.
915 .. option:: allow_mounted_write=bool
917 If this isn't set, fio will abort jobs that are destructive (e.g. that write)
918 to what appears to be a mounted device or partition. This should help catch
919 creating inadvertently destructive tests, not realizing that the test will
920 destroy data on the mounted file system. Note that some platforms don't allow
921 writing against a mounted device regardless of this option. Default: false.
923 .. option:: pre_read=bool
925 If this is given, files will be pre-read into memory before starting the
926 given I/O operation. This will also clear the :option:`invalidate` flag,
927 since it is pointless to pre-read and then drop the cache. This will only
928 work for I/O engines that are seek-able, since they allow you to read the
929 same data multiple times. Thus it will not work on non-seekable I/O engines
930 (e.g. network, splice). Default: false.
932 .. option:: unlink=bool
934 Unlink the job files when done. Not the default, as repeated runs of that
935 job would then waste time recreating the file set again and again. Default:
938 .. option:: unlink_each_loop=bool
940 Unlink job files after each iteration or loop. Default: false.
942 .. option:: zonesize=int
944 Divide a file into zones of the specified size. See :option:`zoneskip`.
946 .. option:: zonerange=int
948 Give size of an I/O zone. See :option:`zoneskip`.
950 .. option:: zoneskip=int
952 Skip the specified number of bytes when :option:`zonesize` data has been
953 read. The two zone options can be used to only do I/O on zones of a file.
959 .. option:: direct=bool
961 If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that
962 OpenBSD and ZFS on Solaris don't support direct I/O. On Windows the synchronous
963 ioengines don't support direct I/O. Default: false.
965 .. option:: atomic=bool
967 If value is true, attempt to use atomic direct I/O. Atomic writes are
968 guaranteed to be stable once acknowledged by the operating system. Only
969 Linux supports O_ATOMIC right now.
971 .. option:: buffered=bool
973 If value is true, use buffered I/O. This is the opposite of the
974 :option:`direct` option. Defaults to true.
976 .. option:: readwrite=str, rw=str
978 Type of I/O pattern. Accepted values are:
985 Sequential trims (Linux block devices only).
991 Random trims (Linux block devices only).
993 Sequential mixed reads and writes.
995 Random mixed reads and writes.
997 Sequential trim+write sequences. Blocks will be trimmed first,
998 then the same blocks will be written to.
1000 Fio defaults to read if the option is not specified. For the mixed I/O
1001 types, the default is to split them 50/50. For certain types of I/O the
1002 result may still be skewed a bit, since the speed may be different.
1004 It is possible to specify the number of I/Os to do before getting a new
1005 offset by appending ``:<nr>`` to the end of the string given. For a
1006 random read, it would look like ``rw=randread:8`` for passing in an offset
1007 modifier with a value of 8. If the suffix is used with a sequential I/O
1008 pattern, then the *<nr>* value specified will be **added** to the generated
1009 offset for each I/O turning sequential I/O into sequential I/O with holes.
1010 For instance, using ``rw=write:4k`` will skip 4k for every write. Also see
1011 the :option:`rw_sequencer` option.
1013 .. option:: rw_sequencer=str
1015 If an offset modifier is given by appending a number to the ``rw=<str>``
1016 line, then this option controls how that number modifies the I/O offset
1017 being generated. Accepted values are:
1020 Generate sequential offset.
1022 Generate the same offset.
1024 ``sequential`` is only useful for random I/O, where fio would normally
1025 generate a new random offset for every I/O. If you append e.g. 8 to randread,
1026 you would get a new random offset for every 8 I/Os. The result would be a
1027 seek for only every 8 I/Os, instead of for every I/O. Use ``rw=randread:8``
1028 to specify that. As sequential I/O is already sequential, setting
1029 ``sequential`` for that would not result in any differences. ``identical``
1030 behaves in a similar fashion, except it sends the same offset 8 number of
1031 times before generating a new offset.
1033 .. option:: unified_rw_reporting=bool
1035 Fio normally reports statistics on a per data direction basis, meaning that
1036 reads, writes, and trims are accounted and reported separately. If this
1037 option is set fio sums the results and report them as "mixed" instead.
1039 .. option:: randrepeat=bool
1041 Seed the random number generator used for random I/O patterns in a
1042 predictable way so the pattern is repeatable across runs. Default: true.
1044 .. option:: allrandrepeat=bool
1046 Seed all random number generators in a predictable way so results are
1047 repeatable across runs. Default: false.
1049 .. option:: randseed=int
1051 Seed the random number generators based on this seed value, to be able to
1052 control what sequence of output is being generated. If not set, the random
1053 sequence depends on the :option:`randrepeat` setting.
1055 .. option:: fallocate=str
1057 Whether pre-allocation is performed when laying down files.
1058 Accepted values are:
1061 Do not pre-allocate space.
1064 Use a platform's native pre-allocation call but fall back to
1065 **none** behavior if it fails/is not implemented.
1068 Pre-allocate via :manpage:`posix_fallocate(3)`.
1071 Pre-allocate via :manpage:`fallocate(2)` with
1072 FALLOC_FL_KEEP_SIZE set.
1075 Backward-compatible alias for **none**.
1078 Backward-compatible alias for **posix**.
1080 May not be available on all supported platforms. **keep** is only available
1081 on Linux. If using ZFS on Solaris this cannot be set to **posix**
1082 because ZFS doesn't support pre-allocation. Default: **native** if any
1083 pre-allocation methods are available, **none** if not.
1085 .. option:: fadvise_hint=str
1087 Use :manpage:`posix_fadvise(2)` to advise the kernel on what I/O patterns
1088 are likely to be issued. Accepted values are:
1091 Backwards-compatible hint for "no hint".
1094 Backwards compatible hint for "advise with fio workload type". This
1095 uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL**
1096 for a sequential workload.
1099 Advise using **FADV_SEQUENTIAL**.
1102 Advise using **FADV_RANDOM**.
1104 .. option:: write_hint=str
1106 Use :manpage:`fcntl(2)` to advise the kernel what life time to expect
1107 from a write. Only supported on Linux, as of version 4.13. Accepted
1111 No particular life time associated with this file.
1114 Data written to this file has a short life time.
1117 Data written to this file has a medium life time.
1120 Data written to this file has a long life time.
1123 Data written to this file has a very long life time.
1125 The values are all relative to each other, and no absolute meaning
1126 should be associated with them.
1128 .. option:: offset=int
1130 Start I/O at the provided offset in the file, given as either a fixed size in
1131 bytes or a percentage. If a percentage is given, the generated offset will be
1132 aligned to the minimum ``blocksize`` or to the value of ``offset_align`` if
1133 provided. Data before the given offset will not be touched. This
1134 effectively caps the file size at `real_size - offset`. Can be combined with
1135 :option:`size` to constrain the start and end range of the I/O workload.
1136 A percentage can be specified by a number between 1 and 100 followed by '%',
1137 for example, ``offset=20%`` to specify 20%.
1139 .. option:: offset_align=int
1141 If set to non-zero value, the byte offset generated by a percentage ``offset``
1142 is aligned upwards to this value. Defaults to 0 meaning that a percentage
1143 offset is aligned to the minimum block size.
1145 .. option:: offset_increment=int
1147 If this is provided, then the real offset becomes `offset + offset_increment
1148 * thread_number`, where the thread number is a counter that starts at 0 and
1149 is incremented for each sub-job (i.e. when :option:`numjobs` option is
1150 specified). This option is useful if there are several jobs which are
1151 intended to operate on a file in parallel disjoint segments, with even
1152 spacing between the starting points.
1154 .. option:: number_ios=int
1156 Fio will normally perform I/Os until it has exhausted the size of the region
1157 set by :option:`size`, or if it exhaust the allocated time (or hits an error
1158 condition). With this setting, the range/size can be set independently of
1159 the number of I/Os to perform. When fio reaches this number, it will exit
1160 normally and report status. Note that this does not extend the amount of I/O
1161 that will be done, it will only stop fio if this condition is met before
1162 other end-of-job criteria.
1164 .. option:: fsync=int
1166 If writing to a file, issue an :manpage:`fsync(2)` (or its equivalent) of
1167 the dirty data for every number of blocks given. For example, if you give 32
1168 as a parameter, fio will sync the file after every 32 writes issued. If fio is
1169 using non-buffered I/O, we may not sync the file. The exception is the sg
1170 I/O engine, which synchronizes the disk cache anyway. Defaults to 0, which
1171 means fio does not periodically issue and wait for a sync to complete. Also
1172 see :option:`end_fsync` and :option:`fsync_on_close`.
1174 .. option:: fdatasync=int
1176 Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and
1177 not metadata blocks. In Windows, FreeBSD, and DragonFlyBSD there is no
1178 :manpage:`fdatasync(2)` so this falls back to using :manpage:`fsync(2)`.
1179 Defaults to 0, which means fio does not periodically issue and wait for a
1180 data-only sync to complete.
1182 .. option:: write_barrier=int
1184 Make every `N-th` write a barrier write.
1186 .. option:: sync_file_range=str:int
1188 Use :manpage:`sync_file_range(2)` for every `int` number of write
1189 operations. Fio will track range of writes that have happened since the last
1190 :manpage:`sync_file_range(2)` call. `str` can currently be one or more of:
1193 SYNC_FILE_RANGE_WAIT_BEFORE
1195 SYNC_FILE_RANGE_WRITE
1197 SYNC_FILE_RANGE_WAIT_AFTER
1199 So if you do ``sync_file_range=wait_before,write:8``, fio would use
1200 ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8
1201 writes. Also see the :manpage:`sync_file_range(2)` man page. This option is
1204 .. option:: overwrite=bool
1206 If true, writes to a file will always overwrite existing data. If the file
1207 doesn't already exist, it will be created before the write phase begins. If
1208 the file exists and is large enough for the specified write phase, nothing
1209 will be done. Default: false.
1211 .. option:: end_fsync=bool
1213 If true, :manpage:`fsync(2)` file contents when a write stage has completed.
1216 .. option:: fsync_on_close=bool
1218 If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs
1219 from :option:`end_fsync` in that it will happen on every file close, not
1220 just at the end of the job. Default: false.
1222 .. option:: rwmixread=int
1224 Percentage of a mixed workload that should be reads. Default: 50.
1226 .. option:: rwmixwrite=int
1228 Percentage of a mixed workload that should be writes. If both
1229 :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not
1230 add up to 100%, the latter of the two will be used to override the
1231 first. This may interfere with a given rate setting, if fio is asked to
1232 limit reads or writes to a certain rate. If that is the case, then the
1233 distribution may be skewed. Default: 50.
1235 .. option:: random_distribution=str:float[,str:float][,str:float]
1237 By default, fio will use a completely uniform random distribution when asked
1238 to perform random I/O. Sometimes it is useful to skew the distribution in
1239 specific ways, ensuring that some parts of the data is more hot than others.
1240 fio includes the following distribution models:
1243 Uniform random distribution
1252 Normal (Gaussian) distribution
1255 Zoned random distribution
1257 When using a **zipf** or **pareto** distribution, an input value is also
1258 needed to define the access pattern. For **zipf**, this is the `Zipf
1259 theta`. For **pareto**, it's the `Pareto power`. Fio includes a test
1260 program, :command:`fio-genzipf`, that can be used visualize what the given input
1261 values will yield in terms of hit rates. If you wanted to use **zipf** with
1262 a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the
1263 option. If a non-uniform model is used, fio will disable use of the random
1264 map. For the **normal** distribution, a normal (Gaussian) deviation is
1265 supplied as a value between 0 and 100.
1267 For a **zoned** distribution, fio supports specifying percentages of I/O
1268 access that should fall within what range of the file or device. For
1269 example, given a criteria of:
1271 * 60% of accesses should be to the first 10%
1272 * 30% of accesses should be to the next 20%
1273 * 8% of accesses should be to the next 30%
1274 * 2% of accesses should be to the next 40%
1276 we can define that through zoning of the random accesses. For the above
1277 example, the user would do::
1279 random_distribution=zoned:60/10:30/20:8/30:2/40
1281 similarly to how :option:`bssplit` works for setting ranges and percentages
1282 of block sizes. Like :option:`bssplit`, it's possible to specify separate
1283 zones for reads, writes, and trims. If just one set is given, it'll apply to
1286 .. option:: percentage_random=int[,int][,int]
1288 For a random workload, set how big a percentage should be random. This
1289 defaults to 100%, in which case the workload is fully random. It can be set
1290 from anywhere from 0 to 100. Setting it to 0 would make the workload fully
1291 sequential. Any setting in between will result in a random mix of sequential
1292 and random I/O, at the given percentages. Comma-separated values may be
1293 specified for reads, writes, and trims as described in :option:`blocksize`.
1295 .. option:: norandommap
1297 Normally fio will cover every block of the file when doing random I/O. If
1298 this option is given, fio will just get a new random offset without looking
1299 at past I/O history. This means that some blocks may not be read or written,
1300 and that some blocks may be read/written more than once. If this option is
1301 used with :option:`verify` and multiple blocksizes (via :option:`bsrange`),
1302 only intact blocks are verified, i.e., partially-overwritten blocks are
1305 .. option:: softrandommap=bool
1307 See :option:`norandommap`. If fio runs with the random block map enabled and
1308 it fails to allocate the map, if this option is set it will continue without
1309 a random block map. As coverage will not be as complete as with random maps,
1310 this option is disabled by default.
1312 .. option:: random_generator=str
1314 Fio supports the following engines for generating I/O offsets for random I/O:
1317 Strong 2^88 cycle random number generator.
1319 Linear feedback shift register generator.
1321 Strong 64-bit 2^258 cycle random number generator.
1323 **tausworthe** is a strong random number generator, but it requires tracking
1324 on the side if we want to ensure that blocks are only read or written
1325 once. **lfsr** guarantees that we never generate the same offset twice, and
1326 it's also less computationally expensive. It's not a true random generator,
1327 however, though for I/O purposes it's typically good enough. **lfsr** only
1328 works with single block sizes, not with workloads that use multiple block
1329 sizes. If used with such a workload, fio may read or write some blocks
1330 multiple times. The default value is **tausworthe**, unless the required
1331 space exceeds 2^32 blocks. If it does, then **tausworthe64** is
1332 selected automatically.
1338 .. option:: blocksize=int[,int][,int], bs=int[,int][,int]
1340 The block size in bytes used for I/O units. Default: 4096. A single value
1341 applies to reads, writes, and trims. Comma-separated values may be
1342 specified for reads, writes, and trims. A value not terminated in a comma
1343 applies to subsequent types.
1348 means 256k for reads, writes and trims.
1351 means 8k for reads, 32k for writes and trims.
1354 means 8k for reads, 32k for writes, and default for trims.
1357 means default for reads, 8k for writes and trims.
1360 means default for reads, 8k for writes, and default for trims.
1362 .. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange]
1364 A range of block sizes in bytes for I/O units. The issued I/O unit will
1365 always be a multiple of the minimum size, unless
1366 :option:`blocksize_unaligned` is set.
1368 Comma-separated ranges may be specified for reads, writes, and trims as
1369 described in :option:`blocksize`.
1371 Example: ``bsrange=1k-4k,2k-8k``.
1373 .. option:: bssplit=str[,str][,str]
1375 Sometimes you want even finer grained control of the block sizes issued, not
1376 just an even split between them. This option allows you to weight various
1377 block sizes, so that you are able to define a specific amount of block sizes
1378 issued. The format for this option is::
1380 bssplit=blocksize/percentage:blocksize/percentage
1382 for as many block sizes as needed. So if you want to define a workload that
1383 has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would write::
1385 bssplit=4k/10:64k/50:32k/40
1387 Ordering does not matter. If the percentage is left blank, fio will fill in
1388 the remaining values evenly. So a bssplit option like this one::
1390 bssplit=4k/50:1k/:32k/
1392 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always add up
1393 to 100, if bssplit is given a range that adds up to more, it will error out.
1395 Comma-separated values may be specified for reads, writes, and trims as
1396 described in :option:`blocksize`.
1398 If you want a workload that has 50% 2k reads and 50% 4k reads, while having
1399 90% 4k writes and 10% 8k writes, you would specify::
1401 bssplit=2k/50:4k/50,4k/90,8k/10
1403 .. option:: blocksize_unaligned, bs_unaligned
1405 If set, fio will issue I/O units with any size within
1406 :option:`blocksize_range`, not just multiples of the minimum size. This
1407 typically won't work with direct I/O, as that normally requires sector
1410 .. option:: bs_is_seq_rand=bool
1412 If this option is set, fio will use the normal read,write blocksize settings
1413 as sequential,random blocksize settings instead. Any random read or write
1414 will use the WRITE blocksize settings, and any sequential read or write will
1415 use the READ blocksize settings.
1417 .. option:: blockalign=int[,int][,int], ba=int[,int][,int]
1419 Boundary to which fio will align random I/O units. Default:
1420 :option:`blocksize`. Minimum alignment is typically 512b for using direct
1421 I/O, though it usually depends on the hardware block size. This option is
1422 mutually exclusive with using a random map for files, so it will turn off
1423 that option. Comma-separated values may be specified for reads, writes, and
1424 trims as described in :option:`blocksize`.
1430 .. option:: zero_buffers
1432 Initialize buffers with all zeros. Default: fill buffers with random data.
1434 .. option:: refill_buffers
1436 If this option is given, fio will refill the I/O buffers on every
1437 submit. Only makes sense if :option:`zero_buffers` isn't specified,
1438 naturally. Defaults to being unset i.e., the buffer is only filled at
1439 init time and the data in it is reused when possible but if any of
1440 :option:`verify`, :option:`buffer_compress_percentage` or
1441 :option:`dedupe_percentage` are enabled then `refill_buffers` is also
1442 automatically enabled.
1444 .. option:: scramble_buffers=bool
1446 If :option:`refill_buffers` is too costly and the target is using data
1447 deduplication, then setting this option will slightly modify the I/O buffer
1448 contents to defeat normal de-dupe attempts. This is not enough to defeat
1449 more clever block compression attempts, but it will stop naive dedupe of
1450 blocks. Default: true.
1452 .. option:: buffer_compress_percentage=int
1454 If this is set, then fio will attempt to provide I/O buffer content
1455 (on WRITEs) that compresses to the specified level. Fio does this by
1456 providing a mix of random data followed by fixed pattern data. The
1457 fixed pattern is either zeros, or the pattern specified by
1458 :option:`buffer_pattern`. If the `buffer_pattern` option is used, it
1459 might skew the compression ratio slightly. Setting
1460 `buffer_compress_percentage` to a value other than 100 will also
1461 enable :option:`refill_buffers` in order to reduce the likelihood that
1462 adjacent blocks are so similar that they over compress when seen
1463 together. See :option:`buffer_compress_chunk` for how to set a finer or
1464 coarser granularity for the random/fixed data region. Defaults to unset
1465 i.e., buffer data will not adhere to any compression level.
1467 .. option:: buffer_compress_chunk=int
1469 This setting allows fio to manage how big the random/fixed data region
1470 is when using :option:`buffer_compress_percentage`. When
1471 `buffer_compress_chunk` is set to some non-zero value smaller than the
1472 block size, fio can repeat the random/fixed region throughout the I/O
1473 buffer at the specified interval (which particularly useful when
1474 bigger block sizes are used for a job). When set to 0, fio will use a
1475 chunk size that matches the block size resulting in a single
1476 random/fixed region within the I/O buffer. Defaults to 512. When the
1477 unit is omitted, the value is interpreted in bytes.
1479 .. option:: buffer_pattern=str
1481 If set, fio will fill the I/O buffers with this pattern or with the contents
1482 of a file. If not set, the contents of I/O buffers are defined by the other
1483 options related to buffer contents. The setting can be any pattern of bytes,
1484 and can be prefixed with 0x for hex values. It may also be a string, where
1485 the string must then be wrapped with ``""``. Or it may also be a filename,
1486 where the filename must be wrapped with ``''`` in which case the file is
1487 opened and read. Note that not all the file contents will be read if that
1488 would cause the buffers to overflow. So, for example::
1490 buffer_pattern='filename'
1494 buffer_pattern="abcd"
1502 buffer_pattern=0xdeadface
1504 Also you can combine everything together in any order::
1506 buffer_pattern=0xdeadface"abcd"-12'filename'
1508 .. option:: dedupe_percentage=int
1510 If set, fio will generate this percentage of identical buffers when
1511 writing. These buffers will be naturally dedupable. The contents of the
1512 buffers depend on what other buffer compression settings have been set. It's
1513 possible to have the individual buffers either fully compressible, or not at
1514 all -- this option only controls the distribution of unique buffers. Setting
1515 this option will also enable :option:`refill_buffers` to prevent every buffer
1518 .. option:: invalidate=bool
1520 Invalidate the buffer/page cache parts of the files to be used prior to
1521 starting I/O if the platform and file type support it. Defaults to true.
1522 This will be ignored if :option:`pre_read` is also specified for the
1525 .. option:: sync=bool
1527 Use synchronous I/O for buffered writes. For the majority of I/O engines,
1528 this means using O_SYNC. Default: false.
1530 .. option:: iomem=str, mem=str
1532 Fio can use various types of memory as the I/O unit buffer. The allowed
1536 Use memory from :manpage:`malloc(3)` as the buffers. Default memory
1540 Use shared memory as the buffers. Allocated through
1541 :manpage:`shmget(2)`.
1544 Same as shm, but use huge pages as backing.
1547 Use :manpage:`mmap(2)` to allocate buffers. May either be anonymous memory, or can
1548 be file backed if a filename is given after the option. The format
1549 is `mem=mmap:/path/to/file`.
1552 Use a memory mapped huge file as the buffer backing. Append filename
1553 after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`.
1556 Same as mmap, but use a MMAP_SHARED mapping.
1559 Use GPU memory as the buffers for GPUDirect RDMA benchmark.
1560 The :option:`ioengine` must be `rdma`.
1562 The area allocated is a function of the maximum allowed bs size for the job,
1563 multiplied by the I/O depth given. Note that for **shmhuge** and
1564 **mmaphuge** to work, the system must have free huge pages allocated. This
1565 can normally be checked and set by reading/writing
1566 :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page
1567 is 4MiB in size. So to calculate the number of huge pages you need for a
1568 given job file, add up the I/O depth of all jobs (normally one unless
1569 :option:`iodepth` is used) and multiply by the maximum bs set. Then divide
1570 that number by the huge page size. You can see the size of the huge pages in
1571 :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero
1572 number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also
1573 see :option:`hugepage-size`.
1575 **mmaphuge** also needs to have hugetlbfs mounted and the file location
1576 should point there. So if it's mounted in :file:`/huge`, you would use
1577 `mem=mmaphuge:/huge/somefile`.
1579 .. option:: iomem_align=int, mem_align=int
1581 This indicates the memory alignment of the I/O memory buffers. Note that
1582 the given alignment is applied to the first I/O unit buffer, if using
1583 :option:`iodepth` the alignment of the following buffers are given by the
1584 :option:`bs` used. In other words, if using a :option:`bs` that is a
1585 multiple of the page sized in the system, all buffers will be aligned to
1586 this value. If using a :option:`bs` that is not page aligned, the alignment
1587 of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and
1590 .. option:: hugepage-size=int
1592 Defines the size of a huge page. Must at least be equal to the system
1593 setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably
1594 always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the
1595 preferred way to set this to avoid setting a non-pow-2 bad value.
1597 .. option:: lockmem=int
1599 Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to
1600 simulate a smaller amount of memory. The amount specified is per worker.
1606 .. option:: size=int
1608 The total size of file I/O for each thread of this job. Fio will run until
1609 this many bytes has been transferred, unless runtime is limited by other options
1610 (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`).
1611 Fio will divide this size between the available files determined by options
1612 such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is
1613 specified by the job. If the result of division happens to be 0, the size is
1614 set to the physical size of the given files or devices if they exist.
1615 If this option is not specified, fio will use the full size of the given
1616 files or devices. If the files do not exist, size must be given. It is also
1617 possible to give size as a percentage between 1 and 100. If ``size=20%`` is
1618 given, fio will use 20% of the full size of the given files or devices.
1619 Can be combined with :option:`offset` to constrain the start and end range
1620 that I/O will be done within.
1622 .. option:: io_size=int, io_limit=int
1624 Normally fio operates within the region set by :option:`size`, which means
1625 that the :option:`size` option sets both the region and size of I/O to be
1626 performed. Sometimes that is not what you want. With this option, it is
1627 possible to define just the amount of I/O that fio should do. For instance,
1628 if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio
1629 will perform I/O within the first 20GiB but exit when 5GiB have been
1630 done. The opposite is also possible -- if :option:`size` is set to 20GiB,
1631 and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within
1632 the 0..20GiB region.
1634 .. option:: filesize=irange(int)
1636 Individual file sizes. May be a range, in which case fio will select sizes
1637 for files at random within the given range and limited to :option:`size` in
1638 total (if that is given). If not given, each created file is the same size.
1639 This option overrides :option:`size` in terms of file size, which means
1640 this value is used as a fixed size or possible range of each file.
1642 .. option:: file_append=bool
1644 Perform I/O after the end of the file. Normally fio will operate within the
1645 size of a file. If this option is set, then fio will append to the file
1646 instead. This has identical behavior to setting :option:`offset` to the size
1647 of a file. This option is ignored on non-regular files.
1649 .. option:: fill_device=bool, fill_fs=bool
1651 Sets size to something really large and waits for ENOSPC (no space left on
1652 device) as the terminating condition. Only makes sense with sequential
1653 write. For a read workload, the mount point will be filled first then I/O
1654 started on the result. This option doesn't make sense if operating on a raw
1655 device node, since the size of that is already known by the file system.
1656 Additionally, writing beyond end-of-device will not return ENOSPC there.
1662 .. option:: ioengine=str
1664 Defines how the job issues I/O to the file. The following types are defined:
1667 Basic :manpage:`read(2)` or :manpage:`write(2)`
1668 I/O. :manpage:`lseek(2)` is used to position the I/O location.
1669 See :option:`fsync` and :option:`fdatasync` for syncing write I/Os.
1672 Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on
1673 all supported operating systems except for Windows.
1676 Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate
1677 queuing by coalescing adjacent I/Os into a single submission.
1680 Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O.
1683 Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O.
1686 Linux native asynchronous I/O. Note that Linux may only support
1687 queued behavior with non-buffered I/O (set ``direct=1`` or
1689 This engine defines engine specific options.
1692 POSIX asynchronous I/O using :manpage:`aio_read(3)` and
1693 :manpage:`aio_write(3)`.
1696 Solaris native asynchronous I/O.
1699 Windows native asynchronous I/O. Default on Windows.
1702 File is memory mapped with :manpage:`mmap(2)` and data copied
1703 to/from using :manpage:`memcpy(3)`.
1706 :manpage:`splice(2)` is used to transfer the data and
1707 :manpage:`vmsplice(2)` to transfer data from user space to the
1711 SCSI generic sg v3 I/O. May either be synchronous using the SG_IO
1712 ioctl, or if the target is an sg character device we use
1713 :manpage:`read(2)` and :manpage:`write(2)` for asynchronous
1714 I/O. Requires :option:`filename` option to specify either block or
1718 Doesn't transfer any data, just pretends to. This is mainly used to
1719 exercise fio itself and for debugging/testing purposes.
1722 Transfer over the network to given ``host:port``. Depending on the
1723 :option:`protocol` used, the :option:`hostname`, :option:`port`,
1724 :option:`listen` and :option:`filename` options are used to specify
1725 what sort of connection to make, while the :option:`protocol` option
1726 determines which protocol will be used. This engine defines engine
1730 Like **net**, but uses :manpage:`splice(2)` and
1731 :manpage:`vmsplice(2)` to map data and send/receive.
1732 This engine defines engine specific options.
1735 Doesn't transfer any data, but burns CPU cycles according to the
1736 :option:`cpuload` and :option:`cpuchunks` options. Setting
1737 :option:`cpuload`\=85 will cause that job to do nothing but burn 85%
1738 of the CPU. In case of SMP machines, use :option:`numjobs`\=<nr_of_cpu>
1739 to get desired CPU usage, as the cpuload only loads a
1740 single CPU at the desired rate. A job never finishes unless there is
1741 at least one non-cpuio job.
1744 The GUASI I/O engine is the Generic Userspace Asyncronous Syscall
1745 Interface approach to async I/O. See
1747 http://www.xmailserver.org/guasi-lib.html
1749 for more info on GUASI.
1752 The RDMA I/O engine supports both RDMA memory semantics
1753 (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the
1754 InfiniBand, RoCE and iWARP protocols. This engine defines engine
1758 I/O engine that does regular fallocate to simulate data transfer as
1762 does fallocate(,mode = FALLOC_FL_KEEP_SIZE,).
1765 does fallocate(,mode = 0).
1768 does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE).
1771 I/O engine that sends :manpage:`ftruncate(2)` operations in response
1772 to write (DDIR_WRITE) events. Each ftruncate issued sets the file's
1773 size to the current block offset. :option:`blocksize` is ignored.
1776 I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate
1777 defragment activity in request to DDIR_WRITE event.
1780 I/O engine supporting direct access to Ceph Rados Block Devices
1781 (RBD) via librbd without the need to use the kernel rbd driver. This
1782 ioengine defines engine specific options.
1785 Using GlusterFS libgfapi sync interface to direct access to
1786 GlusterFS volumes without having to go through FUSE. This ioengine
1787 defines engine specific options.
1790 Using GlusterFS libgfapi async interface to direct access to
1791 GlusterFS volumes without having to go through FUSE. This ioengine
1792 defines engine specific options.
1795 Read and write through Hadoop (HDFS). The :option:`filename` option
1796 is used to specify host,port of the hdfs name-node to connect. This
1797 engine interprets offsets a little differently. In HDFS, files once
1798 created cannot be modified so random writes are not possible. To
1799 imitate this the libhdfs engine expects a bunch of small files to be
1800 created over HDFS and will randomly pick a file from them
1801 based on the offset generated by fio backend (see the example
1802 job file to create such files, use ``rw=write`` option). Please
1803 note, it may be necessary to set environment variables to work
1804 with HDFS/libhdfs properly. Each job uses its own connection to
1808 Read, write and erase an MTD character device (e.g.,
1809 :file:`/dev/mtd0`). Discards are treated as erases. Depending on the
1810 underlying device type, the I/O may have to go in a certain pattern,
1811 e.g., on NAND, writing sequentially to erase blocks and discarding
1812 before overwriting. The `trimwrite` mode works well for this
1816 Read and write using filesystem DAX to a file on a filesystem
1817 mounted with DAX on a persistent memory device through the NVML
1821 Read and write using device DAX to a persistent memory device (e.g.,
1822 /dev/dax0.0) through the NVML libpmem library.
1825 Prefix to specify loading an external I/O engine object file. Append
1826 the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load
1827 ioengine :file:`foo.o` in :file:`/tmp`. The path can be either
1828 absolute or relative. See :file:`engines/skeleton_external.c` for
1829 details of writing an external I/O engine.
1832 Simply create the files and do no I/O to them. You still need to
1833 set `filesize` so that all the accounting still occurs, but no
1834 actual I/O will be done other than creating the file.
1837 Read and write using mmap I/O to a file on a filesystem
1838 mounted with DAX on a persistent memory device through the NVML
1841 I/O engine specific parameters
1842 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1844 In addition, there are some parameters which are only valid when a specific
1845 :option:`ioengine` is in use. These are used identically to normal parameters,
1846 with the caveat that when used on the command line, they must come after the
1847 :option:`ioengine` that defines them is selected.
1849 .. option:: userspace_reap : [libaio]
1851 Normally, with the libaio engine in use, fio will use the
1852 :manpage:`io_getevents(2)` system call to reap newly returned events. With
1853 this flag turned on, the AIO ring will be read directly from user-space to
1854 reap events. The reaping mode is only enabled when polling for a minimum of
1855 0 events (e.g. when :option:`iodepth_batch_complete` `=0`).
1857 .. option:: hipri : [pvsync2]
1859 Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority
1862 .. option:: hipri_percentage : [pvsync2]
1864 When hipri is set this determines the probability of a pvsync2 I/O being high
1865 priority. The default is 100%.
1867 .. option:: cpuload=int : [cpuio]
1869 Attempt to use the specified percentage of CPU cycles. This is a mandatory
1870 option when using cpuio I/O engine.
1872 .. option:: cpuchunks=int : [cpuio]
1874 Split the load into cycles of the given time. In microseconds.
1876 .. option:: exit_on_io_done=bool : [cpuio]
1878 Detect when I/O threads are done, then exit.
1880 .. option:: namenode=str : [libhdfs]
1882 The hostname or IP address of a HDFS cluster namenode to contact.
1884 .. option:: port=int
1888 The listening port of the HFDS cluster namenode.
1892 The TCP or UDP port to bind to or connect to. If this is used with
1893 :option:`numjobs` to spawn multiple instances of the same job type, then
1894 this will be the starting port number since fio will use a range of
1899 The port to use for RDMA-CM communication. This should be the same value
1900 on the client and the server side.
1902 .. option:: hostname=str : [netsplice] [net] [rdma]
1904 The hostname or IP address to use for TCP, UDP or RDMA-CM based I/O. If the job
1905 is a TCP listener or UDP reader, the hostname is not used and must be omitted
1906 unless it is a valid UDP multicast address.
1908 .. option:: interface=str : [netsplice] [net]
1910 The IP address of the network interface used to send or receive UDP
1913 .. option:: ttl=int : [netsplice] [net]
1915 Time-to-live value for outgoing UDP multicast packets. Default: 1.
1917 .. option:: nodelay=bool : [netsplice] [net]
1919 Set TCP_NODELAY on TCP connections.
1921 .. option:: protocol=str, proto=str : [netsplice] [net]
1923 The network protocol to use. Accepted values are:
1926 Transmission control protocol.
1928 Transmission control protocol V6.
1930 User datagram protocol.
1932 User datagram protocol V6.
1936 When the protocol is TCP or UDP, the port must also be given, as well as the
1937 hostname if the job is a TCP listener or UDP reader. For unix sockets, the
1938 normal :option:`filename` option should be used and the port is invalid.
1940 .. option:: listen : [netsplice] [net]
1942 For TCP network connections, tell fio to listen for incoming connections
1943 rather than initiating an outgoing connection. The :option:`hostname` must
1944 be omitted if this option is used.
1946 .. option:: pingpong : [netsplice] [net]
1948 Normally a network writer will just continue writing data, and a network
1949 reader will just consume packages. If ``pingpong=1`` is set, a writer will
1950 send its normal payload to the reader, then wait for the reader to send the
1951 same payload back. This allows fio to measure network latencies. The
1952 submission and completion latencies then measure local time spent sending or
1953 receiving, and the completion latency measures how long it took for the
1954 other end to receive and send back. For UDP multicast traffic
1955 ``pingpong=1`` should only be set for a single reader when multiple readers
1956 are listening to the same address.
1958 .. option:: window_size : [netsplice] [net]
1960 Set the desired socket buffer size for the connection.
1962 .. option:: mss : [netsplice] [net]
1964 Set the TCP maximum segment size (TCP_MAXSEG).
1966 .. option:: donorname=str : [e4defrag]
1968 File will be used as a block donor (swap extents between files).
1970 .. option:: inplace=int : [e4defrag]
1972 Configure donor file blocks allocation strategy:
1975 Default. Preallocate donor's file on init.
1977 Allocate space immediately inside defragment event, and free right
1980 .. option:: clustername=str : [rbd]
1982 Specifies the name of the Ceph cluster.
1984 .. option:: rbdname=str : [rbd]
1986 Specifies the name of the RBD.
1988 .. option:: pool=str : [rbd]
1990 Specifies the name of the Ceph pool containing RBD.
1992 .. option:: clientname=str : [rbd]
1994 Specifies the username (without the 'client.' prefix) used to access the
1995 Ceph cluster. If the *clustername* is specified, the *clientname* shall be
1996 the full *type.id* string. If no type. prefix is given, fio will add
1997 'client.' by default.
1999 .. option:: skip_bad=bool : [mtd]
2001 Skip operations against known bad blocks.
2003 .. option:: hdfsdirectory : [libhdfs]
2005 libhdfs will create chunk in this HDFS directory.
2007 .. option:: chunk_size : [libhdfs]
2009 The size of the chunk to use for each file.
2011 .. option:: verb=str : [rdma]
2013 The RDMA verb to use on this side of the RDMA ioengine connection. Valid
2014 values are write, read, send and recv. These correspond to the equivalent
2015 RDMA verbs (e.g. write = rdma_write etc.). Note that this only needs to be
2016 specified on the client side of the connection. See the examples folder.
2018 .. option:: bindname=str : [rdma]
2020 The name to use to bind the local RDMA-CM connection to a local RDMA device.
2021 This could be a hostname or an IPv4 or IPv6 address. On the server side this
2022 will be passed into the rdma_bind_addr() function and on the client site it
2023 will be used in the rdma_resolve_add() function. This can be useful when
2024 multiple paths exist between the client and the server or in certain loopback
2030 .. option:: iodepth=int
2032 Number of I/O units to keep in flight against the file. Note that
2033 increasing *iodepth* beyond 1 will not affect synchronous ioengines (except
2034 for small degrees when :option:`verify_async` is in use). Even async
2035 engines may impose OS restrictions causing the desired depth not to be
2036 achieved. This may happen on Linux when using libaio and not setting
2037 :option:`direct`\=1, since buffered I/O is not async on that OS. Keep an
2038 eye on the I/O depth distribution in the fio output to verify that the
2039 achieved depth is as expected. Default: 1.
2041 .. option:: iodepth_batch_submit=int, iodepth_batch=int
2043 This defines how many pieces of I/O to submit at once. It defaults to 1
2044 which means that we submit each I/O as soon as it is available, but can be
2045 raised to submit bigger batches of I/O at the time. If it is set to 0 the
2046 :option:`iodepth` value will be used.
2048 .. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int
2050 This defines how many pieces of I/O to retrieve at once. It defaults to 1
2051 which means that we'll ask for a minimum of 1 I/O in the retrieval process
2052 from the kernel. The I/O retrieval will go on until we hit the limit set by
2053 :option:`iodepth_low`. If this variable is set to 0, then fio will always
2054 check for completed events before queuing more I/O. This helps reduce I/O
2055 latency, at the cost of more retrieval system calls.
2057 .. option:: iodepth_batch_complete_max=int
2059 This defines maximum pieces of I/O to retrieve at once. This variable should
2060 be used along with :option:`iodepth_batch_complete_min`\=int variable,
2061 specifying the range of min and max amount of I/O which should be
2062 retrieved. By default it is equal to the :option:`iodepth_batch_complete_min`
2067 iodepth_batch_complete_min=1
2068 iodepth_batch_complete_max=<iodepth>
2070 which means that we will retrieve at least 1 I/O and up to the whole
2071 submitted queue depth. If none of I/O has been completed yet, we will wait.
2075 iodepth_batch_complete_min=0
2076 iodepth_batch_complete_max=<iodepth>
2078 which means that we can retrieve up to the whole submitted queue depth, but
2079 if none of I/O has been completed yet, we will NOT wait and immediately exit
2080 the system call. In this example we simply do polling.
2082 .. option:: iodepth_low=int
2084 The low water mark indicating when to start filling the queue
2085 again. Defaults to the same as :option:`iodepth`, meaning that fio will
2086 attempt to keep the queue full at all times. If :option:`iodepth` is set to
2087 e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of
2088 16 requests, it will let the depth drain down to 4 before starting to fill
2091 .. option:: serialize_overlap=bool
2093 Serialize in-flight I/Os that might otherwise cause or suffer from data races.
2094 When two or more I/Os are submitted simultaneously, there is no guarantee that
2095 the I/Os will be processed or completed in the submitted order. Further, if
2096 two or more of those I/Os are writes, any overlapping region between them can
2097 become indeterminate/undefined on certain storage. These issues can cause
2098 verification to fail erratically when at least one of the racing I/Os is
2099 changing data and the overlapping region has a non-zero size. Setting
2100 ``serialize_overlap`` tells fio to avoid provoking this behavior by explicitly
2101 serializing in-flight I/Os that have a non-zero overlap. Note that setting
2102 this option can reduce both performance and the :option:`iodepth` achieved.
2103 Additionally this option does not work when :option:`io_submit_mode` is set to
2104 offload. Default: false.
2106 .. option:: io_submit_mode=str
2108 This option controls how fio submits the I/O to the I/O engine. The default
2109 is `inline`, which means that the fio job threads submit and reap I/O
2110 directly. If set to `offload`, the job threads will offload I/O submission
2111 to a dedicated pool of I/O threads. This requires some coordination and thus
2112 has a bit of extra overhead, especially for lower queue depth I/O where it
2113 can increase latencies. The benefit is that fio can manage submission rates
2114 independently of the device completion rates. This avoids skewed latency
2115 reporting if I/O gets backed up on the device side (the coordinated omission
2122 .. option:: thinktime=time
2124 Stall the job for the specified period of time after an I/O has completed before issuing the
2125 next. May be used to simulate processing being done by an application.
2126 When the unit is omitted, the value is interpreted in microseconds. See
2127 :option:`thinktime_blocks` and :option:`thinktime_spin`.
2129 .. option:: thinktime_spin=time
2131 Only valid if :option:`thinktime` is set - pretend to spend CPU time doing
2132 something with the data received, before falling back to sleeping for the
2133 rest of the period specified by :option:`thinktime`. When the unit is
2134 omitted, the value is interpreted in microseconds.
2136 .. option:: thinktime_blocks=int
2138 Only valid if :option:`thinktime` is set - control how many blocks to issue,
2139 before waiting :option:`thinktime` usecs. If not set, defaults to 1 which will make
2140 fio wait :option:`thinktime` usecs after every block. This effectively makes any
2141 queue depth setting redundant, since no more than 1 I/O will be queued
2142 before we have to complete it and do our :option:`thinktime`. In other words, this
2143 setting effectively caps the queue depth if the latter is larger.
2145 .. option:: rate=int[,int][,int]
2147 Cap the bandwidth used by this job. The number is in bytes/sec, the normal
2148 suffix rules apply. Comma-separated values may be specified for reads,
2149 writes, and trims as described in :option:`blocksize`.
2151 For example, using `rate=1m,500k` would limit reads to 1MiB/sec and writes to
2152 500KiB/sec. Capping only reads or writes can be done with `rate=,500k` or
2153 `rate=500k,` where the former will only limit writes (to 500KiB/sec) and the
2154 latter will only limit reads.
2156 .. option:: rate_min=int[,int][,int]
2158 Tell fio to do whatever it can to maintain at least this bandwidth. Failing
2159 to meet this requirement will cause the job to exit. Comma-separated values
2160 may be specified for reads, writes, and trims as described in
2161 :option:`blocksize`.
2163 .. option:: rate_iops=int[,int][,int]
2165 Cap the bandwidth to this number of IOPS. Basically the same as
2166 :option:`rate`, just specified independently of bandwidth. If the job is
2167 given a block size range instead of a fixed value, the smallest block size
2168 is used as the metric. Comma-separated values may be specified for reads,
2169 writes, and trims as described in :option:`blocksize`.
2171 .. option:: rate_iops_min=int[,int][,int]
2173 If fio doesn't meet this rate of I/O, it will cause the job to exit.
2174 Comma-separated values may be specified for reads, writes, and trims as
2175 described in :option:`blocksize`.
2177 .. option:: rate_process=str
2179 This option controls how fio manages rated I/O submissions. The default is
2180 `linear`, which submits I/O in a linear fashion with fixed delays between
2181 I/Os that gets adjusted based on I/O completion rates. If this is set to
2182 `poisson`, fio will submit I/O based on a more real world random request
2183 flow, known as the Poisson process
2184 (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be
2185 10^6 / IOPS for the given workload.
2191 .. option:: latency_target=time
2193 If set, fio will attempt to find the max performance point that the given
2194 workload will run at while maintaining a latency below this target. When
2195 the unit is omitted, the value is interpreted in microseconds. See
2196 :option:`latency_window` and :option:`latency_percentile`.
2198 .. option:: latency_window=time
2200 Used with :option:`latency_target` to specify the sample window that the job
2201 is run at varying queue depths to test the performance. When the unit is
2202 omitted, the value is interpreted in microseconds.
2204 .. option:: latency_percentile=float
2206 The percentage of I/Os that must fall within the criteria specified by
2207 :option:`latency_target` and :option:`latency_window`. If not set, this
2208 defaults to 100.0, meaning that all I/Os must be equal or below to the value
2209 set by :option:`latency_target`.
2211 .. option:: max_latency=time
2213 If set, fio will exit the job with an ETIMEDOUT error if it exceeds this
2214 maximum latency. When the unit is omitted, the value is interpreted in
2217 .. option:: rate_cycle=int
2219 Average bandwidth for :option:`rate` and :option:`rate_min` over this number
2220 of milliseconds. Defaults to 1000.
2226 .. option:: write_iolog=str
2228 Write the issued I/O patterns to the specified file. See
2229 :option:`read_iolog`. Specify a separate file for each job, otherwise the
2230 iologs will be interspersed and the file may be corrupt.
2232 .. option:: read_iolog=str
2234 Open an iolog with the specified filename and replay the I/O patterns it
2235 contains. This can be used to store a workload and replay it sometime
2236 later. The iolog given may also be a blktrace binary file, which allows fio
2237 to replay a workload captured by :command:`blktrace`. See
2238 :manpage:`blktrace(8)` for how to capture such logging data. For blktrace
2239 replay, the file needs to be turned into a blkparse binary data file first
2240 (``blkparse <device> -o /dev/null -d file_for_fio.bin``).
2242 .. option:: replay_no_stall=bool
2244 When replaying I/O with :option:`read_iolog` the default behavior is to
2245 attempt to respect the timestamps within the log and replay them with the
2246 appropriate delay between IOPS. By setting this variable fio will not
2247 respect the timestamps and attempt to replay them as fast as possible while
2248 still respecting ordering. The result is the same I/O pattern to a given
2249 device, but different timings.
2251 .. option:: replay_redirect=str
2253 While replaying I/O patterns using :option:`read_iolog` the default behavior
2254 is to replay the IOPS onto the major/minor device that each IOP was recorded
2255 from. This is sometimes undesirable because on a different machine those
2256 major/minor numbers can map to a different device. Changing hardware on the
2257 same system can also result in a different major/minor mapping.
2258 ``replay_redirect`` causes all I/Os to be replayed onto the single specified
2259 device regardless of the device it was recorded
2260 from. i.e. :option:`replay_redirect`\= :file:`/dev/sdc` would cause all I/O
2261 in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means
2262 multiple devices will be replayed onto a single device, if the trace
2263 contains multiple devices. If you want multiple devices to be replayed
2264 concurrently to multiple redirected devices you must blkparse your trace
2265 into separate traces and replay them with independent fio invocations.
2266 Unfortunately this also breaks the strict time ordering between multiple
2269 .. option:: replay_align=int
2271 Force alignment of I/O offsets and lengths in a trace to this power of 2
2274 .. option:: replay_scale=int
2276 Scale sector offsets down by this factor when replaying traces.
2279 Threads, processes and job synchronization
2280 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2284 Fio defaults to creating jobs by using fork, however if this option is
2285 given, fio will create jobs by using POSIX Threads' function
2286 :manpage:`pthread_create(3)` to create threads instead.
2288 .. option:: wait_for=str
2290 If set, the current job won't be started until all workers of the specified
2291 waitee job are done.
2293 ``wait_for`` operates on the job name basis, so there are a few
2294 limitations. First, the waitee must be defined prior to the waiter job
2295 (meaning no forward references). Second, if a job is being referenced as a
2296 waitee, it must have a unique name (no duplicate waitees).
2298 .. option:: nice=int
2300 Run the job with the given nice value. See man :manpage:`nice(2)`.
2302 On Windows, values less than -15 set the process class to "High"; -1 through
2303 -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle"
2306 .. option:: prio=int
2308 Set the I/O priority value of this job. Linux limits us to a positive value
2309 between 0 and 7, with 0 being the highest. See man
2310 :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating
2311 systems since meaning of priority may differ.
2313 .. option:: prioclass=int
2315 Set the I/O priority class. See man :manpage:`ionice(1)`.
2317 .. option:: cpumask=int
2319 Set the CPU affinity of this job. The parameter given is a bit mask of
2320 allowed CPUs the job may run on. So if you want the allowed CPUs to be 1
2321 and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man
2322 :manpage:`sched_setaffinity(2)`. This may not work on all supported
2323 operating systems or kernel versions. This option doesn't work well for a
2324 higher CPU count than what you can store in an integer mask, so it can only
2325 control cpus 1-32. For boxes with larger CPU counts, use
2326 :option:`cpus_allowed`.
2328 .. option:: cpus_allowed=str
2330 Controls the same options as :option:`cpumask`, but accepts a textual
2331 specification of the permitted CPUs instead. So to use CPUs 1 and 5 you
2332 would specify ``cpus_allowed=1,5``. This option also allows a range of CPUs
2333 to be specified -- say you wanted a binding to CPUs 1, 5, and 8 to 15, you
2334 would set ``cpus_allowed=1,5,8-15``.
2336 .. option:: cpus_allowed_policy=str
2338 Set the policy of how fio distributes the CPUs specified by
2339 :option:`cpus_allowed` or :option:`cpumask`. Two policies are supported:
2342 All jobs will share the CPU set specified.
2344 Each job will get a unique CPU from the CPU set.
2346 **shared** is the default behavior, if the option isn't specified. If
2347 **split** is specified, then fio will will assign one cpu per job. If not
2348 enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs
2351 .. option:: numa_cpu_nodes=str
2353 Set this job running on specified NUMA nodes' CPUs. The arguments allow
2354 comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable
2355 NUMA options support, fio must be built on a system with libnuma-dev(el)
2358 .. option:: numa_mem_policy=str
2360 Set this job's memory policy and corresponding NUMA nodes. Format of the
2365 ``mode`` is one of the following memory poicies: ``default``, ``prefer``,
2366 ``bind``, ``interleave`` or ``local``. For ``default`` and ``local`` memory
2367 policies, no node needs to be specified. For ``prefer``, only one node is
2368 allowed. For ``bind`` and ``interleave`` the ``nodelist`` may be as
2369 follows: a comma delimited list of numbers, A-B ranges, or `all`.
2371 .. option:: cgroup=str
2373 Add job to this control group. If it doesn't exist, it will be created. The
2374 system must have a mounted cgroup blkio mount point for this to work. If
2375 your system doesn't have it mounted, you can do so with::
2377 # mount -t cgroup -o blkio none /cgroup
2379 .. option:: cgroup_weight=int
2381 Set the weight of the cgroup to this value. See the documentation that comes
2382 with the kernel, allowed values are in the range of 100..1000.
2384 .. option:: cgroup_nodelete=bool
2386 Normally fio will delete the cgroups it has created after the job
2387 completion. To override this behavior and to leave cgroups around after the
2388 job completion, set ``cgroup_nodelete=1``. This can be useful if one wants
2389 to inspect various cgroup files after job completion. Default: false.
2391 .. option:: flow_id=int
2393 The ID of the flow. If not specified, it defaults to being a global
2394 flow. See :option:`flow`.
2396 .. option:: flow=int
2398 Weight in token-based flow control. If this value is used, then there is a
2399 'flow counter' which is used to regulate the proportion of activity between
2400 two or more jobs. Fio attempts to keep this flow counter near zero. The
2401 ``flow`` parameter stands for how much should be added or subtracted to the
2402 flow counter on each iteration of the main I/O loop. That is, if one job has
2403 ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8
2404 ratio in how much one runs vs the other.
2406 .. option:: flow_watermark=int
2408 The maximum value that the absolute value of the flow counter is allowed to
2409 reach before the job must wait for a lower value of the counter.
2411 .. option:: flow_sleep=int
2413 The period of time, in microseconds, to wait after the flow watermark has
2414 been exceeded before retrying operations.
2416 .. option:: stonewall, wait_for_previous
2418 Wait for preceding jobs in the job file to exit, before starting this
2419 one. Can be used to insert serialization points in the job file. A stone
2420 wall also implies starting a new reporting group, see
2421 :option:`group_reporting`.
2425 By default, fio will continue running all other jobs when one job finishes
2426 but sometimes this is not the desired action. Setting ``exitall`` will
2427 instead make fio terminate all other jobs when one job finishes.
2429 .. option:: exec_prerun=str
2431 Before running this job, issue the command specified through
2432 :manpage:`system(3)`. Output is redirected in a file called
2433 :file:`jobname.prerun.txt`.
2435 .. option:: exec_postrun=str
2437 After the job completes, issue the command specified though
2438 :manpage:`system(3)`. Output is redirected in a file called
2439 :file:`jobname.postrun.txt`.
2443 Instead of running as the invoking user, set the user ID to this value
2444 before the thread/process does any work.
2448 Set group ID, see :option:`uid`.
2454 .. option:: verify_only
2456 Do not perform specified workload, only verify data still matches previous
2457 invocation of this workload. This option allows one to check data multiple
2458 times at a later date without overwriting it. This option makes sense only
2459 for workloads that write data, and does not support workloads with the
2460 :option:`time_based` option set.
2462 .. option:: do_verify=bool
2464 Run the verify phase after a write phase. Only valid if :option:`verify` is
2467 .. option:: verify=str
2469 If writing to a file, fio can verify the file contents after each iteration
2470 of the job. Each verification method also implies verification of special
2471 header, which is written to the beginning of each block. This header also
2472 includes meta information, like offset of the block, block number, timestamp
2473 when block was written, etc. :option:`verify` can be combined with
2474 :option:`verify_pattern` option. The allowed values are:
2477 Use an md5 sum of the data area and store it in the header of
2481 Use an experimental crc64 sum of the data area and store it in the
2482 header of each block.
2485 Use a crc32c sum of the data area and store it in the header of
2486 each block. This will automatically use hardware acceleration
2487 (e.g. SSE4.2 on an x86 or CRC crypto extensions on ARM64) but will
2488 fall back to software crc32c if none is found. Generally the
2489 fatest checksum fio supports when hardware accelerated.
2495 Use a crc32 sum of the data area and store it in the header of each
2499 Use a crc16 sum of the data area and store it in the header of each
2503 Use a crc7 sum of the data area and store it in the header of each
2507 Use xxhash as the checksum function. Generally the fastest software
2508 checksum that fio supports.
2511 Use sha512 as the checksum function.
2514 Use sha256 as the checksum function.
2517 Use optimized sha1 as the checksum function.
2520 Use optimized sha3-224 as the checksum function.
2523 Use optimized sha3-256 as the checksum function.
2526 Use optimized sha3-384 as the checksum function.
2529 Use optimized sha3-512 as the checksum function.
2532 This option is deprecated, since now meta information is included in
2533 generic verification header and meta verification happens by
2534 default. For detailed information see the description of the
2535 :option:`verify` setting. This option is kept because of
2536 compatibility's sake with old configurations. Do not use it.
2539 Verify a strict pattern. Normally fio includes a header with some
2540 basic information and checksumming, but if this option is set, only
2541 the specific pattern set with :option:`verify_pattern` is verified.
2544 Only pretend to verify. Useful for testing internals with
2545 :option:`ioengine`\=null, not for much else.
2547 This option can be used for repeated burn-in tests of a system to make sure
2548 that the written data is also correctly read back. If the data direction
2549 given is a read or random read, fio will assume that it should verify a
2550 previously written file. If the data direction includes any form of write,
2551 the verify will be of the newly written data.
2553 .. option:: verifysort=bool
2555 If true, fio will sort written verify blocks when it deems it faster to read
2556 them back in a sorted manner. This is often the case when overwriting an
2557 existing file, since the blocks are already laid out in the file system. You
2558 can ignore this option unless doing huge amounts of really fast I/O where
2559 the red-black tree sorting CPU time becomes significant. Default: true.
2561 .. option:: verifysort_nr=int
2563 Pre-load and sort verify blocks for a read workload.
2565 .. option:: verify_offset=int
2567 Swap the verification header with data somewhere else in the block before
2568 writing. It is swapped back before verifying.
2570 .. option:: verify_interval=int
2572 Write the verification header at a finer granularity than the
2573 :option:`blocksize`. It will be written for chunks the size of
2574 ``verify_interval``. :option:`blocksize` should divide this evenly.
2576 .. option:: verify_pattern=str
2578 If set, fio will fill the I/O buffers with this pattern. Fio defaults to
2579 filling with totally random bytes, but sometimes it's interesting to fill
2580 with a known pattern for I/O verification purposes. Depending on the width
2581 of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time (it can
2582 be either a decimal or a hex number). The ``verify_pattern`` if larger than
2583 a 32-bit quantity has to be a hex number that starts with either "0x" or
2584 "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o
2585 format, which means that for each block offset will be written and then
2586 verified back, e.g.::
2590 Or use combination of everything::
2592 verify_pattern=0xff%o"abcd"-12
2594 .. option:: verify_fatal=bool
2596 Normally fio will keep checking the entire contents before quitting on a
2597 block verification failure. If this option is set, fio will exit the job on
2598 the first observed failure. Default: false.
2600 .. option:: verify_dump=bool
2602 If set, dump the contents of both the original data block and the data block
2603 we read off disk to files. This allows later analysis to inspect just what
2604 kind of data corruption occurred. Off by default.
2606 .. option:: verify_async=int
2608 Fio will normally verify I/O inline from the submitting thread. This option
2609 takes an integer describing how many async offload threads to create for I/O
2610 verification instead, causing fio to offload the duty of verifying I/O
2611 contents to one or more separate threads. If using this offload option, even
2612 sync I/O engines can benefit from using an :option:`iodepth` setting higher
2613 than 1, as it allows them to have I/O in flight while verifies are running.
2614 Defaults to 0 async threads, i.e. verification is not asynchronous.
2616 .. option:: verify_async_cpus=str
2618 Tell fio to set the given CPU affinity on the async I/O verification
2619 threads. See :option:`cpus_allowed` for the format used.
2621 .. option:: verify_backlog=int
2623 Fio will normally verify the written contents of a job that utilizes verify
2624 once that job has completed. In other words, everything is written then
2625 everything is read back and verified. You may want to verify continually
2626 instead for a variety of reasons. Fio stores the meta data associated with
2627 an I/O block in memory, so for large verify workloads, quite a bit of memory
2628 would be used up holding this meta data. If this option is enabled, fio will
2629 write only N blocks before verifying these blocks.
2631 .. option:: verify_backlog_batch=int
2633 Control how many blocks fio will verify if :option:`verify_backlog` is
2634 set. If not set, will default to the value of :option:`verify_backlog`
2635 (meaning the entire queue is read back and verified). If
2636 ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all
2637 blocks will be verified, if ``verify_backlog_batch`` is larger than
2638 :option:`verify_backlog`, some blocks will be verified more than once.
2640 .. option:: verify_state_save=bool
2642 When a job exits during the write phase of a verify workload, save its
2643 current state. This allows fio to replay up until that point, if the verify
2644 state is loaded for the verify read phase. The format of the filename is,
2647 <type>-<jobname>-<jobindex>-verify.state.
2649 <type> is "local" for a local run, "sock" for a client/server socket
2650 connection, and "ip" (192.168.0.1, for instance) for a networked
2651 client/server connection. Defaults to true.
2653 .. option:: verify_state_load=bool
2655 If a verify termination trigger was used, fio stores the current write state
2656 of each thread. This can be used at verification time so that fio knows how
2657 far it should verify. Without this information, fio will run a full
2658 verification pass, according to the settings in the job file used. Default
2661 .. option:: trim_percentage=int
2663 Number of verify blocks to discard/trim.
2665 .. option:: trim_verify_zero=bool
2667 Verify that trim/discarded blocks are returned as zeros.
2669 .. option:: trim_backlog=int
2671 Trim after this number of blocks are written.
2673 .. option:: trim_backlog_batch=int
2675 Trim this number of I/O blocks.
2677 .. option:: experimental_verify=bool
2679 Enable experimental verification.
2684 .. option:: steadystate=str:float, ss=str:float
2686 Define the criterion and limit for assessing steady state performance. The
2687 first parameter designates the criterion whereas the second parameter sets
2688 the threshold. When the criterion falls below the threshold for the
2689 specified duration, the job will stop. For example, `iops_slope:0.1%` will
2690 direct fio to terminate the job when the least squares regression slope
2691 falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled
2692 this will apply to all jobs in the group. Below is the list of available
2693 steady state assessment criteria. All assessments are carried out using only
2694 data from the rolling collection window. Threshold limits can be expressed
2695 as a fixed value or as a percentage of the mean in the collection window.
2698 Collect IOPS data. Stop the job if all individual IOPS measurements
2699 are within the specified limit of the mean IOPS (e.g., ``iops:2``
2700 means that all individual IOPS values must be within 2 of the mean,
2701 whereas ``iops:0.2%`` means that all individual IOPS values must be
2702 within 0.2% of the mean IOPS to terminate the job).
2705 Collect IOPS data and calculate the least squares regression
2706 slope. Stop the job if the slope falls below the specified limit.
2709 Collect bandwidth data. Stop the job if all individual bandwidth
2710 measurements are within the specified limit of the mean bandwidth.
2713 Collect bandwidth data and calculate the least squares regression
2714 slope. Stop the job if the slope falls below the specified limit.
2716 .. option:: steadystate_duration=time, ss_dur=time
2718 A rolling window of this duration will be used to judge whether steady state
2719 has been reached. Data will be collected once per second. The default is 0
2720 which disables steady state detection. When the unit is omitted, the
2721 value is interpreted in seconds.
2723 .. option:: steadystate_ramp_time=time, ss_ramp=time
2725 Allow the job to run for the specified duration before beginning data
2726 collection for checking the steady state job termination criterion. The
2727 default is 0. When the unit is omitted, the value is interpreted in seconds.
2730 Measurements and reporting
2731 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2733 .. option:: per_job_logs=bool
2735 If set, this generates bw/clat/iops log with per file private filenames. If
2736 not set, jobs with identical names will share the log filename. Default:
2739 .. option:: group_reporting
2741 It may sometimes be interesting to display statistics for groups of jobs as
2742 a whole instead of for each individual job. This is especially true if
2743 :option:`numjobs` is used; looking at individual thread/process output
2744 quickly becomes unwieldy. To see the final report per-group instead of
2745 per-job, use :option:`group_reporting`. Jobs in a file will be part of the
2746 same reporting group, unless if separated by a :option:`stonewall`, or by
2747 using :option:`new_group`.
2749 .. option:: new_group
2751 Start a new reporting group. See: :option:`group_reporting`. If not given,
2752 all jobs in a file will be part of the same reporting group, unless
2753 separated by a :option:`stonewall`.
2755 .. option:: stats=bool
2757 By default, fio collects and shows final output results for all jobs
2758 that run. If this option is set to 0, then fio will ignore it in
2759 the final stat output.
2761 .. option:: write_bw_log=str
2763 If given, write a bandwidth log for this job. Can be used to store data of
2764 the bandwidth of the jobs in their lifetime.
2766 If no str argument is given, the default filename of
2767 :file:`jobname_type.x.log` is used. Even when the argument is given, fio
2768 will still append the type of log. So if one specifies::
2772 The actual log name will be :file:`foo_bw.x.log` where `x` is the index
2773 of the job (`1..N`, where `N` is the number of jobs). If
2774 :option:`per_job_logs` is false, then the filename will not include the
2777 The included :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these
2778 text files into nice graphs. See `Log File Formats`_ for how data is
2779 structured within the file.
2781 .. option:: write_lat_log=str
2783 Same as :option:`write_bw_log`, except this option creates I/O
2784 submission (e.g., :file:`name_slat.x.log`), completion (e.g.,
2785 :file:`name_clat.x.log`), and total (e.g., :file:`name_lat.x.log`)
2786 latency files instead. See :option:`write_bw_log` for details about
2787 the filename format and `Log File Formats`_ for how data is structured
2790 .. option:: write_hist_log=str
2792 Same as :option:`write_bw_log` but writes an I/O completion latency
2793 histogram file (e.g., :file:`name_hist.x.log`) instead. Note that this
2794 file will be empty unless :option:`log_hist_msec` has also been set.
2795 See :option:`write_bw_log` for details about the filename format and
2796 `Log File Formats`_ for how data is structured within the file.
2798 .. option:: write_iops_log=str
2800 Same as :option:`write_bw_log`, but writes an IOPS file (e.g.
2801 :file:`name_iops.x.log`) instead. See :option:`write_bw_log` for
2802 details about the filename format and `Log File Formats`_ for how data
2803 is structured within the file.
2805 .. option:: log_avg_msec=int
2807 By default, fio will log an entry in the iops, latency, or bw log for every
2808 I/O that completes. When writing to the disk log, that can quickly grow to a
2809 very large size. Setting this option makes fio average the each log entry
2810 over the specified period of time, reducing the resolution of the log. See
2811 :option:`log_max_value` as well. Defaults to 0, logging all entries.
2812 Also see `Log File Formats`_.
2814 .. option:: log_hist_msec=int
2816 Same as :option:`log_avg_msec`, but logs entries for completion latency
2817 histograms. Computing latency percentiles from averages of intervals using
2818 :option:`log_avg_msec` is inaccurate. Setting this option makes fio log
2819 histogram entries over the specified period of time, reducing log sizes for
2820 high IOPS devices while retaining percentile accuracy. See
2821 :option:`log_hist_coarseness` and :option:`write_hist_log` as well.
2822 Defaults to 0, meaning histogram logging is disabled.
2824 .. option:: log_hist_coarseness=int
2826 Integer ranging from 0 to 6, defining the coarseness of the resolution of
2827 the histogram logs enabled with :option:`log_hist_msec`. For each increment
2828 in coarseness, fio outputs half as many bins. Defaults to 0, for which
2829 histogram logs contain 1216 latency bins. See :option:`write_hist_log`
2830 and `Log File Formats`_.
2832 .. option:: log_max_value=bool
2834 If :option:`log_avg_msec` is set, fio logs the average over that window. If
2835 you instead want to log the maximum value, set this option to 1. Defaults to
2836 0, meaning that averaged values are logged.
2838 .. option:: log_offset=bool
2840 If this is set, the iolog options will include the byte offset for the I/O
2841 entry as well as the other data values. Defaults to 0 meaning that
2842 offsets are not present in logs. Also see `Log File Formats`_.
2844 .. option:: log_compression=int
2846 If this is set, fio will compress the I/O logs as it goes, to keep the
2847 memory footprint lower. When a log reaches the specified size, that chunk is
2848 removed and compressed in the background. Given that I/O logs are fairly
2849 highly compressible, this yields a nice memory savings for longer runs. The
2850 downside is that the compression will consume some background CPU cycles, so
2851 it may impact the run. This, however, is also true if the logging ends up
2852 consuming most of the system memory. So pick your poison. The I/O logs are
2853 saved normally at the end of a run, by decompressing the chunks and storing
2854 them in the specified log file. This feature depends on the availability of
2857 .. option:: log_compression_cpus=str
2859 Define the set of CPUs that are allowed to handle online log compression for
2860 the I/O jobs. This can provide better isolation between performance
2861 sensitive jobs, and background compression work.
2863 .. option:: log_store_compressed=bool
2865 If set, fio will store the log files in a compressed format. They can be
2866 decompressed with fio, using the :option:`--inflate-log` command line
2867 parameter. The files will be stored with a :file:`.fz` suffix.
2869 .. option:: log_unix_epoch=bool
2871 If set, fio will log Unix timestamps to the log files produced by enabling
2872 write_type_log for each log type, instead of the default zero-based
2875 .. option:: block_error_percentiles=bool
2877 If set, record errors in trim block-sized units from writes and trims and
2878 output a histogram of how many trims it took to get to errors, and what kind
2879 of error was encountered.
2881 .. option:: bwavgtime=int
2883 Average the calculated bandwidth over the given time. Value is specified in
2884 milliseconds. If the job also does bandwidth logging through
2885 :option:`write_bw_log`, then the minimum of this option and
2886 :option:`log_avg_msec` will be used. Default: 500ms.
2888 .. option:: iopsavgtime=int
2890 Average the calculated IOPS over the given time. Value is specified in
2891 milliseconds. If the job also does IOPS logging through
2892 :option:`write_iops_log`, then the minimum of this option and
2893 :option:`log_avg_msec` will be used. Default: 500ms.
2895 .. option:: disk_util=bool
2897 Generate disk utilization statistics, if the platform supports it.
2900 .. option:: disable_lat=bool
2902 Disable measurements of total latency numbers. Useful only for cutting back
2903 the number of calls to :manpage:`gettimeofday(2)`, as that does impact
2904 performance at really high IOPS rates. Note that to really get rid of a
2905 large amount of these calls, this option must be used with
2906 :option:`disable_slat` and :option:`disable_bw_measurement` as well.
2908 .. option:: disable_clat=bool
2910 Disable measurements of completion latency numbers. See
2911 :option:`disable_lat`.
2913 .. option:: disable_slat=bool
2915 Disable measurements of submission latency numbers. See
2916 :option:`disable_lat`.
2918 .. option:: disable_bw_measurement=bool, disable_bw=bool
2920 Disable measurements of throughput/bandwidth numbers. See
2921 :option:`disable_lat`.
2923 .. option:: clat_percentiles=bool
2925 Enable the reporting of percentiles of completion latencies. This
2926 option is mutually exclusive with :option:`lat_percentiles`.
2928 .. option:: lat_percentiles=bool
2930 Enable the reporting of percentiles of I/O latencies. This is similar
2931 to :option:`clat_percentiles`, except that this includes the
2932 submission latency. This option is mutually exclusive with
2933 :option:`clat_percentiles`.
2935 .. option:: percentile_list=float_list
2937 Overwrite the default list of percentiles for completion latencies and the
2938 block error histogram. Each number is a floating number in the range
2939 (0,100], and the maximum length of the list is 20. Use ``:`` to separate the
2940 numbers, and list the numbers in ascending order. For example,
2941 ``--percentile_list=99.5:99.9`` will cause fio to report the values of
2942 completion latency below which 99.5% and 99.9% of the observed latencies
2945 .. option:: significant_figures=int
2947 If using :option:`--output-format` of `normal`, set the significant figures
2948 to this value. Higher values will yield more precise IOPS and throughput
2949 units, while lower values will round. Requires a minimum value of 1 and a
2950 maximum value of 10. Defaults to 4.
2956 .. option:: exitall_on_error
2958 When one job finishes in error, terminate the rest. The default is to wait
2959 for each job to finish.
2961 .. option:: continue_on_error=str
2963 Normally fio will exit the job on the first observed failure. If this option
2964 is set, fio will continue the job when there is a 'non-fatal error' (EIO or
2965 EILSEQ) until the runtime is exceeded or the I/O size specified is
2966 completed. If this option is used, there are two more stats that are
2967 appended, the total error count and the first error. The error field given
2968 in the stats is the first error that was hit during the run.
2970 The allowed values are:
2973 Exit on any I/O or verify errors.
2976 Continue on read errors, exit on all others.
2979 Continue on write errors, exit on all others.
2982 Continue on any I/O error, exit on all others.
2985 Continue on verify errors, exit on all others.
2988 Continue on all errors.
2991 Backward-compatible alias for 'none'.
2994 Backward-compatible alias for 'all'.
2996 .. option:: ignore_error=str
2998 Sometimes you want to ignore some errors during test in that case you can
2999 specify error list for each error type, instead of only being able to
3000 ignore the default 'non-fatal error' using :option:`continue_on_error`.
3001 ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for
3002 given error type is separated with ':'. Error may be symbol ('ENOSPC',
3003 'ENOMEM') or integer. Example::
3005 ignore_error=EAGAIN,ENOSPC:122
3007 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from
3008 WRITE. This option works by overriding :option:`continue_on_error` with
3009 the list of errors for each error type if any.
3011 .. option:: error_dump=bool
3013 If set dump every error even if it is non fatal, true by default. If
3014 disabled only fatal error will be dumped.
3016 Running predefined workloads
3017 ----------------------------
3019 Fio includes predefined profiles that mimic the I/O workloads generated by
3022 .. option:: profile=str
3024 The predefined workload to run. Current profiles are:
3027 Threaded I/O bench (tiotest/tiobench) like workload.
3030 Aerospike Certification Tool (ACT) like workload.
3032 To view a profile's additional options use :option:`--cmdhelp` after specifying
3033 the profile. For example::
3035 $ fio --profile=act --cmdhelp
3040 .. option:: device-names=str
3045 .. option:: load=int
3048 ACT load multiplier. Default: 1.
3050 .. option:: test-duration=time
3053 How long the entire test takes to run. When the unit is omitted, the value
3054 is given in seconds. Default: 24h.
3056 .. option:: threads-per-queue=int
3059 Number of read I/O threads per device. Default: 8.
3061 .. option:: read-req-num-512-blocks=int
3064 Number of 512B blocks to read at the time. Default: 3.
3066 .. option:: large-block-op-kbytes=int
3069 Size of large block ops in KiB (writes). Default: 131072.
3074 Set to run ACT prep phase.
3076 Tiobench profile options
3077 ~~~~~~~~~~~~~~~~~~~~~~~~
3079 .. option:: size=str
3084 .. option:: block=int
3087 Block size in bytes. Default: 4096.
3089 .. option:: numruns=int
3099 .. option:: threads=int
3104 Interpreting the output
3105 -----------------------
3108 Example output was based on the following:
3109 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --time_based \
3110 --rate=1256k --bs=14K --name=quick --runtime=1s --name=mixed \
3111 --runtime=2m --rw=rw
3113 Fio spits out a lot of output. While running, fio will display the status of the
3114 jobs created. An example of that would be::
3116 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]
3118 The characters inside the first set of square brackets denote the current status of
3119 each thread. The first character is the first job defined in the job file, and so
3120 forth. The possible values (in typical life cycle order) are:
3122 +------+-----+-----------------------------------------------------------+
3124 +======+=====+===========================================================+
3125 | P | | Thread setup, but not started. |
3126 +------+-----+-----------------------------------------------------------+
3127 | C | | Thread created. |
3128 +------+-----+-----------------------------------------------------------+
3129 | I | | Thread initialized, waiting or generating necessary data. |
3130 +------+-----+-----------------------------------------------------------+
3131 | | p | Thread running pre-reading file(s). |
3132 +------+-----+-----------------------------------------------------------+
3133 | | / | Thread is in ramp period. |
3134 +------+-----+-----------------------------------------------------------+
3135 | | R | Running, doing sequential reads. |
3136 +------+-----+-----------------------------------------------------------+
3137 | | r | Running, doing random reads. |
3138 +------+-----+-----------------------------------------------------------+
3139 | | W | Running, doing sequential writes. |
3140 +------+-----+-----------------------------------------------------------+
3141 | | w | Running, doing random writes. |
3142 +------+-----+-----------------------------------------------------------+
3143 | | M | Running, doing mixed sequential reads/writes. |
3144 +------+-----+-----------------------------------------------------------+
3145 | | m | Running, doing mixed random reads/writes. |
3146 +------+-----+-----------------------------------------------------------+
3147 | | D | Running, doing sequential trims. |
3148 +------+-----+-----------------------------------------------------------+
3149 | | d | Running, doing random trims. |
3150 +------+-----+-----------------------------------------------------------+
3151 | | F | Running, currently waiting for :manpage:`fsync(2)`. |
3152 +------+-----+-----------------------------------------------------------+
3153 | | V | Running, doing verification of written data. |
3154 +------+-----+-----------------------------------------------------------+
3155 | f | | Thread finishing. |
3156 +------+-----+-----------------------------------------------------------+
3157 | E | | Thread exited, not reaped by main thread yet. |
3158 +------+-----+-----------------------------------------------------------+
3159 | _ | | Thread reaped. |
3160 +------+-----+-----------------------------------------------------------+
3161 | X | | Thread reaped, exited with an error. |
3162 +------+-----+-----------------------------------------------------------+
3163 | K | | Thread reaped, exited due to signal. |
3164 +------+-----+-----------------------------------------------------------+
3167 Example output was based on the following:
3168 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --runtime=58m \
3169 --time_based --rate=2512k --bs=256K --numjobs=10 \
3170 --name=readers --rw=read --name=writers --rw=write
3172 Fio will condense the thread string as not to take up more space on the command
3173 line than needed. For instance, if you have 10 readers and 10 writers running,
3174 the output would look like this::
3176 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]
3178 Note that the status string is displayed in order, so it's possible to tell which of
3179 the jobs are currently doing what. In the example above this means that jobs 1--10
3180 are readers and 11--20 are writers.
3182 The other values are fairly self explanatory -- number of threads currently
3183 running and doing I/O, the number of currently open files (f=), the estimated
3184 completion percentage, the rate of I/O since last check (read speed listed first,
3185 then write speed and optionally trim speed) in terms of bandwidth and IOPS,
3186 and time to completion for the current running group. It's impossible to estimate
3187 runtime of the following groups (if any).
3190 Example output was based on the following:
3191 TZ=UTC fio --iodepth=16 --ioengine=posixaio --filename=/tmp/fiofile \
3192 --direct=1 --size=100M --time_based --runtime=50s --rate_iops=89 \
3193 --bs=7K --name=Client1 --rw=write
3195 When fio is done (or interrupted by :kbd:`Ctrl-C`), it will show the data for
3196 each thread, group of threads, and disks in that order. For each overall thread (or
3197 group) the output looks like::
3199 Client1: (groupid=0, jobs=1): err= 0: pid=16109: Sat Jun 24 12:07:54 2017
3200 write: IOPS=88, BW=623KiB/s (638kB/s)(30.4MiB/50032msec)
3201 slat (nsec): min=500, max=145500, avg=8318.00, stdev=4781.50
3202 clat (usec): min=170, max=78367, avg=4019.02, stdev=8293.31
3203 lat (usec): min=174, max=78375, avg=4027.34, stdev=8291.79
3204 clat percentiles (usec):
3205 | 1.00th=[ 302], 5.00th=[ 326], 10.00th=[ 343], 20.00th=[ 363],
3206 | 30.00th=[ 392], 40.00th=[ 404], 50.00th=[ 416], 60.00th=[ 445],
3207 | 70.00th=[ 816], 80.00th=[ 6718], 90.00th=[12911], 95.00th=[21627],
3208 | 99.00th=[43779], 99.50th=[51643], 99.90th=[68682], 99.95th=[72877],
3210 bw ( KiB/s): min= 532, max= 686, per=0.10%, avg=622.87, stdev=24.82, samples= 100
3211 iops : min= 76, max= 98, avg=88.98, stdev= 3.54, samples= 100
3212 lat (usec) : 250=0.04%, 500=64.11%, 750=4.81%, 1000=2.79%
3213 lat (msec) : 2=4.16%, 4=1.84%, 10=4.90%, 20=11.33%, 50=5.37%
3214 lat (msec) : 100=0.65%
3215 cpu : usr=0.27%, sys=0.18%, ctx=12072, majf=0, minf=21
3216 IO depths : 1=85.0%, 2=13.1%, 4=1.8%, 8=0.1%, 16=0.0%, 32=0.0%, >=64=0.0%
3217 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3218 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3219 issued rwt: total=0,4450,0, short=0,0,0, dropped=0,0,0
3220 latency : target=0, window=0, percentile=100.00%, depth=8
3222 The job name (or first job's name when using :option:`group_reporting`) is printed,
3223 along with the group id, count of jobs being aggregated, last error id seen (which
3224 is 0 when there are no errors), pid/tid of that thread and the time the job/group
3225 completed. Below are the I/O statistics for each data direction performed (showing
3226 writes in the example above). In the order listed, they denote:
3229 The string before the colon shows the I/O direction the statistics
3230 are for. **IOPS** is the average I/Os performed per second. **BW**
3231 is the average bandwidth rate shown as: value in power of 2 format
3232 (value in power of 10 format). The last two values show: (**total
3233 I/O performed** in power of 2 format / **runtime** of that thread).
3236 Submission latency (**min** being the minimum, **max** being the
3237 maximum, **avg** being the average, **stdev** being the standard
3238 deviation). This is the time it took to submit the I/O. For
3239 sync I/O this row is not displayed as the slat is really the
3240 completion latency (since queue/complete is one operation there).
3241 This value can be in nanoseconds, microseconds or milliseconds ---
3242 fio will choose the most appropriate base and print that (in the
3243 example above nanoseconds was the best scale). Note: in :option:`--minimal` mode
3244 latencies are always expressed in microseconds.
3247 Completion latency. Same names as slat, this denotes the time from
3248 submission to completion of the I/O pieces. For sync I/O, clat will
3249 usually be equal (or very close) to 0, as the time from submit to
3250 complete is basically just CPU time (I/O has already been done, see slat
3254 Total latency. Same names as slat and clat, this denotes the time from
3255 when fio created the I/O unit to completion of the I/O operation.
3258 Bandwidth statistics based on samples. Same names as the xlat stats,
3259 but also includes the number of samples taken (**samples**) and an
3260 approximate percentage of total aggregate bandwidth this thread
3261 received in its group (**per**). This last value is only really
3262 useful if the threads in this group are on the same disk, since they
3263 are then competing for disk access.
3266 IOPS statistics based on samples. Same names as bw.
3268 **lat (nsec/usec/msec)**
3269 The distribution of I/O completion latencies. This is the time from when
3270 I/O leaves fio and when it gets completed. Unlike the separate
3271 read/write/trim sections above, the data here and in the remaining
3272 sections apply to all I/Os for the reporting group. 250=0.04% means that
3273 0.04% of the I/Os completed in under 250us. 500=64.11% means that 64.11%
3274 of the I/Os required 250 to 499us for completion.
3277 CPU usage. User and system time, along with the number of context
3278 switches this thread went through, usage of system and user time, and
3279 finally the number of major and minor page faults. The CPU utilization
3280 numbers are averages for the jobs in that reporting group, while the
3281 context and fault counters are summed.
3284 The distribution of I/O depths over the job lifetime. The numbers are
3285 divided into powers of 2 and each entry covers depths from that value
3286 up to those that are lower than the next entry -- e.g., 16= covers
3287 depths from 16 to 31. Note that the range covered by a depth
3288 distribution entry can be different to the range covered by the
3289 equivalent submit/complete distribution entry.
3292 How many pieces of I/O were submitting in a single submit call. Each
3293 entry denotes that amount and below, until the previous entry -- e.g.,
3294 16=100% means that we submitted anywhere between 9 to 16 I/Os per submit
3295 call. Note that the range covered by a submit distribution entry can
3296 be different to the range covered by the equivalent depth distribution
3300 Like the above submit number, but for completions instead.
3303 The number of read/write/trim requests issued, and how many of them were
3307 These values are for :option:`latency_target` and related options. When
3308 these options are engaged, this section describes the I/O depth required
3309 to meet the specified latency target.
3312 Example output was based on the following:
3313 TZ=UTC fio --ioengine=null --iodepth=2 --size=100M --numjobs=2 \
3314 --rate_process=poisson --io_limit=32M --name=read --bs=128k \
3315 --rate=11M --name=write --rw=write --bs=2k --rate=700k
3317 After each client has been listed, the group statistics are printed. They
3318 will look like this::
3320 Run status group 0 (all jobs):
3321 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
3322 WRITE: bw=1231KiB/s (1261kB/s), 616KiB/s-621KiB/s (630kB/s-636kB/s), io=64.0MiB (67.1MB), run=52747-53223msec
3324 For each data direction it prints:
3327 Aggregate bandwidth of threads in this group followed by the
3328 minimum and maximum bandwidth of all the threads in this group.
3329 Values outside of brackets are power-of-2 format and those
3330 within are the equivalent value in a power-of-10 format.
3332 Aggregate I/O performed of all threads in this group. The
3333 format is the same as bw.
3335 The smallest and longest runtimes of the threads in this group.
3337 And finally, the disk statistics are printed. This is Linux specific. They will look like this::
3339 Disk stats (read/write):
3340 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
3342 Each value is printed for both reads and writes, with reads first. The
3346 Number of I/Os performed by all groups.
3348 Number of merges performed by the I/O scheduler.
3350 Number of ticks we kept the disk busy.
3352 Total time spent in the disk queue.
3354 The disk utilization. A value of 100% means we kept the disk
3355 busy constantly, 50% would be a disk idling half of the time.
3357 It is also possible to get fio to dump the current output while it is running,
3358 without terminating the job. To do that, send fio the **USR1** signal. You can
3359 also get regularly timed dumps by using the :option:`--status-interval`
3360 parameter, or by creating a file in :file:`/tmp` named
3361 :file:`fio-dump-status`. If fio sees this file, it will unlink it and dump the
3362 current output status.
3368 For scripted usage where you typically want to generate tables or graphs of the
3369 results, fio can output the results in a semicolon separated format. The format
3370 is one long line of values, such as::
3372 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%
3373 A description of this job goes here.
3375 The job description (if provided) follows on a second line.
3377 To enable terse output, use the :option:`--minimal` or
3378 :option:`--output-format`\=terse command line options. The
3379 first value is the version of the terse output format. If the output has to be
3380 changed for some reason, this number will be incremented by 1 to signify that
3383 Split up, the format is as follows (comments in brackets denote when a
3384 field was introduced or whether it's specific to some terse version):
3388 terse version, fio version [v3], jobname, groupid, error
3392 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3393 Submission latency: min, max, mean, stdev (usec)
3394 Completion latency: min, max, mean, stdev (usec)
3395 Completion latency percentiles: 20 fields (see below)
3396 Total latency: min, max, mean, stdev (usec)
3397 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3398 IOPS [v5]: min, max, mean, stdev, number of samples
3404 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3405 Submission latency: min, max, mean, stdev (usec)
3406 Completion latency: min, max, mean, stdev (usec)
3407 Completion latency percentiles: 20 fields (see below)
3408 Total latency: min, max, mean, stdev (usec)
3409 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3410 IOPS [v5]: min, max, mean, stdev, number of samples
3412 TRIM status [all but version 3]:
3414 Fields are similar to READ/WRITE status.
3418 user, system, context switches, major faults, minor faults
3422 <=1, 2, 4, 8, 16, 32, >=64
3424 I/O latencies microseconds::
3426 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
3428 I/O latencies milliseconds::
3430 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
3432 Disk utilization [v3]::
3434 disk name, read ios, write ios, read merges, write merges, read ticks, write ticks,
3435 time spent in queue, disk utilization percentage
3437 Additional Info (dependent on continue_on_error, default off)::
3439 total # errors, first error code
3441 Additional Info (dependent on description being set)::
3445 Completion latency percentiles can be a grouping of up to 20 sets, so for the
3446 terse output fio writes all of them. Each field will look like this::
3450 which is the Xth percentile, and the `usec` latency associated with it.
3452 For `Disk utilization`, all disks used by fio are shown. So for each disk there
3453 will be a disk utilization section.
3455 Below is a single line containing short names for each of the fields in the
3456 minimal output v3, separated by semicolons::
3458 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
3464 The `json` output format is intended to be both human readable and convenient
3465 for automated parsing. For the most part its sections mirror those of the
3466 `normal` output. The `runtime` value is reported in msec and the `bw` value is
3467 reported in 1024 bytes per second units.
3473 The `json+` output format is identical to the `json` output format except that it
3474 adds a full dump of the completion latency bins. Each `bins` object contains a
3475 set of (key, value) pairs where keys are latency durations and values count how
3476 many I/Os had completion latencies of the corresponding duration. For example,
3479 "bins" : { "87552" : 1, "89600" : 1, "94720" : 1, "96768" : 1, "97792" : 1, "99840" : 1, "100864" : 2, "103936" : 6, "104960" : 534, "105984" : 5995, "107008" : 7529, ... }
3481 This data indicates that one I/O required 87,552ns to complete, two I/Os required
3482 100,864ns to complete, and 7529 I/Os required 107,008ns to complete.
3484 Also included with fio is a Python script `fio_jsonplus_clat2csv` that takes
3485 json+ output and generates CSV-formatted latency data suitable for plotting.
3487 The latency durations actually represent the midpoints of latency intervals.
3488 For details refer to :file:`stat.h`.
3494 There are two trace file format that you can encounter. The older (v1) format is
3495 unsupported since version 1.20-rc3 (March 2008). It will still be described
3496 below in case that you get an old trace and want to understand it.
3498 In any case the trace is a simple text file with a single action per line.
3501 Trace file format v1
3502 ~~~~~~~~~~~~~~~~~~~~
3504 Each line represents a single I/O action in the following format::
3508 where `rw=0/1` for read/write, and the `offset` and `length` entries being in bytes.
3510 This format is not supported in fio versions >= 1.20-rc3.
3513 Trace file format v2
3514 ~~~~~~~~~~~~~~~~~~~~
3516 The second version of the trace file format was added in fio version 1.17. It
3517 allows to access more then one file per trace and has a bigger set of possible
3520 The first line of the trace file has to be::
3524 Following this can be lines in two different formats, which are described below.
3526 The file management format::
3530 The `filename` is given as an absolute path. The `action` can be one of these:
3533 Add the given `filename` to the trace.
3535 Open the file with the given `filename`. The `filename` has to have
3536 been added with the **add** action before.
3538 Close the file with the given `filename`. The file has to have been
3542 The file I/O action format::
3544 filename action offset length
3546 The `filename` is given as an absolute path, and has to have been added and
3547 opened before it can be used with this format. The `offset` and `length` are
3548 given in bytes. The `action` can be one of these:
3551 Wait for `offset` microseconds. Everything below 100 is discarded.
3552 The time is relative to the previous `wait` statement.
3554 Read `length` bytes beginning from `offset`.
3556 Write `length` bytes beginning from `offset`.
3558 :manpage:`fsync(2)` the file.
3560 :manpage:`fdatasync(2)` the file.
3562 Trim the given file from the given `offset` for `length` bytes.
3564 CPU idleness profiling
3565 ----------------------
3567 In some cases, we want to understand CPU overhead in a test. For example, we
3568 test patches for the specific goodness of whether they reduce CPU usage.
3569 Fio implements a balloon approach to create a thread per CPU that runs at idle
3570 priority, meaning that it only runs when nobody else needs the cpu.
3571 By measuring the amount of work completed by the thread, idleness of each CPU
3572 can be derived accordingly.
3574 An unit work is defined as touching a full page of unsigned characters. Mean and
3575 standard deviation of time to complete an unit work is reported in "unit work"
3576 section. Options can be chosen to report detailed percpu idleness or overall
3577 system idleness by aggregating percpu stats.
3580 Verification and triggers
3581 -------------------------
3583 Fio is usually run in one of two ways, when data verification is done. The first
3584 is a normal write job of some sort with verify enabled. When the write phase has
3585 completed, fio switches to reads and verifies everything it wrote. The second
3586 model is running just the write phase, and then later on running the same job
3587 (but with reads instead of writes) to repeat the same I/O patterns and verify
3588 the contents. Both of these methods depend on the write phase being completed,
3589 as fio otherwise has no idea how much data was written.
3591 With verification triggers, fio supports dumping the current write state to
3592 local files. Then a subsequent read verify workload can load this state and know
3593 exactly where to stop. This is useful for testing cases where power is cut to a
3594 server in a managed fashion, for instance.
3596 A verification trigger consists of two things:
3598 1) Storing the write state of each job.
3599 2) Executing a trigger command.
3601 The write state is relatively small, on the order of hundreds of bytes to single
3602 kilobytes. It contains information on the number of completions done, the last X
3605 A trigger is invoked either through creation ('touch') of a specified file in
3606 the system, or through a timeout setting. If fio is run with
3607 :option:`--trigger-file`\= :file:`/tmp/trigger-file`, then it will continually
3608 check for the existence of :file:`/tmp/trigger-file`. When it sees this file, it
3609 will fire off the trigger (thus saving state, and executing the trigger
3612 For client/server runs, there's both a local and remote trigger. If fio is
3613 running as a server backend, it will send the job states back to the client for
3614 safe storage, then execute the remote trigger, if specified. If a local trigger
3615 is specified, the server will still send back the write state, but the client
3616 will then execute the trigger.
3618 Verification trigger example
3619 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3621 Let's say we want to run a powercut test on the remote Linux machine 'server'.
3622 Our write workload is in :file:`write-test.fio`. We want to cut power to 'server' at
3623 some point during the run, and we'll run this test from the safety or our local
3624 machine, 'localbox'. On the server, we'll start the fio backend normally::
3626 server# fio --server
3628 and on the client, we'll fire off the workload::
3630 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger-remote="bash -c \"echo b > /proc/sysrq-triger\""
3632 We set :file:`/tmp/my-trigger` as the trigger file, and we tell fio to execute::
3634 echo b > /proc/sysrq-trigger
3636 on the server once it has received the trigger and sent us the write state. This
3637 will work, but it's not **really** cutting power to the server, it's merely
3638 abruptly rebooting it. If we have a remote way of cutting power to the server
3639 through IPMI or similar, we could do that through a local trigger command
3640 instead. Let's assume we have a script that does IPMI reboot of a given hostname,
3641 ipmi-reboot. On localbox, we could then have run fio with a local trigger
3644 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger="ipmi-reboot server"
3646 For this case, fio would wait for the server to send us the write state, then
3647 execute ``ipmi-reboot server`` when that happened.
3649 Loading verify state
3650 ~~~~~~~~~~~~~~~~~~~~
3652 To load stored write state, a read verification job file must contain the
3653 :option:`verify_state_load` option. If that is set, fio will load the previously
3654 stored state. For a local fio run this is done by loading the files directly,
3655 and on a client/server run, the server backend will ask the client to send the
3656 files over and load them from there.
3662 Fio supports a variety of log file formats, for logging latencies, bandwidth,
3663 and IOPS. The logs share a common format, which looks like this:
3665 *time* (`msec`), *value*, *data direction*, *block size* (`bytes`),
3668 *Time* for the log entry is always in milliseconds. The *value* logged depends
3669 on the type of log, it will be one of the following:
3672 Value is latency in nsecs
3678 *Data direction* is one of the following:
3687 The entry's *block size* is always in bytes. The *offset* is the offset, in bytes,
3688 from the start of the file, for that particular I/O. The logging of the offset can be
3689 toggled with :option:`log_offset`.
3691 Fio defaults to logging every individual I/O. When IOPS are logged for individual
3692 I/Os the *value* entry will always be 1. If windowed logging is enabled through
3693 :option:`log_avg_msec`, fio logs the average values over the specified period of time.
3694 If windowed logging is enabled and :option:`log_max_value` is set, then fio logs
3695 maximum values in that window instead of averages. Since *data direction*, *block
3696 size* and *offset* are per-I/O values, if windowed logging is enabled they
3697 aren't applicable and will be 0.
3702 Normally fio is invoked as a stand-alone application on the machine where the
3703 I/O workload should be generated. However, the backend and frontend of fio can
3704 be run separately i.e., the fio server can generate an I/O workload on the "Device
3705 Under Test" while being controlled by a client on another machine.
3707 Start the server on the machine which has access to the storage DUT::
3711 where `args` defines what fio listens to. The arguments are of the form
3712 ``type,hostname`` or ``IP,port``. *type* is either ``ip`` (or ip4) for TCP/IP
3713 v4, ``ip6`` for TCP/IP v6, or ``sock`` for a local unix domain socket.
3714 *hostname* is either a hostname or IP address, and *port* is the port to listen
3715 to (only valid for TCP/IP, not a local socket). Some examples:
3719 Start a fio server, listening on all interfaces on the default port (8765).
3721 2) ``fio --server=ip:hostname,4444``
3723 Start a fio server, listening on IP belonging to hostname and on port 4444.
3725 3) ``fio --server=ip6:::1,4444``
3727 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
3729 4) ``fio --server=,4444``
3731 Start a fio server, listening on all interfaces on port 4444.
3733 5) ``fio --server=1.2.3.4``
3735 Start a fio server, listening on IP 1.2.3.4 on the default port.
3737 6) ``fio --server=sock:/tmp/fio.sock``
3739 Start a fio server, listening on the local socket :file:`/tmp/fio.sock`.
3741 Once a server is running, a "client" can connect to the fio server with::
3743 fio <local-args> --client=<server> <remote-args> <job file(s)>
3745 where `local-args` are arguments for the client where it is running, `server`
3746 is the connect string, and `remote-args` and `job file(s)` are sent to the
3747 server. The `server` string follows the same format as it does on the server
3748 side, to allow IP/hostname/socket and port strings.
3750 Fio can connect to multiple servers this way::
3752 fio --client=<server1> <job file(s)> --client=<server2> <job file(s)>
3754 If the job file is located on the fio server, then you can tell the server to
3755 load a local file as well. This is done by using :option:`--remote-config` ::
3757 fio --client=server --remote-config /path/to/file.fio
3759 Then fio will open this local (to the server) job file instead of being passed
3760 one from the client.
3762 If you have many servers (example: 100 VMs/containers), you can input a pathname
3763 of a file containing host IPs/names as the parameter value for the
3764 :option:`--client` option. For example, here is an example :file:`host.list`
3765 file containing 2 hostnames::
3767 host1.your.dns.domain
3768 host2.your.dns.domain
3770 The fio command would then be::
3772 fio --client=host.list <job file(s)>
3774 In this mode, you cannot input server-specific parameters or job files -- all
3775 servers receive the same job file.
3777 In order to let ``fio --client`` runs use a shared filesystem from multiple
3778 hosts, ``fio --client`` now prepends the IP address of the server to the
3779 filename. For example, if fio is using the directory :file:`/mnt/nfs/fio` and is
3780 writing filename :file:`fileio.tmp`, with a :option:`--client` `hostfile`
3781 containing two hostnames ``h1`` and ``h2`` with IP addresses 192.168.10.120 and
3782 192.168.10.121, then fio will create two files::
3784 /mnt/nfs/fio/192.168.10.120.fileio.tmp
3785 /mnt/nfs/fio/192.168.10.121.fileio.tmp