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 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:: --bandwidth-log
109 Generate aggregate bandwidth logs.
111 .. option:: --minimal
113 Print statistics in a terse, semicolon-delimited format.
115 .. option:: --append-terse
117 Print statistics in selected mode AND terse, semicolon-delimited format.
118 **deprecated**, use :option:`--output-format` instead to select multiple
121 .. option:: --output-format=type
123 Set the reporting format to `normal`, `terse`, `json`, or `json+`. Multiple
124 formats can be selected, separate by a comma. `terse` is a CSV based
125 format. `json+` is like `json`, except it adds a full dump of the latency
128 .. option:: --terse-version=type
130 Set terse version output format (default 3, or 2 or 4).
132 .. option:: --version
134 Print version info 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 builtin checksumming functions. If no argument is
147 given, all of them are tested. Or a comma separated list can be passed, in
148 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 :option:`ioengine`, or print help for `command`
157 defined by :option:`ioengine`. If no :option:`ioengine` is given, list all
160 .. option:: --showcmd=jobfile
162 Turn a job file into 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 When real-time ETA estimate should be printed. May be `always`, `never` or
178 .. option:: --eta-newline=time
180 Force a new line for every `time` period passed.
182 .. option:: --status-interval=time
184 Force full status dump every `time` period passed.
186 .. option:: --section=name
188 Only run specified section in job file. Multiple sections can be specified.
189 The ``--section`` option allows one to combine related jobs into one file.
190 E.g. one job file could define light, moderate, and heavy sections. Tell
191 fio to run only the "heavy" section by giving ``--section=heavy``
192 command line option. One can also specify the "write" operations in one
193 section and "verify" operation in another section. The ``--section`` option
194 only applies to job sections. The reserved *global* section is always
197 .. option:: --alloc-size=kb
199 Set the internal smalloc pool to this size in kb (def 1024). The
200 ``--alloc-size`` switch allows one to use a larger pool size for smalloc.
201 If running large jobs with randommap enabled, fio can run out of memory.
202 Smalloc is an internal allocator for shared structures from a fixed size
203 memory pool. The pool size defaults to 16M and can grow to 8 pools.
205 NOTE: While running :file:`.fio_smalloc.*` backing store files are visible
208 .. option:: --warnings-fatal
210 All fio parser warnings are fatal, causing fio to exit with an
213 .. option:: --max-jobs=nr
215 Maximum number of threads/processes to support.
217 .. option:: --server=args
219 Start a backend server, with `args` specifying what to listen to.
220 See `Client/Server`_ section.
222 .. option:: --daemonize=pidfile
224 Background a fio server, writing the pid to the given `pidfile` file.
226 .. option:: --client=hostname
228 Instead of running the jobs locally, send and run them on the given host or
229 set of hosts. See `Client/Server`_ section.
231 .. option:: --remote-config=file
233 Tell fio server to load this local file.
235 .. option:: --idle-prof=option
237 Report cpu idleness on a system or percpu basis
238 ``--idle-prof=system,percpu`` or
239 run unit work calibration only ``--idle-prof=calibrate``.
241 .. option:: --inflate-log=log
243 Inflate and output compressed log.
245 .. option:: --trigger-file=file
247 Execute trigger cmd when file exists.
249 .. option:: --trigger-timeout=t
251 Execute trigger at this time.
253 .. option:: --trigger=cmd
255 Set this command as local trigger.
257 .. option:: --trigger-remote=cmd
259 Set this command as remote trigger.
261 .. option:: --aux-path=path
263 Use this path for fio state generated files.
265 Any parameters following the options will be assumed to be job files, unless
266 they match a job file parameter. Multiple job files can be listed and each job
267 file will be regarded as a separate group. Fio will :option:`stonewall`
268 execution between each group.
274 As previously described, fio accepts one or more job files describing what it is
275 supposed to do. The job file format is the classic ini file, where the names
276 enclosed in [] brackets define the job name. You are free to use any ASCII name
277 you want, except *global* which has special meaning. Following the job name is
278 a sequence of zero or more parameters, one per line, that define the behavior of
279 the job. If the first character in a line is a ';' or a '#', the entire line is
280 discarded as a comment.
282 A *global* section sets defaults for the jobs described in that file. A job may
283 override a *global* section parameter, and a job file may even have several
284 *global* sections if so desired. A job is only affected by a *global* section
287 The :option:`--cmdhelp` option also lists all options. If used with an `option`
288 argument, :option:`--cmdhelp` will detail the given `option`.
290 See the `examples/` directory for inspiration on how to write job files. Note
291 the copyright and license requirements currently apply to `examples/` files.
293 So let's look at a really simple job file that defines two processes, each
294 randomly reading from a 128MiB file:
298 ; -- start job file --
309 As you can see, the job file sections themselves are empty as all the described
310 parameters are shared. As no :option:`filename` option is given, fio makes up a
311 `filename` for each of the jobs as it sees fit. On the command line, this job
312 would look as follows::
314 $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
317 Let's look at an example that has a number of processes writing randomly to
322 ; -- start job file --
333 Here we have no *global* section, as we only have one job defined anyway. We
334 want to use async I/O here, with a depth of 4 for each file. We also increased
335 the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical
336 jobs. The result is 4 processes each randomly writing to their own 64MiB
337 file. Instead of using the above job file, you could have given the parameters
338 on the command line. For this case, you would specify::
340 $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
342 When fio is utilized as a basis of any reasonably large test suite, it might be
343 desirable to share a set of standardized settings across multiple job files.
344 Instead of copy/pasting such settings, any section may pull in an external
345 :file:`filename.fio` file with *include filename* directive, as in the following
348 ; -- start job file including.fio --
352 include glob-include.fio
359 include test-include.fio
360 ; -- end job file including.fio --
364 ; -- start job file glob-include.fio --
367 ; -- end job file glob-include.fio --
371 ; -- start job file test-include.fio --
374 ; -- end job file test-include.fio --
376 Settings pulled into a section apply to that section only (except *global*
377 section). Include directives may be nested in that any included file may contain
378 further include directive(s). Include files may not contain [] sections.
381 Environment variables
382 ~~~~~~~~~~~~~~~~~~~~~
384 Fio also supports environment variable expansion in job files. Any sub-string of
385 the form ``${VARNAME}`` as part of an option value (in other words, on the right
386 of the '='), will be expanded to the value of the environment variable called
387 `VARNAME`. If no such environment variable is defined, or `VARNAME` is the
388 empty string, the empty string will be substituted.
390 As an example, let's look at a sample fio invocation and job file::
392 $ SIZE=64m NUMJOBS=4 fio jobfile.fio
396 ; -- start job file --
403 This will expand to the following equivalent job file at runtime:
407 ; -- start job file --
414 Fio ships with a few example job files, you can also look there for inspiration.
419 Additionally, fio has a set of reserved keywords that will be replaced
420 internally with the appropriate value. Those keywords are:
424 The architecture page size of the running system.
428 Megabytes of total memory in the system.
432 Number of online available CPUs.
434 These can be used on the command line or in the job file, and will be
435 automatically substituted with the current system values when the job is
436 run. Simple math is also supported on these keywords, so you can perform actions
441 and get that properly expanded to 8 times the size of memory in the machine.
447 This section describes in details each parameter associated with a job. Some
448 parameters take an option of a given type, such as an integer or a
449 string. Anywhere a numeric value is required, an arithmetic expression may be
450 used, provided it is surrounded by parentheses. Supported operators are:
459 For time values in expressions, units are microseconds by default. This is
460 different than for time values not in expressions (not enclosed in
461 parentheses). The following types are used:
468 String. This is a sequence of alpha characters.
471 Integer with possible time suffix. In seconds unless otherwise
472 specified, use e.g. 10m for 10 minutes. Accepts s/m/h for seconds, minutes,
473 and hours, and accepts 'ms' (or 'msec') for milliseconds, and 'us' (or
474 'usec') for microseconds.
479 Integer. A whole number value, which may contain an integer prefix
480 and an integer suffix:
482 [*integer prefix*] **number** [*integer suffix*]
484 The optional *integer prefix* specifies the number's base. The default
485 is decimal. *0x* specifies hexadecimal.
487 The optional *integer suffix* specifies the number's units, and includes an
488 optional unit prefix and an optional unit. For quantities of data, the
489 default unit is bytes. For quantities of time, the default unit is seconds.
491 With :option:`kb_base` =1000, fio follows international standards for unit
492 prefixes. To specify power-of-10 decimal values defined in the
493 International System of Units (SI):
495 * *Ki* -- means kilo (K) or 1000
496 * *Mi* -- means mega (M) or 1000**2
497 * *Gi* -- means giga (G) or 1000**3
498 * *Ti* -- means tera (T) or 1000**4
499 * *Pi* -- means peta (P) or 1000**5
501 To specify power-of-2 binary values defined in IEC 80000-13:
503 * *k* -- means kibi (Ki) or 1024
504 * *M* -- means mebi (Mi) or 1024**2
505 * *G* -- means gibi (Gi) or 1024**3
506 * *T* -- means tebi (Ti) or 1024**4
507 * *P* -- means pebi (Pi) or 1024**5
509 With :option:`kb_base` =1024 (the default), the unit prefixes are opposite
510 from those specified in the SI and IEC 80000-13 standards to provide
511 compatibility with old scripts. For example, 4k means 4096.
513 For quantities of data, an optional unit of 'B' may be included
514 (e.g., 'kB' is the same as 'k').
516 The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega,
517 not milli). 'b' and 'B' both mean byte, not bit.
519 Examples with :option:`kb_base` =1000:
521 * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB
522 * *1 MiB*: 1048576, 1mi, 1024ki
523 * *1 MB*: 1000000, 1m, 1000k
524 * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi
525 * *1 TB*: 1000000000, 1t, 1000m, 1000000k
527 Examples with :option:`kb_base` =1024 (default):
529 * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
530 * *1 MiB*: 1048576, 1m, 1024k
531 * *1 MB*: 1000000, 1mi, 1000ki
532 * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m
533 * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki
535 To specify times (units are not case sensitive):
539 * *M* -- mean minutes
540 * *s* -- or sec means seconds (default)
541 * *ms* -- or *msec* means milliseconds
542 * *us* -- or *usec* means microseconds
544 If the option accepts an upper and lower range, use a colon ':' or
545 minus '-' to separate such values. See :ref:`irange <irange>`.
550 Boolean. Usually parsed as an integer, however only defined for
551 true and false (1 and 0).
556 Integer range with suffix. Allows value range to be given, such as
557 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the
558 option allows two sets of ranges, they can be specified with a ',' or '/'
559 delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`.
562 A list of floating point numbers, separated by a ':' character.
568 .. option:: kb_base=int
570 Select the interpretation of unit prefixes in input parameters.
573 Inputs comply with IEC 80000-13 and the International
574 System of Units (SI). Use:
576 - power-of-2 values with IEC prefixes (e.g., KiB)
577 - power-of-10 values with SI prefixes (e.g., kB)
580 Compatibility mode (default). To avoid breaking old scripts:
582 - power-of-2 values with SI prefixes
583 - power-of-10 values with IEC prefixes
585 See :option:`bs` for more details on input parameters.
587 Outputs always use correct prefixes. Most outputs include both
590 bw=2383.3kB/s (2327.4KiB/s)
592 If only one value is reported, then kb_base selects the one to use:
594 **1000** -- SI prefixes
596 **1024** -- IEC prefixes
598 .. option:: unit_base=int
600 Base unit for reporting. Allowed values are:
603 Use auto-detection (default).
610 With the above in mind, here follows the complete list of fio job parameters.
618 ASCII name of the job. This may be used to override the name printed by fio
619 for this job. Otherwise the job name is used. On the command line this
620 parameter has the special purpose of also signaling the start of a new job.
622 .. option:: description=str
624 Text description of the job. Doesn't do anything except dump this text
625 description when this job is run. It's not parsed.
627 .. option:: loops=int
629 Run the specified number of iterations of this job. Used to repeat the same
630 workload a given number of times. Defaults to 1.
632 .. option:: numjobs=int
634 Create the specified number of clones of this job. Each clone of job
635 is spawned as an independent thread or process. May be used to setup a
636 larger number of threads/processes doing the same thing. Each thread is
637 reported separately; to see statistics for all clones as a whole, use
638 :option:`group_reporting` in conjunction with :option:`new_group`.
639 See :option:`--max-jobs`.
642 Time related parameters
643 ~~~~~~~~~~~~~~~~~~~~~~~
645 .. option:: runtime=time
647 Tell fio to terminate processing after the specified period of time. It
648 can be quite hard to determine for how long a specified job will run, so
649 this parameter is handy to cap the total runtime to a given time. When
650 the unit is omitted, the value is given in seconds.
652 .. option:: time_based
654 If set, fio will run for the duration of the :option:`runtime` specified
655 even if the file(s) are completely read or written. It will simply loop over
656 the same workload as many times as the :option:`runtime` allows.
658 .. option:: startdelay=irange(time)
660 Delay start of job for the specified number of seconds. Supports all time
661 suffixes to allow specification of hours, minutes, seconds and milliseconds
662 -- seconds are the default if a unit is omitted. Can be given as a range
663 which causes each thread to choose randomly out of the range.
665 .. option:: ramp_time=time
667 If set, fio will run the specified workload for this amount of time before
668 logging any performance numbers. Useful for letting performance settle
669 before logging results, thus minimizing the runtime required for stable
670 results. Note that the ``ramp_time`` is considered lead in time for a job,
671 thus it will increase the total runtime if a special timeout or
672 :option:`runtime` is specified. When the unit is omitted, the value is
675 .. option:: clocksource=str
677 Use the given clocksource as the base of timing. The supported options are:
680 :manpage:`gettimeofday(2)`
683 :manpage:`clock_gettime(2)`
686 Internal CPU clock source
688 cpu is the preferred clocksource if it is reliable, as it is very fast (and
689 fio is heavy on time calls). Fio will automatically use this clocksource if
690 it's supported and considered reliable on the system it is running on,
691 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
692 means supporting TSC Invariant.
694 .. option:: gtod_reduce=bool
696 Enable all of the :manpage:`gettimeofday(2)` reducing options
697 (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus
698 reduce precision of the timeout somewhat to really shrink the
699 :manpage:`gettimeofday(2)` call count. With this option enabled, we only do
700 about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all
701 time keeping was enabled.
703 .. option:: gtod_cpu=int
705 Sometimes it's cheaper to dedicate a single thread of execution to just
706 getting the current time. Fio (and databases, for instance) are very
707 intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set
708 one CPU aside for doing nothing but logging current time to a shared memory
709 location. Then the other threads/processes that run I/O workloads need only
710 copy that segment, instead of entering the kernel with a
711 :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time
712 calls will be excluded from other uses. Fio will manually clear it from the
713 CPU mask of other jobs.
719 .. option:: directory=str
721 Prefix filenames with this directory. Used to place files in a different
722 location than :file:`./`. You can specify a number of directories by
723 separating the names with a ':' character. These directories will be
724 assigned equally distributed to job clones creates with :option:`numjobs` as
725 long as they are using generated filenames. If specific `filename(s)` are
726 set fio will use the first listed directory, and thereby matching the
727 `filename` semantic which generates a file each clone if not specified, but
728 let all clones use the same if set.
730 See the :option:`filename` option for escaping certain characters.
732 .. option:: filename=str
734 Fio normally makes up a `filename` based on the job name, thread number, and
735 file number. If you want to share files between threads in a job or several
736 jobs with fixed file paths, specify a `filename` for each of them to override
737 the default. If the ioengine is file based, you can specify a number of files
738 by separating the names with a ':' colon. So if you wanted a job to open
739 :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use
740 ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is
741 specified, :option:`nrfiles` is ignored. The size of regular files specified
742 by this option will be :option:`size` divided by number of files unless
743 explicit size is specified by :option:`filesize`.
745 On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for
746 the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc.
747 Note: Windows and FreeBSD prevent write access to areas
748 of the disk containing in-use data (e.g. filesystems). If the wanted
749 `filename` does need to include a colon, then escape that with a ``\``
750 character. For instance, if the `filename` is :file:`/dev/dsk/foo@3,0:c`,
751 then you would use ``filename="/dev/dsk/foo@3,0\:c"``. The
752 :file:`-` is a reserved name, meaning stdin or stdout. Which of the two
753 depends on the read/write direction set.
755 .. option:: filename_format=str
757 If sharing multiple files between jobs, it is usually necessary to have fio
758 generate the exact names that you want. By default, fio will name a file
759 based on the default file format specification of
760 :file:`jobname.jobnumber.filenumber`. With this option, that can be
761 customized. Fio will recognize and replace the following keywords in this
765 The name of the worker thread or process.
767 The incremental number of the worker thread or process.
769 The incremental number of the file for that worker thread or
772 To have dependent jobs share a set of files, this option can be set to have
773 fio generate filenames that are shared between the two. For instance, if
774 :file:`testfiles.$filenum` is specified, file number 4 for any job will be
775 named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum`
776 will be used if no other format specifier is given.
778 .. option:: unique_filename=bool
780 To avoid collisions between networked clients, fio defaults to prefixing any
781 generated filenames (with a directory specified) with the source of the
782 client connecting. To disable this behavior, set this option to 0.
784 .. option:: opendir=str
786 Recursively open any files below directory `str`.
788 .. option:: lockfile=str
790 Fio defaults to not locking any files before it does I/O to them. If a file
791 or file descriptor is shared, fio can serialize I/O to that file to make the
792 end result consistent. This is usual for emulating real workloads that share
793 files. The lock modes are:
796 No locking. The default.
798 Only one thread or process may do I/O at a time, excluding all
801 Read-write locking on the file. Many readers may
802 access the file at the same time, but writes get exclusive access.
804 .. option:: nrfiles=int
806 Number of files to use for this job. Defaults to 1. The size of files
807 will be :option:`size` divided by this unless explicit size is specified by
808 :option:`filesize`. Files are created for each thread separately, and each
809 file will have a file number within its name by default, as explained in
810 :option:`filename` section.
813 .. option:: openfiles=int
815 Number of files to keep open at the same time. Defaults to the same as
816 :option:`nrfiles`, can be set smaller to limit the number simultaneous
819 .. option:: file_service_type=str
821 Defines how fio decides which file from a job to service next. The following
825 Choose a file at random.
828 Round robin over opened files. This is the default.
831 Finish one file before moving on to the next. Multiple files can
832 still be open depending on 'openfiles'.
835 Use a *Zipf* distribution to decide what file to access.
838 Use a *Pareto* distribution to decide what file to access.
841 Use a *Gaussian* (normal) distribution to decide what file to
844 For *random*, *roundrobin*, and *sequential*, a postfix can be appended to
845 tell fio how many I/Os to issue before switching to a new file. For example,
846 specifying ``file_service_type=random:8`` would cause fio to issue
847 8 I/Os before selecting a new file at random. For the non-uniform
848 distributions, a floating point postfix can be given to influence how the
849 distribution is skewed. See :option:`random_distribution` for a description
850 of how that would work.
852 .. option:: ioscheduler=str
854 Attempt to switch the device hosting the file to the specified I/O scheduler
857 .. option:: create_serialize=bool
859 If true, serialize the file creation for the jobs. This may be handy to
860 avoid interleaving of data files, which may greatly depend on the filesystem
861 used and even the number of processors in the system.
863 .. option:: create_fsync=bool
865 fsync the data file after creation. This is the default.
867 .. option:: create_on_open=bool
869 Don't pre-setup the files for I/O, just create open() when it's time to do
872 .. option:: create_only=bool
874 If true, fio will only run the setup phase of the job. If files need to be
875 laid out or updated on disk, only that will be done. The actual job contents
878 .. option:: allow_file_create=bool
880 If true, fio is permitted to create files as part of its workload. This is
881 the default behavior. If this option is false, then fio will error out if
882 the files it needs to use don't already exist. Default: true.
884 .. option:: allow_mounted_write=bool
886 If this isn't set, fio will abort jobs that are destructive (e.g. that write)
887 to what appears to be a mounted device or partition. This should help catch
888 creating inadvertently destructive tests, not realizing that the test will
889 destroy data on the mounted file system. Note that some platforms don't allow
890 writing against a mounted device regardless of this option. Default: false.
892 .. option:: pre_read=bool
894 If this is given, files will be pre-read into memory before starting the
895 given I/O operation. This will also clear the :option:`invalidate` flag,
896 since it is pointless to pre-read and then drop the cache. This will only
897 work for I/O engines that are seek-able, since they allow you to read the
898 same data multiple times. Thus it will not work on e.g. network or splice I/O.
900 .. option:: unlink=bool
902 Unlink the job files when done. Not the default, as repeated runs of that
903 job would then waste time recreating the file set again and again.
905 .. option:: unlink_each_loop=bool
907 Unlink job files after each iteration or loop.
909 .. option:: zonesize=int
911 Divide a file into zones of the specified size. See :option:`zoneskip`.
913 .. option:: zonerange=int
915 Give size of an I/O zone. See :option:`zoneskip`.
917 .. option:: zoneskip=int
919 Skip the specified number of bytes when :option:`zonesize` data has been
920 read. The two zone options can be used to only do I/O on zones of a file.
926 .. option:: direct=bool
928 If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that
929 ZFS on Solaris doesn't support direct I/O. On Windows the synchronous
930 ioengines don't support direct I/O. Default: false.
932 .. option:: atomic=bool
934 If value is true, attempt to use atomic direct I/O. Atomic writes are
935 guaranteed to be stable once acknowledged by the operating system. Only
936 Linux supports O_ATOMIC right now.
938 .. option:: buffered=bool
940 If value is true, use buffered I/O. This is the opposite of the
941 :option:`direct` option. Defaults to true.
943 .. option:: readwrite=str, rw=str
945 Type of I/O pattern. Accepted values are:
952 Sequential trims (Linux block devices only).
958 Random trims (Linux block devices only).
960 Sequential mixed reads and writes.
962 Random mixed reads and writes.
964 Sequential trim+write sequences. Blocks will be trimmed first,
965 then the same blocks will be written to.
967 Fio defaults to read if the option is not specified. For the mixed I/O
968 types, the default is to split them 50/50. For certain types of I/O the
969 result may still be skewed a bit, since the speed may be different. It is
970 possible to specify a number of I/O's to do before getting a new offset,
971 this is done by appending a ``:<nr>`` to the end of the string given. For a
972 random read, it would look like ``rw=randread:8`` for passing in an offset
973 modifier with a value of 8. If the suffix is used with a sequential I/O
974 pattern, then the value specified will be added to the generated offset for
975 each I/O. For instance, using ``rw=write:4k`` will skip 4k for every
976 write. It turns sequential I/O into sequential I/O with holes. See the
977 :option:`rw_sequencer` option.
979 .. option:: rw_sequencer=str
981 If an offset modifier is given by appending a number to the ``rw=<str>``
982 line, then this option controls how that number modifies the I/O offset
983 being generated. Accepted values are:
986 Generate sequential offset.
988 Generate the same offset.
990 ``sequential`` is only useful for random I/O, where fio would normally
991 generate a new random offset for every I/O. If you append e.g. 8 to randread,
992 you would get a new random offset for every 8 I/O's. The result would be a
993 seek for only every 8 I/O's, instead of for every I/O. Use ``rw=randread:8``
994 to specify that. As sequential I/O is already sequential, setting
995 ``sequential`` for that would not result in any differences. ``identical``
996 behaves in a similar fashion, except it sends the same offset 8 number of
997 times before generating a new offset.
999 .. option:: unified_rw_reporting=bool
1001 Fio normally reports statistics on a per data direction basis, meaning that
1002 reads, writes, and trims are accounted and reported separately. If this
1003 option is set fio sums the results and report them as "mixed" instead.
1005 .. option:: randrepeat=bool
1007 Seed the random number generator used for random I/O patterns in a
1008 predictable way so the pattern is repeatable across runs. Default: true.
1010 .. option:: allrandrepeat=bool
1012 Seed all random number generators in a predictable way so results are
1013 repeatable across runs. Default: false.
1015 .. option:: randseed=int
1017 Seed the random number generators based on this seed value, to be able to
1018 control what sequence of output is being generated. If not set, the random
1019 sequence depends on the :option:`randrepeat` setting.
1021 .. option:: fallocate=str
1023 Whether pre-allocation is performed when laying down files.
1024 Accepted values are:
1027 Do not pre-allocate space.
1030 Pre-allocate via :manpage:`posix_fallocate(3)`.
1033 Pre-allocate via :manpage:`fallocate(2)` with
1034 FALLOC_FL_KEEP_SIZE set.
1037 Backward-compatible alias for **none**.
1040 Backward-compatible alias for **posix**.
1042 May not be available on all supported platforms. **keep** is only available
1043 on Linux. If using ZFS on Solaris this must be set to **none** because ZFS
1044 doesn't support it. Default: **posix**.
1046 .. option:: fadvise_hint=str
1048 Use :manpage:`posix_fadvise(2)` to advise the kernel on what I/O patterns
1049 are likely to be issued. Accepted values are:
1052 Backwards-compatible hint for "no hint".
1055 Backwards compatible hint for "advise with fio workload type". This
1056 uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL**
1057 for a sequential workload.
1060 Advise using **FADV_SEQUENTIAL**.
1063 Advise using **FADV_RANDOM**.
1065 .. option:: fadvise_stream=int
1067 Use :manpage:`posix_fadvise(2)` to advise the kernel what stream ID the
1068 writes issued belong to. Only supported on Linux. Note, this option may
1069 change going forward.
1071 .. option:: offset=int
1073 Start I/O at the given offset in the file. The data before the given offset
1074 will not be touched. This effectively caps the file size at `real_size -
1075 offset`. Can be combined with :option:`size` to constrain the start and
1076 end range that I/O will be done within.
1078 .. option:: offset_increment=int
1080 If this is provided, then the real offset becomes `offset + offset_increment
1081 * thread_number`, where the thread number is a counter that starts at 0 and
1082 is incremented for each sub-job (i.e. when :option:`numjobs` option is
1083 specified). This option is useful if there are several jobs which are
1084 intended to operate on a file in parallel disjoint segments, with even
1085 spacing between the starting points.
1087 .. option:: number_ios=int
1089 Fio will normally perform I/Os until it has exhausted the size of the region
1090 set by :option:`size`, or if it exhaust the allocated time (or hits an error
1091 condition). With this setting, the range/size can be set independently of
1092 the number of I/Os to perform. When fio reaches this number, it will exit
1093 normally and report status. Note that this does not extend the amount of I/O
1094 that will be done, it will only stop fio if this condition is met before
1095 other end-of-job criteria.
1097 .. option:: fsync=int
1099 If writing to a file, issue a sync of the dirty data for every number of
1100 blocks given. For example, if you give 32 as a parameter, fio will sync the
1101 file for every 32 writes issued. If fio is using non-buffered I/O, we may
1102 not sync the file. The exception is the sg I/O engine, which synchronizes
1103 the disk cache anyway.
1105 .. option:: fdatasync=int
1107 Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and
1108 not metadata blocks. In Windows, FreeBSD, and DragonFlyBSD there is no
1109 :manpage:`fdatasync(2)`, this falls back to using :manpage:`fsync(2)`.
1111 .. option:: write_barrier=int
1113 Make every `N-th` write a barrier write.
1115 .. option:: sync_file_range=str:val
1117 Use :manpage:`sync_file_range(2)` for every `val` number of write
1118 operations. Fio will track range of writes that have happened since the last
1119 :manpage:`sync_file_range(2)` call. `str` can currently be one or more of:
1122 SYNC_FILE_RANGE_WAIT_BEFORE
1124 SYNC_FILE_RANGE_WRITE
1126 SYNC_FILE_RANGE_WAIT_AFTER
1128 So if you do ``sync_file_range=wait_before,write:8``, fio would use
1129 ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8
1130 writes. Also see the :manpage:`sync_file_range(2)` man page. This option is
1133 .. option:: overwrite=bool
1135 If true, writes to a file will always overwrite existing data. If the file
1136 doesn't already exist, it will be created before the write phase begins. If
1137 the file exists and is large enough for the specified write phase, nothing
1140 .. option:: end_fsync=bool
1142 If true, fsync file contents when a write stage has completed.
1144 .. option:: fsync_on_close=bool
1146 If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs
1147 from end_fsync in that it will happen on every file close, not just at the
1150 .. option:: rwmixread=int
1152 Percentage of a mixed workload that should be reads. Default: 50.
1154 .. option:: rwmixwrite=int
1156 Percentage of a mixed workload that should be writes. If both
1157 :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not
1158 add up to 100%, the latter of the two will be used to override the
1159 first. This may interfere with a given rate setting, if fio is asked to
1160 limit reads or writes to a certain rate. If that is the case, then the
1161 distribution may be skewed. Default: 50.
1163 .. option:: random_distribution=str:float[,str:float][,str:float]
1165 By default, fio will use a completely uniform random distribution when asked
1166 to perform random I/O. Sometimes it is useful to skew the distribution in
1167 specific ways, ensuring that some parts of the data is more hot than others.
1168 fio includes the following distribution models:
1171 Uniform random distribution
1180 Normal (Gaussian) distribution
1183 Zoned random distribution
1185 When using a **zipf** or **pareto** distribution, an input value is also
1186 needed to define the access pattern. For **zipf**, this is the `zipf
1187 theta`. For **pareto**, it's the `Pareto power`. Fio includes a test
1188 program, :command:`genzipf`, that can be used visualize what the given input
1189 values will yield in terms of hit rates. If you wanted to use **zipf** with
1190 a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the
1191 option. If a non-uniform model is used, fio will disable use of the random
1192 map. For the **gauss** distribution, a normal deviation is supplied as a
1193 value between 0 and 100.
1195 For a **zoned** distribution, fio supports specifying percentages of I/O
1196 access that should fall within what range of the file or device. For
1197 example, given a criteria of:
1199 * 60% of accesses should be to the first 10%
1200 * 30% of accesses should be to the next 20%
1201 * 8% of accesses should be to to the next 30%
1202 * 2% of accesses should be to the next 40%
1204 we can define that through zoning of the random accesses. For the above
1205 example, the user would do::
1207 random_distribution=zoned:60/10:30/20:8/30:2/40
1209 similarly to how :option:`bssplit` works for setting ranges and percentages
1210 of block sizes. Like :option:`bssplit`, it's possible to specify separate
1211 zones for reads, writes, and trims. If just one set is given, it'll apply to
1214 .. option:: percentage_random=int[,int][,int]
1216 For a random workload, set how big a percentage should be random. This
1217 defaults to 100%, in which case the workload is fully random. It can be set
1218 from anywhere from 0 to 100. Setting it to 0 would make the workload fully
1219 sequential. Any setting in between will result in a random mix of sequential
1220 and random I/O, at the given percentages. Comma-separated values may be
1221 specified for reads, writes, and trims as described in :option:`blocksize`.
1223 .. option:: norandommap
1225 Normally fio will cover every block of the file when doing random I/O. If
1226 this option is given, fio will just get a new random offset without looking
1227 at past I/O history. This means that some blocks may not be read or written,
1228 and that some blocks may be read/written more than once. If this option is
1229 used with :option:`verify` and multiple blocksizes (via :option:`bsrange`),
1230 only intact blocks are verified, i.e., partially-overwritten blocks are
1233 .. option:: softrandommap=bool
1235 See :option:`norandommap`. If fio runs with the random block map enabled and
1236 it fails to allocate the map, if this option is set it will continue without
1237 a random block map. As coverage will not be as complete as with random maps,
1238 this option is disabled by default.
1240 .. option:: random_generator=str
1242 Fio supports the following engines for generating
1243 I/O offsets for random I/O:
1246 Strong 2^88 cycle random number generator
1248 Linear feedback shift register generator
1250 Strong 64-bit 2^258 cycle random number generator
1252 **tausworthe** is a strong random number generator, but it requires tracking
1253 on the side if we want to ensure that blocks are only read or written
1254 once. **LFSR** guarantees that we never generate the same offset twice, and
1255 it's also less computationally expensive. It's not a true random generator,
1256 however, though for I/O purposes it's typically good enough. **LFSR** only
1257 works with single block sizes, not with workloads that use multiple block
1258 sizes. If used with such a workload, fio may read or write some blocks
1259 multiple times. The default value is **tausworthe**, unless the required
1260 space exceeds 2^32 blocks. If it does, then **tausworthe64** is
1261 selected automatically.
1267 .. option:: blocksize=int[,int][,int], bs=int[,int][,int]
1269 The block size in bytes used for I/O units. Default: 4096. A single value
1270 applies to reads, writes, and trims. Comma-separated values may be
1271 specified for reads, writes, and trims. A value not terminated in a comma
1272 applies to subsequent types.
1277 means 256k for reads, writes and trims.
1280 means 8k for reads, 32k for writes and trims.
1283 means 8k for reads, 32k for writes, and default for trims.
1286 means default for reads, 8k for writes and trims.
1289 means default for reads, 8k for writes, and default for writes.
1291 .. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange]
1293 A range of block sizes in bytes for I/O units. The issued I/O unit will
1294 always be a multiple of the minimum size, unless
1295 :option:`blocksize_unaligned` is set.
1297 Comma-separated ranges may be specified for reads, writes, and trims as
1298 described in :option:`blocksize`.
1300 Example: ``bsrange=1k-4k,2k-8k``.
1302 .. option:: bssplit=str[,str][,str]
1304 Sometimes you want even finer grained control of the block sizes issued, not
1305 just an even split between them. This option allows you to weight various
1306 block sizes, so that you are able to define a specific amount of block sizes
1307 issued. The format for this option is::
1309 bssplit=blocksize/percentage:blocksize/percentage
1311 for as many block sizes as needed. So if you want to define a workload that
1312 has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would write::
1314 bssplit=4k/10:64k/50:32k/40
1316 Ordering does not matter. If the percentage is left blank, fio will fill in
1317 the remaining values evenly. So a bssplit option like this one::
1319 bssplit=4k/50:1k/:32k/
1321 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always add up
1322 to 100, if bssplit is given a range that adds up to more, it will error out.
1324 Comma-separated values may be specified for reads, writes, and trims as
1325 described in :option:`blocksize`.
1327 If you want a workload that has 50% 2k reads and 50% 4k reads, while having
1328 90% 4k writes and 10% 8k writes, you would specify::
1330 bssplit=2k/50:4k/50,4k/90,8k/10
1332 .. option:: blocksize_unaligned, bs_unaligned
1334 If set, fio will issue I/O units with any size within
1335 :option:`blocksize_range`, not just multiples of the minimum size. This
1336 typically won't work with direct I/O, as that normally requires sector
1339 .. option:: bs_is_seq_rand
1341 If this option is set, fio will use the normal read,write blocksize settings
1342 as sequential,random blocksize settings instead. Any random read or write
1343 will use the WRITE blocksize settings, and any sequential read or write will
1344 use the READ blocksize settings.
1346 .. option:: blockalign=int[,int][,int], ba=int[,int][,int]
1348 Boundary to which fio will align random I/O units. Default:
1349 :option:`blocksize`. Minimum alignment is typically 512b for using direct
1350 I/O, though it usually depends on the hardware block size. This option is
1351 mutually exclusive with using a random map for files, so it will turn off
1352 that option. Comma-separated values may be specified for reads, writes, and
1353 trims as described in :option:`blocksize`.
1359 .. option:: zero_buffers
1361 Initialize buffers with all zeros. Default: fill buffers with random data.
1363 .. option:: refill_buffers
1365 If this option is given, fio will refill the I/O buffers on every
1366 submit. The default is to only fill it at init time and reuse that
1367 data. Only makes sense if zero_buffers isn't specified, naturally. If data
1368 verification is enabled, `refill_buffers` is also automatically enabled.
1370 .. option:: scramble_buffers=bool
1372 If :option:`refill_buffers` is too costly and the target is using data
1373 deduplication, then setting this option will slightly modify the I/O buffer
1374 contents to defeat normal de-dupe attempts. This is not enough to defeat
1375 more clever block compression attempts, but it will stop naive dedupe of
1376 blocks. Default: true.
1378 .. option:: buffer_compress_percentage=int
1380 If this is set, then fio will attempt to provide I/O buffer content (on
1381 WRITEs) that compress to the specified level. Fio does this by providing a
1382 mix of random data and a fixed pattern. The fixed pattern is either zeroes,
1383 or the pattern specified by :option:`buffer_pattern`. If the pattern option
1384 is used, it might skew the compression ratio slightly. Note that this is per
1385 block size unit, for file/disk wide compression level that matches this
1386 setting, you'll also want to set :option:`refill_buffers`.
1388 .. option:: buffer_compress_chunk=int
1390 See :option:`buffer_compress_percentage`. This setting allows fio to manage
1391 how big the ranges of random data and zeroed data is. Without this set, fio
1392 will provide :option:`buffer_compress_percentage` of blocksize random data,
1393 followed by the remaining zeroed. With this set to some chunk size smaller
1394 than the block size, fio can alternate random and zeroed data throughout the
1397 .. option:: buffer_pattern=str
1399 If set, fio will fill the I/O buffers with this pattern. If not set, the
1400 contents of I/O buffers is defined by the other options related to buffer
1401 contents. The setting can be any pattern of bytes, and can be prefixed with
1402 0x for hex values. It may also be a string, where the string must then be
1403 wrapped with ``""``, e.g.::
1405 buffer_pattern="abcd"
1413 buffer_pattern=0xdeadface
1415 Also you can combine everything together in any order::
1417 buffer_pattern=0xdeadface"abcd"-12
1419 .. option:: dedupe_percentage=int
1421 If set, fio will generate this percentage of identical buffers when
1422 writing. These buffers will be naturally dedupable. The contents of the
1423 buffers depend on what other buffer compression settings have been set. It's
1424 possible to have the individual buffers either fully compressible, or not at
1425 all. This option only controls the distribution of unique buffers.
1427 .. option:: invalidate=bool
1429 Invalidate the buffer/page cache parts for this file prior to starting
1430 I/O if the platform and file type support it. Defaults to true.
1431 This will be ignored if :option:`pre_read` is also specified for the
1434 .. option:: sync=bool
1436 Use synchronous I/O for buffered writes. For the majority of I/O engines,
1437 this means using O_SYNC. Default: false.
1439 .. option:: iomem=str, mem=str
1441 Fio can use various types of memory as the I/O unit buffer. The allowed
1445 Use memory from :manpage:`malloc(3)` as the buffers. Default memory
1449 Use shared memory as the buffers. Allocated through
1450 :manpage:`shmget(2)`.
1453 Same as shm, but use huge pages as backing.
1456 Use mmap to allocate buffers. May either be anonymous memory, or can
1457 be file backed if a filename is given after the option. The format
1458 is `mem=mmap:/path/to/file`.
1461 Use a memory mapped huge file as the buffer backing. Append filename
1462 after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`.
1465 Same as mmap, but use a MMAP_SHARED mapping.
1467 The area allocated is a function of the maximum allowed bs size for the job,
1468 multiplied by the I/O depth given. Note that for **shmhuge** and
1469 **mmaphuge** to work, the system must have free huge pages allocated. This
1470 can normally be checked and set by reading/writing
1471 :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page
1472 is 4MiB in size. So to calculate the number of huge pages you need for a
1473 given job file, add up the I/O depth of all jobs (normally one unless
1474 :option:`iodepth` is used) and multiply by the maximum bs set. Then divide
1475 that number by the huge page size. You can see the size of the huge pages in
1476 :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero
1477 number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also
1478 see :option:`hugepage-size`.
1480 **mmaphuge** also needs to have hugetlbfs mounted and the file location
1481 should point there. So if it's mounted in :file:`/huge`, you would use
1482 `mem=mmaphuge:/huge/somefile`.
1484 .. option:: iomem_align=int
1486 This indicates the memory alignment of the I/O memory buffers. Note that
1487 the given alignment is applied to the first I/O unit buffer, if using
1488 :option:`iodepth` the alignment of the following buffers are given by the
1489 :option:`bs` used. In other words, if using a :option:`bs` that is a
1490 multiple of the page sized in the system, all buffers will be aligned to
1491 this value. If using a :option:`bs` that is not page aligned, the alignment
1492 of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and
1495 .. option:: hugepage-size=int
1497 Defines the size of a huge page. Must at least be equal to the system
1498 setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably
1499 always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the
1500 preferred way to set this to avoid setting a non-pow-2 bad value.
1502 .. option:: lockmem=int
1504 Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to
1505 simulate a smaller amount of memory. The amount specified is per worker.
1511 .. option:: size=int
1513 The total size of file I/O for each thread of this job. Fio will run until
1514 this many bytes has been transferred, unless runtime is limited by other options
1515 (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`).
1516 Fio will divide this size between the available files determined by options
1517 such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is
1518 specified by the job. If the result of division happens to be 0, the size is
1519 set to the physical size of the given files or devices if they exist.
1520 If this option is not specified, fio will use the full size of the given
1521 files or devices. If the files do not exist, size must be given. It is also
1522 possible to give size as a percentage between 1 and 100. If ``size=20%`` is
1523 given, fio will use 20% of the full size of the given files or devices.
1524 Can be combined with :option:`offset` to constrain the start and end range
1525 that I/O will be done within.
1527 .. option:: io_size=int, io_limit=int
1529 Normally fio operates within the region set by :option:`size`, which means
1530 that the :option:`size` option sets both the region and size of I/O to be
1531 performed. Sometimes that is not what you want. With this option, it is
1532 possible to define just the amount of I/O that fio should do. For instance,
1533 if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio
1534 will perform I/O within the first 20GiB but exit when 5GiB have been
1535 done. The opposite is also possible -- if :option:`size` is set to 20GiB,
1536 and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within
1537 the 0..20GiB region.
1539 .. option:: filesize=int
1541 Individual file sizes. May be a range, in which case fio will select sizes
1542 for files at random within the given range and limited to :option:`size` in
1543 total (if that is given). If not given, each created file is the same size.
1544 This option overrides :option:`size` in terms of file size, which means
1545 this value is used as a fixed size or possible range of each file.
1547 .. option:: file_append=bool
1549 Perform I/O after the end of the file. Normally fio will operate within the
1550 size of a file. If this option is set, then fio will append to the file
1551 instead. This has identical behavior to setting :option:`offset` to the size
1552 of a file. This option is ignored on non-regular files.
1554 .. option:: fill_device=bool, fill_fs=bool
1556 Sets size to something really large and waits for ENOSPC (no space left on
1557 device) as the terminating condition. Only makes sense with sequential
1558 write. For a read workload, the mount point will be filled first then I/O
1559 started on the result. This option doesn't make sense if operating on a raw
1560 device node, since the size of that is already known by the file system.
1561 Additionally, writing beyond end-of-device will not return ENOSPC there.
1567 .. option:: ioengine=str
1569 Defines how the job issues I/O to the file. The following types are defined:
1572 Basic :manpage:`read(2)` or :manpage:`write(2)`
1573 I/O. :manpage:`lseek(2)` is used to position the I/O location.
1576 Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on
1577 all supported operating systems except for Windows.
1580 Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate
1581 queuing by coalescing adjacent I/Os into a single submission.
1584 Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O.
1587 Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O.
1590 Linux native asynchronous I/O. Note that Linux may only support
1591 queued behaviour with non-buffered I/O (set ``direct=1`` or
1593 This engine defines engine specific options.
1596 POSIX asynchronous I/O using :manpage:`aio_read(3)` and
1597 :manpage:`aio_write(3)`.
1600 Solaris native asynchronous I/O.
1603 Windows native asynchronous I/O. Default on Windows.
1606 File is memory mapped with :manpage:`mmap(2)` and data copied
1607 to/from using :manpage:`memcpy(3)`.
1610 :manpage:`splice(2)` is used to transfer the data and
1611 :manpage:`vmsplice(2)` to transfer data from user space to the
1615 SCSI generic sg v3 I/O. May either be synchronous using the SG_IO
1616 ioctl, or if the target is an sg character device we use
1617 :manpage:`read(2)` and :manpage:`write(2)` for asynchronous
1618 I/O. Requires filename option to specify either block or character
1622 Doesn't transfer any data, just pretends to. This is mainly used to
1623 exercise fio itself and for debugging/testing purposes.
1626 Transfer over the network to given ``host:port``. Depending on the
1627 :option:`protocol` used, the :option:`hostname`, :option:`port`,
1628 :option:`listen` and :option:`filename` options are used to specify
1629 what sort of connection to make, while the :option:`protocol` option
1630 determines which protocol will be used. This engine defines engine
1634 Like **net**, but uses :manpage:`splice(2)` and
1635 :manpage:`vmsplice(2)` to map data and send/receive.
1636 This engine defines engine specific options.
1639 Doesn't transfer any data, but burns CPU cycles according to the
1640 :option:`cpuload` and :option:`cpuchunks` options. Setting
1641 :option:`cpuload` =85 will cause that job to do nothing but burn 85%
1642 of the CPU. In case of SMP machines, use :option:`numjobs`
1643 =<no_of_cpu> to get desired CPU usage, as the cpuload only loads a
1644 single CPU at the desired rate. A job never finishes unless there is
1645 at least one non-cpuio job.
1648 The GUASI I/O engine is the Generic Userspace Asyncronous Syscall
1649 Interface approach to async I/O. See
1651 http://www.xmailserver.org/guasi-lib.html
1653 for more info on GUASI.
1656 The RDMA I/O engine supports both RDMA memory semantics
1657 (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the
1658 InfiniBand, RoCE and iWARP protocols.
1661 I/O engine that does regular fallocate to simulate data transfer as
1665 does fallocate(,mode = FALLOC_FL_KEEP_SIZE,).
1668 does fallocate(,mode = 0).
1671 does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE).
1674 I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate
1675 defragment activity in request to DDIR_WRITE event.
1678 I/O engine supporting direct access to Ceph Rados Block Devices
1679 (RBD) via librbd without the need to use the kernel rbd driver. This
1680 ioengine defines engine specific options.
1683 Using Glusterfs libgfapi sync interface to direct access to
1684 Glusterfs volumes without having to go through FUSE. This ioengine
1685 defines engine specific options.
1688 Using Glusterfs libgfapi async interface to direct access to
1689 Glusterfs volumes without having to go through FUSE. This ioengine
1690 defines engine specific options.
1693 Read and write through Hadoop (HDFS). The :file:`filename` option
1694 is used to specify host,port of the hdfs name-node to connect. This
1695 engine interprets offsets a little differently. In HDFS, files once
1696 created cannot be modified. So random writes are not possible. To
1697 imitate this, libhdfs engine expects bunch of small files to be
1698 created over HDFS, and engine will randomly pick a file out of those
1699 files based on the offset generated by fio backend. (see the example
1700 job file to create such files, use ``rw=write`` option). Please
1701 note, you might want to set necessary environment variables to work
1702 with hdfs/libhdfs properly. Each job uses its own connection to
1706 Read, write and erase an MTD character device (e.g.,
1707 :file:`/dev/mtd0`). Discards are treated as erases. Depending on the
1708 underlying device type, the I/O may have to go in a certain pattern,
1709 e.g., on NAND, writing sequentially to erase blocks and discarding
1710 before overwriting. The writetrim mode works well for this
1714 Read and write using filesystem DAX to a file on a filesystem
1715 mounted with DAX on a persistent memory device through the NVML
1719 Read and write using device DAX to a persistent memory device (e.g.,
1720 /dev/dax0.0) through the NVML libpmem library.
1723 Prefix to specify loading an external I/O engine object file. Append
1724 the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load
1725 ioengine :file:`foo.o` in :file:`/tmp`.
1728 I/O engine specific parameters
1729 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1731 In addition, there are some parameters which are only valid when a specific
1732 ioengine is in use. These are used identically to normal parameters, with the
1733 caveat that when used on the command line, they must come after the
1734 :option:`ioengine` that defines them is selected.
1736 .. option:: userspace_reap : [libaio]
1738 Normally, with the libaio engine in use, fio will use the
1739 :manpage:`io_getevents(2)` system call to reap newly returned events. With
1740 this flag turned on, the AIO ring will be read directly from user-space to
1741 reap events. The reaping mode is only enabled when polling for a minimum of
1742 0 events (e.g. when :option:`iodepth_batch_complete` `=0`).
1744 .. option:: hipri : [pvsync2]
1746 Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority
1749 .. option:: cpuload=int : [cpuio]
1751 Attempt to use the specified percentage of CPU cycles. This is a mandatory
1752 option when using cpuio I/O engine.
1754 .. option:: cpuchunks=int : [cpuio]
1756 Split the load into cycles of the given time. In microseconds.
1758 .. option:: exit_on_io_done=bool : [cpuio]
1760 Detect when I/O threads are done, then exit.
1762 .. option:: hostname=str : [netsplice] [net]
1764 The host name or IP address to use for TCP or UDP based I/O. If the job is
1765 a TCP listener or UDP reader, the host name is not used and must be omitted
1766 unless it is a valid UDP multicast address.
1768 .. option:: namenode=str : [libhdfs]
1770 The host name or IP address of a HDFS cluster namenode to contact.
1772 .. option:: port=int
1776 The TCP or UDP port to bind to or connect to. If this is used with
1777 :option:`numjobs` to spawn multiple instances of the same job type, then
1778 this will be the starting port number since fio will use a range of
1783 the listening port of the HFDS cluster namenode.
1785 .. option:: interface=str : [netsplice] [net]
1787 The IP address of the network interface used to send or receive UDP
1790 .. option:: ttl=int : [netsplice] [net]
1792 Time-to-live value for outgoing UDP multicast packets. Default: 1.
1794 .. option:: nodelay=bool : [netsplice] [net]
1796 Set TCP_NODELAY on TCP connections.
1798 .. option:: protocol=str : [netsplice] [net]
1800 .. option:: proto=str : [netsplice] [net]
1802 The network protocol to use. Accepted values are:
1805 Transmission control protocol.
1807 Transmission control protocol V6.
1809 User datagram protocol.
1811 User datagram protocol V6.
1815 When the protocol is TCP or UDP, the port must also be given, as well as the
1816 hostname if the job is a TCP listener or UDP reader. For unix sockets, the
1817 normal filename option should be used and the port is invalid.
1819 .. option:: listen : [net]
1821 For TCP network connections, tell fio to listen for incoming connections
1822 rather than initiating an outgoing connection. The :option:`hostname` must
1823 be omitted if this option is used.
1825 .. option:: pingpong : [net]
1827 Normally a network writer will just continue writing data, and a network
1828 reader will just consume packages. If ``pingpong=1`` is set, a writer will
1829 send its normal payload to the reader, then wait for the reader to send the
1830 same payload back. This allows fio to measure network latencies. The
1831 submission and completion latencies then measure local time spent sending or
1832 receiving, and the completion latency measures how long it took for the
1833 other end to receive and send back. For UDP multicast traffic
1834 ``pingpong=1`` should only be set for a single reader when multiple readers
1835 are listening to the same address.
1837 .. option:: window_size : [net]
1839 Set the desired socket buffer size for the connection.
1841 .. option:: mss : [net]
1843 Set the TCP maximum segment size (TCP_MAXSEG).
1845 .. option:: donorname=str : [e4defrag]
1847 File will be used as a block donor(swap extents between files).
1849 .. option:: inplace=int : [e4defrag]
1851 Configure donor file blocks allocation strategy:
1854 Default. Preallocate donor's file on init.
1856 Allocate space immediately inside defragment event, and free right
1859 .. option:: clustername=str : [rbd]
1861 Specifies the name of the Ceph cluster.
1863 .. option:: rbdname=str : [rbd]
1865 Specifies the name of the RBD.
1867 .. option:: pool=str : [rbd]
1869 Specifies the name of the Ceph pool containing RBD.
1871 .. option:: clientname=str : [rbd]
1873 Specifies the username (without the 'client.' prefix) used to access the
1874 Ceph cluster. If the *clustername* is specified, the *clientname* shall be
1875 the full *type.id* string. If no type. prefix is given, fio will add
1876 'client.' by default.
1878 .. option:: skip_bad=bool : [mtd]
1880 Skip operations against known bad blocks.
1882 .. option:: hdfsdirectory : [libhdfs]
1884 libhdfs will create chunk in this HDFS directory.
1886 .. option:: chunk_size : [libhdfs]
1888 the size of the chunk to use for each file.
1894 .. option:: iodepth=int
1896 Number of I/O units to keep in flight against the file. Note that
1897 increasing *iodepth* beyond 1 will not affect synchronous ioengines (except
1898 for small degrees when :option:`verify_async` is in use). Even async
1899 engines may impose OS restrictions causing the desired depth not to be
1900 achieved. This may happen on Linux when using libaio and not setting
1901 :option:`direct` =1, since buffered I/O is not async on that OS. Keep an
1902 eye on the I/O depth distribution in the fio output to verify that the
1903 achieved depth is as expected. Default: 1.
1905 .. option:: iodepth_batch_submit=int, iodepth_batch=int
1907 This defines how many pieces of I/O to submit at once. It defaults to 1
1908 which means that we submit each I/O as soon as it is available, but can be
1909 raised to submit bigger batches of I/O at the time. If it is set to 0 the
1910 :option:`iodepth` value will be used.
1912 .. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int
1914 This defines how many pieces of I/O to retrieve at once. It defaults to 1
1915 which means that we'll ask for a minimum of 1 I/O in the retrieval process
1916 from the kernel. The I/O retrieval will go on until we hit the limit set by
1917 :option:`iodepth_low`. If this variable is set to 0, then fio will always
1918 check for completed events before queuing more I/O. This helps reduce I/O
1919 latency, at the cost of more retrieval system calls.
1921 .. option:: iodepth_batch_complete_max=int
1923 This defines maximum pieces of I/O to retrieve at once. This variable should
1924 be used along with :option:`iodepth_batch_complete_min` =int variable,
1925 specifying the range of min and max amount of I/O which should be
1926 retrieved. By default it is equal to :option:`iodepth_batch_complete_min`
1931 iodepth_batch_complete_min=1
1932 iodepth_batch_complete_max=<iodepth>
1934 which means that we will retrieve at least 1 I/O and up to the whole
1935 submitted queue depth. If none of I/O has been completed yet, we will wait.
1939 iodepth_batch_complete_min=0
1940 iodepth_batch_complete_max=<iodepth>
1942 which means that we can retrieve up to the whole submitted queue depth, but
1943 if none of I/O has been completed yet, we will NOT wait and immediately exit
1944 the system call. In this example we simply do polling.
1946 .. option:: iodepth_low=int
1948 The low water mark indicating when to start filling the queue
1949 again. Defaults to the same as :option:`iodepth`, meaning that fio will
1950 attempt to keep the queue full at all times. If :option:`iodepth` is set to
1951 e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of
1952 16 requests, it will let the depth drain down to 4 before starting to fill
1955 .. option:: io_submit_mode=str
1957 This option controls how fio submits the I/O to the I/O engine. The default
1958 is `inline`, which means that the fio job threads submit and reap I/O
1959 directly. If set to `offload`, the job threads will offload I/O submission
1960 to a dedicated pool of I/O threads. This requires some coordination and thus
1961 has a bit of extra overhead, especially for lower queue depth I/O where it
1962 can increase latencies. The benefit is that fio can manage submission rates
1963 independently of the device completion rates. This avoids skewed latency
1964 reporting if I/O gets back up on the device side (the coordinated omission
1971 .. option:: thinktime=time
1973 Stall the job for the specified period of time after an I/O has completed before issuing the
1974 next. May be used to simulate processing being done by an application.
1975 When the unit is omitted, the value is given in microseconds. See
1976 :option:`thinktime_blocks` and :option:`thinktime_spin`.
1978 .. option:: thinktime_spin=time
1980 Only valid if :option:`thinktime` is set - pretend to spend CPU time doing
1981 something with the data received, before falling back to sleeping for the
1982 rest of the period specified by :option:`thinktime`. When the unit is
1983 omitted, the value is given in microseconds.
1985 .. option:: thinktime_blocks=int
1987 Only valid if :option:`thinktime` is set - control how many blocks to issue,
1988 before waiting `thinktime` usecs. If not set, defaults to 1 which will make
1989 fio wait `thinktime` usecs after every block. This effectively makes any
1990 queue depth setting redundant, since no more than 1 I/O will be queued
1991 before we have to complete it and do our thinktime. In other words, this
1992 setting effectively caps the queue depth if the latter is larger.
1994 .. option:: rate=int[,int][,int]
1996 Cap the bandwidth used by this job. The number is in bytes/sec, the normal
1997 suffix rules apply. Comma-separated values may be specified for reads,
1998 writes, and trims as described in :option:`blocksize`.
2000 .. option:: rate_min=int[,int][,int]
2002 Tell fio to do whatever it can to maintain at least this bandwidth. Failing
2003 to meet this requirement will cause the job to exit. Comma-separated values
2004 may be specified for reads, writes, and trims as described in
2005 :option:`blocksize`.
2007 .. option:: rate_iops=int[,int][,int]
2009 Cap the bandwidth to this number of IOPS. Basically the same as
2010 :option:`rate`, just specified independently of bandwidth. If the job is
2011 given a block size range instead of a fixed value, the smallest block size
2012 is used as the metric. Comma-separated values may be specified for reads,
2013 writes, and trims as described in :option:`blocksize`.
2015 .. option:: rate_iops_min=int[,int][,int]
2017 If fio doesn't meet this rate of I/O, it will cause the job to exit.
2018 Comma-separated values may be specified for reads, writes, and trims as
2019 described in :option:`blocksize`.
2021 .. option:: rate_process=str
2023 This option controls how fio manages rated I/O submissions. The default is
2024 `linear`, which submits I/O in a linear fashion with fixed delays between
2025 I/Os that gets adjusted based on I/O completion rates. If this is set to
2026 `poisson`, fio will submit I/O based on a more real world random request
2027 flow, known as the Poisson process
2028 (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be
2029 10^6 / IOPS for the given workload.
2035 .. option:: latency_target=time
2037 If set, fio will attempt to find the max performance point that the given
2038 workload will run at while maintaining a latency below this target. When
2039 the unit is omitted, the value is given in microseconds. See
2040 :option:`latency_window` and :option:`latency_percentile`.
2042 .. option:: latency_window=time
2044 Used with :option:`latency_target` to specify the sample window that the job
2045 is run at varying queue depths to test the performance. When the unit is
2046 omitted, the value is given in microseconds.
2048 .. option:: latency_percentile=float
2050 The percentage of I/Os that must fall within the criteria specified by
2051 :option:`latency_target` and :option:`latency_window`. If not set, this
2052 defaults to 100.0, meaning that all I/Os must be equal or below to the value
2053 set by :option:`latency_target`.
2055 .. option:: max_latency=time
2057 If set, fio will exit the job with an ETIMEDOUT error if it exceeds this
2058 maximum latency. When the unit is omitted, the value is given in
2061 .. option:: rate_cycle=int
2063 Average bandwidth for :option:`rate` and :option:`rate_min` over this number
2070 .. option:: write_iolog=str
2072 Write the issued I/O patterns to the specified file. See
2073 :option:`read_iolog`. Specify a separate file for each job, otherwise the
2074 iologs will be interspersed and the file may be corrupt.
2076 .. option:: read_iolog=str
2078 Open an iolog with the specified file name and replay the I/O patterns it
2079 contains. This can be used to store a workload and replay it sometime
2080 later. The iolog given may also be a blktrace binary file, which allows fio
2081 to replay a workload captured by :command:`blktrace`. See
2082 :manpage:`blktrace(8)` for how to capture such logging data. For blktrace
2083 replay, the file needs to be turned into a blkparse binary data file first
2084 (``blkparse <device> -o /dev/null -d file_for_fio.bin``).
2086 .. option:: replay_no_stall=int
2088 When replaying I/O with :option:`read_iolog` the default behavior is to
2089 attempt to respect the time stamps within the log and replay them with the
2090 appropriate delay between IOPS. By setting this variable fio will not
2091 respect the timestamps and attempt to replay them as fast as possible while
2092 still respecting ordering. The result is the same I/O pattern to a given
2093 device, but different timings.
2095 .. option:: replay_redirect=str
2097 While replaying I/O patterns using :option:`read_iolog` the default behavior
2098 is to replay the IOPS onto the major/minor device that each IOP was recorded
2099 from. This is sometimes undesirable because on a different machine those
2100 major/minor numbers can map to a different device. Changing hardware on the
2101 same system can also result in a different major/minor mapping.
2102 ``replay_redirect`` causes all IOPS to be replayed onto the single specified
2103 device regardless of the device it was recorded
2104 from. i.e. :option:`replay_redirect` = :file:`/dev/sdc` would cause all I/O
2105 in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means
2106 multiple devices will be replayed onto a single device, if the trace
2107 contains multiple devices. If you want multiple devices to be replayed
2108 concurrently to multiple redirected devices you must blkparse your trace
2109 into separate traces and replay them with independent fio invocations.
2110 Unfortunately this also breaks the strict time ordering between multiple
2113 .. option:: replay_align=int
2115 Force alignment of I/O offsets and lengths in a trace to this power of 2
2118 .. option:: replay_scale=int
2120 Scale sector offsets down by this factor when replaying traces.
2123 Threads, processes and job synchronization
2124 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2128 Fio defaults to forking jobs, however if this option is given, fio will use
2129 POSIX Threads function :manpage:`pthread_create(3)` to create threads instead
2130 of forking processes.
2132 .. option:: wait_for=str
2134 Specifies the name of the already defined job to wait for. Single waitee
2135 name only may be specified. If set, the job won't be started until all
2136 workers of the waitee job are done.
2138 ``wait_for`` operates on the job name basis, so there are a few
2139 limitations. First, the waitee must be defined prior to the waiter job
2140 (meaning no forward references). Second, if a job is being referenced as a
2141 waitee, it must have a unique name (no duplicate waitees).
2143 .. option:: nice=int
2145 Run the job with the given nice value. See man :manpage:`nice(2)`.
2147 On Windows, values less than -15 set the process class to "High"; -1 through
2148 -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle"
2151 .. option:: prio=int
2153 Set the I/O priority value of this job. Linux limits us to a positive value
2154 between 0 and 7, with 0 being the highest. See man
2155 :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating
2156 systems since meaning of priority may differ.
2158 .. option:: prioclass=int
2160 Set the I/O priority class. See man :manpage:`ionice(1)`.
2162 .. option:: cpumask=int
2164 Set the CPU affinity of this job. The parameter given is a bitmask of
2165 allowed CPU's the job may run on. So if you want the allowed CPUs to be 1
2166 and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man
2167 :manpage:`sched_setaffinity(2)`. This may not work on all supported
2168 operating systems or kernel versions. This option doesn't work well for a
2169 higher CPU count than what you can store in an integer mask, so it can only
2170 control cpus 1-32. For boxes with larger CPU counts, use
2171 :option:`cpus_allowed`.
2173 .. option:: cpus_allowed=str
2175 Controls the same options as :option:`cpumask`, but it allows a text setting
2176 of the permitted CPUs instead. So to use CPUs 1 and 5, you would specify
2177 ``cpus_allowed=1,5``. This options also allows a range of CPUs. Say you
2178 wanted a binding to CPUs 1, 5, and 8-15, you would set
2179 ``cpus_allowed=1,5,8-15``.
2181 .. option:: cpus_allowed_policy=str
2183 Set the policy of how fio distributes the CPUs specified by
2184 :option:`cpus_allowed` or cpumask. Two policies are supported:
2187 All jobs will share the CPU set specified.
2189 Each job will get a unique CPU from the CPU set.
2191 **shared** is the default behaviour, if the option isn't specified. If
2192 **split** is specified, then fio will will assign one cpu per job. If not
2193 enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs
2196 .. option:: numa_cpu_nodes=str
2198 Set this job running on specified NUMA nodes' CPUs. The arguments allow
2199 comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable
2200 numa options support, fio must be built on a system with libnuma-dev(el)
2203 .. option:: numa_mem_policy=str
2205 Set this job's memory policy and corresponding NUMA nodes. Format of the
2210 ``mode`` is one of the following memory policy: ``default``, ``prefer``,
2211 ``bind``, ``interleave``, ``local`` For ``default`` and ``local`` memory
2212 policy, no node is needed to be specified. For ``prefer``, only one node is
2213 allowed. For ``bind`` and ``interleave``, it allow comma delimited list of
2214 numbers, A-B ranges, or `all`.
2216 .. option:: cgroup=str
2218 Add job to this control group. If it doesn't exist, it will be created. The
2219 system must have a mounted cgroup blkio mount point for this to work. If
2220 your system doesn't have it mounted, you can do so with::
2222 # mount -t cgroup -o blkio none /cgroup
2224 .. option:: cgroup_weight=int
2226 Set the weight of the cgroup to this value. See the documentation that comes
2227 with the kernel, allowed values are in the range of 100..1000.
2229 .. option:: cgroup_nodelete=bool
2231 Normally fio will delete the cgroups it has created after the job
2232 completion. To override this behavior and to leave cgroups around after the
2233 job completion, set ``cgroup_nodelete=1``. This can be useful if one wants
2234 to inspect various cgroup files after job completion. Default: false.
2236 .. option:: flow_id=int
2238 The ID of the flow. If not specified, it defaults to being a global
2239 flow. See :option:`flow`.
2241 .. option:: flow=int
2243 Weight in token-based flow control. If this value is used, then there is a
2244 'flow counter' which is used to regulate the proportion of activity between
2245 two or more jobs. Fio attempts to keep this flow counter near zero. The
2246 ``flow`` parameter stands for how much should be added or subtracted to the
2247 flow counter on each iteration of the main I/O loop. That is, if one job has
2248 ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8
2249 ratio in how much one runs vs the other.
2251 .. option:: flow_watermark=int
2253 The maximum value that the absolute value of the flow counter is allowed to
2254 reach before the job must wait for a lower value of the counter.
2256 .. option:: flow_sleep=int
2258 The period of time, in microseconds, to wait after the flow watermark has
2259 been exceeded before retrying operations.
2261 .. option:: stonewall, wait_for_previous
2263 Wait for preceding jobs in the job file to exit, before starting this
2264 one. Can be used to insert serialization points in the job file. A stone
2265 wall also implies starting a new reporting group, see
2266 :option:`group_reporting`.
2270 When one job finishes, terminate the rest. The default is to wait for each
2271 job to finish, sometimes that is not the desired action.
2273 .. option:: exec_prerun=str
2275 Before running this job, issue the command specified through
2276 :manpage:`system(3)`. Output is redirected in a file called
2277 :file:`jobname.prerun.txt`.
2279 .. option:: exec_postrun=str
2281 After the job completes, issue the command specified though
2282 :manpage:`system(3)`. Output is redirected in a file called
2283 :file:`jobname.postrun.txt`.
2287 Instead of running as the invoking user, set the user ID to this value
2288 before the thread/process does any work.
2292 Set group ID, see :option:`uid`.
2298 .. option:: verify_only
2300 Do not perform specified workload, only verify data still matches previous
2301 invocation of this workload. This option allows one to check data multiple
2302 times at a later date without overwriting it. This option makes sense only
2303 for workloads that write data, and does not support workloads with the
2304 :option:`time_based` option set.
2306 .. option:: do_verify=bool
2308 Run the verify phase after a write phase. Only valid if :option:`verify` is
2311 .. option:: verify=str
2313 If writing to a file, fio can verify the file contents after each iteration
2314 of the job. Each verification method also implies verification of special
2315 header, which is written to the beginning of each block. This header also
2316 includes meta information, like offset of the block, block number, timestamp
2317 when block was written, etc. :option:`verify` can be combined with
2318 :option:`verify_pattern` option. The allowed values are:
2321 Use an md5 sum of the data area and store it in the header of
2325 Use an experimental crc64 sum of the data area and store it in the
2326 header of each block.
2329 Use a crc32c sum of the data area and store it in the header of each
2333 Use hardware assisted crc32c calculation provided on SSE4.2 enabled
2334 processors. Falls back to regular software crc32c, if not supported
2338 Use a crc32 sum of the data area and store it in the header of each
2342 Use a crc16 sum of the data area and store it in the header of each
2346 Use a crc7 sum of the data area and store it in the header of each
2350 Use xxhash as the checksum function. Generally the fastest software
2351 checksum that fio supports.
2354 Use sha512 as the checksum function.
2357 Use sha256 as the checksum function.
2360 Use optimized sha1 as the checksum function.
2363 Use optimized sha3-224 as the checksum function.
2366 Use optimized sha3-256 as the checksum function.
2369 Use optimized sha3-384 as the checksum function.
2372 Use optimized sha3-512 as the checksum function.
2375 This option is deprecated, since now meta information is included in
2376 generic verification header and meta verification happens by
2377 default. For detailed information see the description of the
2378 :option:`verify` setting. This option is kept because of
2379 compatibility's sake with old configurations. Do not use it.
2382 Verify a strict pattern. Normally fio includes a header with some
2383 basic information and checksumming, but if this option is set, only
2384 the specific pattern set with :option:`verify_pattern` is verified.
2387 Only pretend to verify. Useful for testing internals with
2388 :option:`ioengine` `=null`, not for much else.
2390 This option can be used for repeated burn-in tests of a system to make sure
2391 that the written data is also correctly read back. If the data direction
2392 given is a read or random read, fio will assume that it should verify a
2393 previously written file. If the data direction includes any form of write,
2394 the verify will be of the newly written data.
2396 .. option:: verifysort=bool
2398 If true, fio will sort written verify blocks when it deems it faster to read
2399 them back in a sorted manner. This is often the case when overwriting an
2400 existing file, since the blocks are already laid out in the file system. You
2401 can ignore this option unless doing huge amounts of really fast I/O where
2402 the red-black tree sorting CPU time becomes significant. Default: true.
2404 .. option:: verifysort_nr=int
2406 Pre-load and sort verify blocks for a read workload.
2408 .. option:: verify_offset=int
2410 Swap the verification header with data somewhere else in the block before
2411 writing. It is swapped back before verifying.
2413 .. option:: verify_interval=int
2415 Write the verification header at a finer granularity than the
2416 :option:`blocksize`. It will be written for chunks the size of
2417 ``verify_interval``. :option:`blocksize` should divide this evenly.
2419 .. option:: verify_pattern=str
2421 If set, fio will fill the I/O buffers with this pattern. Fio defaults to
2422 filling with totally random bytes, but sometimes it's interesting to fill
2423 with a known pattern for I/O verification purposes. Depending on the width
2424 of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time(it can
2425 be either a decimal or a hex number). The ``verify_pattern`` if larger than
2426 a 32-bit quantity has to be a hex number that starts with either "0x" or
2427 "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o
2428 format, which means that for each block offset will be written and then
2429 verified back, e.g.::
2433 Or use combination of everything::
2435 verify_pattern=0xff%o"abcd"-12
2437 .. option:: verify_fatal=bool
2439 Normally fio will keep checking the entire contents before quitting on a
2440 block verification failure. If this option is set, fio will exit the job on
2441 the first observed failure. Default: false.
2443 .. option:: verify_dump=bool
2445 If set, dump the contents of both the original data block and the data block
2446 we read off disk to files. This allows later analysis to inspect just what
2447 kind of data corruption occurred. Off by default.
2449 .. option:: verify_async=int
2451 Fio will normally verify I/O inline from the submitting thread. This option
2452 takes an integer describing how many async offload threads to create for I/O
2453 verification instead, causing fio to offload the duty of verifying I/O
2454 contents to one or more separate threads. If using this offload option, even
2455 sync I/O engines can benefit from using an :option:`iodepth` setting higher
2456 than 1, as it allows them to have I/O in flight while verifies are running.
2458 .. option:: verify_async_cpus=str
2460 Tell fio to set the given CPU affinity on the async I/O verification
2461 threads. See :option:`cpus_allowed` for the format used.
2463 .. option:: verify_backlog=int
2465 Fio will normally verify the written contents of a job that utilizes verify
2466 once that job has completed. In other words, everything is written then
2467 everything is read back and verified. You may want to verify continually
2468 instead for a variety of reasons. Fio stores the meta data associated with
2469 an I/O block in memory, so for large verify workloads, quite a bit of memory
2470 would be used up holding this meta data. If this option is enabled, fio will
2471 write only N blocks before verifying these blocks.
2473 .. option:: verify_backlog_batch=int
2475 Control how many blocks fio will verify if :option:`verify_backlog` is
2476 set. If not set, will default to the value of :option:`verify_backlog`
2477 (meaning the entire queue is read back and verified). If
2478 ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all
2479 blocks will be verified, if ``verify_backlog_batch`` is larger than
2480 :option:`verify_backlog`, some blocks will be verified more than once.
2482 .. option:: verify_state_save=bool
2484 When a job exits during the write phase of a verify workload, save its
2485 current state. This allows fio to replay up until that point, if the verify
2486 state is loaded for the verify read phase. The format of the filename is,
2489 <type>-<jobname>-<jobindex>-verify.state.
2491 <type> is "local" for a local run, "sock" for a client/server socket
2492 connection, and "ip" (192.168.0.1, for instance) for a networked
2493 client/server connection.
2495 .. option:: verify_state_load=bool
2497 If a verify termination trigger was used, fio stores the current write state
2498 of each thread. This can be used at verification time so that fio knows how
2499 far it should verify. Without this information, fio will run a full
2500 verification pass, according to the settings in the job file used.
2502 .. option:: trim_percentage=int
2504 Number of verify blocks to discard/trim.
2506 .. option:: trim_verify_zero=bool
2508 Verify that trim/discarded blocks are returned as zeroes.
2510 .. option:: trim_backlog=int
2512 Verify that trim/discarded blocks are returned as zeroes.
2514 .. option:: trim_backlog_batch=int
2516 Trim this number of I/O blocks.
2518 .. option:: experimental_verify=bool
2520 Enable experimental verification.
2526 .. option:: steadystate=str:float, ss=str:float
2528 Define the criterion and limit for assessing steady state performance. The
2529 first parameter designates the criterion whereas the second parameter sets
2530 the threshold. When the criterion falls below the threshold for the
2531 specified duration, the job will stop. For example, `iops_slope:0.1%` will
2532 direct fio to terminate the job when the least squares regression slope
2533 falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled
2534 this will apply to all jobs in the group. Below is the list of available
2535 steady state assessment criteria. All assessments are carried out using only
2536 data from the rolling collection window. Threshold limits can be expressed
2537 as a fixed value or as a percentage of the mean in the collection window.
2540 Collect IOPS data. Stop the job if all individual IOPS measurements
2541 are within the specified limit of the mean IOPS (e.g., ``iops:2``
2542 means that all individual IOPS values must be within 2 of the mean,
2543 whereas ``iops:0.2%`` means that all individual IOPS values must be
2544 within 0.2% of the mean IOPS to terminate the job).
2547 Collect IOPS data and calculate the least squares regression
2548 slope. Stop the job if the slope falls below the specified limit.
2551 Collect bandwidth data. Stop the job if all individual bandwidth
2552 measurements are within the specified limit of the mean bandwidth.
2555 Collect bandwidth data and calculate the least squares regression
2556 slope. Stop the job if the slope falls below the specified limit.
2558 .. option:: steadystate_duration=time, ss_dur=time
2560 A rolling window of this duration will be used to judge whether steady state
2561 has been reached. Data will be collected once per second. The default is 0
2562 which disables steady state detection. When the unit is omitted, the
2563 value is given in seconds.
2565 .. option:: steadystate_ramp_time=time, ss_ramp=time
2567 Allow the job to run for the specified duration before beginning data
2568 collection for checking the steady state job termination criterion. The
2569 default is 0. When the unit is omitted, the value is given in seconds.
2572 Measurements and reporting
2573 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2575 .. option:: per_job_logs=bool
2577 If set, this generates bw/clat/iops log with per file private filenames. If
2578 not set, jobs with identical names will share the log filename. Default:
2581 .. option:: group_reporting
2583 It may sometimes be interesting to display statistics for groups of jobs as
2584 a whole instead of for each individual job. This is especially true if
2585 :option:`numjobs` is used; looking at individual thread/process output
2586 quickly becomes unwieldy. To see the final report per-group instead of
2587 per-job, use :option:`group_reporting`. Jobs in a file will be part of the
2588 same reporting group, unless if separated by a :option:`stonewall`, or by
2589 using :option:`new_group`.
2591 .. option:: new_group
2593 Start a new reporting group. See: :option:`group_reporting`. If not given,
2594 all jobs in a file will be part of the same reporting group, unless
2595 separated by a :option:`stonewall`.
2599 By default, fio collects and shows final output results for all jobs
2600 that run. If this option is set to 0, then fio will ignore it in
2601 the final stat output.
2603 .. option:: write_bw_log=str
2605 If given, write a bandwidth log for this job. Can be used to store data of
2606 the bandwidth of the jobs in their lifetime. The included
2607 :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these
2608 text files into nice graphs. See :option:`write_lat_log` for behaviour of
2609 given filename. For this option, the postfix is :file:`_bw.x.log`, where `x`
2610 is the index of the job (`1..N`, where `N` is the number of jobs). If
2611 :option:`per_job_logs` is false, then the filename will not include the job
2612 index. See `Log File Formats`_.
2614 .. option:: write_lat_log=str
2616 Same as :option:`write_bw_log`, except that this option stores I/O
2617 submission, completion, and total latencies instead. If no filename is given
2618 with this option, the default filename of :file:`jobname_type.log` is
2619 used. Even if the filename is given, fio will still append the type of
2620 log. So if one specifies::
2624 The actual log names will be :file:`foo_slat.x.log`, :file:`foo_clat.x.log`,
2625 and :file:`foo_lat.x.log`, where `x` is the index of the job (1..N, where N
2626 is the number of jobs). This helps :command:`fio_generate_plot` find the
2627 logs automatically. If :option:`per_job_logs` is false, then the filename
2628 will not include the job index. See `Log File Formats`_.
2630 .. option:: write_hist_log=str
2632 Same as :option:`write_lat_log`, but writes I/O completion latency
2633 histograms. If no filename is given with this option, the default filename
2634 of :file:`jobname_clat_hist.x.log` is used, where `x` is the index of the
2635 job (1..N, where `N` is the number of jobs). Even if the filename is given,
2636 fio will still append the type of log. If :option:`per_job_logs` is false,
2637 then the filename will not include the job index. See `Log File Formats`_.
2639 .. option:: write_iops_log=str
2641 Same as :option:`write_bw_log`, but writes IOPS. If no filename is given
2642 with this option, the default filename of :file:`jobname_type.x.log` is
2643 used,where `x` is the index of the job (1..N, where `N` is the number of
2644 jobs). Even if the filename is given, fio will still append the type of
2645 log. If :option:`per_job_logs` is false, then the filename will not include
2646 the job index. See `Log File Formats`_.
2648 .. option:: log_avg_msec=int
2650 By default, fio will log an entry in the iops, latency, or bw log for every
2651 I/O that completes. When writing to the disk log, that can quickly grow to a
2652 very large size. Setting this option makes fio average the each log entry
2653 over the specified period of time, reducing the resolution of the log. See
2654 :option:`log_max_value` as well. Defaults to 0, logging all entries.
2656 .. option:: log_hist_msec=int
2658 Same as :option:`log_avg_msec`, but logs entries for completion latency
2659 histograms. Computing latency percentiles from averages of intervals using
2660 :option:`log_avg_msec` is inaccurate. Setting this option makes fio log
2661 histogram entries over the specified period of time, reducing log sizes for
2662 high IOPS devices while retaining percentile accuracy. See
2663 :option:`log_hist_coarseness` as well. Defaults to 0, meaning histogram
2664 logging is disabled.
2666 .. option:: log_hist_coarseness=int
2668 Integer ranging from 0 to 6, defining the coarseness of the resolution of
2669 the histogram logs enabled with :option:`log_hist_msec`. For each increment
2670 in coarseness, fio outputs half as many bins. Defaults to 0, for which
2671 histogram logs contain 1216 latency bins. See `Log File Formats`_.
2673 .. option:: log_max_value=bool
2675 If :option:`log_avg_msec` is set, fio logs the average over that window. If
2676 you instead want to log the maximum value, set this option to 1. Defaults to
2677 0, meaning that averaged values are logged.
2679 .. option:: log_offset=int
2681 If this is set, the iolog options will include the byte offset for the I/O
2682 entry as well as the other data values.
2684 .. option:: log_compression=int
2686 If this is set, fio will compress the I/O logs as it goes, to keep the
2687 memory footprint lower. When a log reaches the specified size, that chunk is
2688 removed and compressed in the background. Given that I/O logs are fairly
2689 highly compressible, this yields a nice memory savings for longer runs. The
2690 downside is that the compression will consume some background CPU cycles, so
2691 it may impact the run. This, however, is also true if the logging ends up
2692 consuming most of the system memory. So pick your poison. The I/O logs are
2693 saved normally at the end of a run, by decompressing the chunks and storing
2694 them in the specified log file. This feature depends on the availability of
2697 .. option:: log_compression_cpus=str
2699 Define the set of CPUs that are allowed to handle online log compression for
2700 the I/O jobs. This can provide better isolation between performance
2701 sensitive jobs, and background compression work.
2703 .. option:: log_store_compressed=bool
2705 If set, fio will store the log files in a compressed format. They can be
2706 decompressed with fio, using the :option:`--inflate-log` command line
2707 parameter. The files will be stored with a :file:`.fz` suffix.
2709 .. option:: log_unix_epoch=bool
2711 If set, fio will log Unix timestamps to the log files produced by enabling
2712 write_type_log for each log type, instead of the default zero-based
2715 .. option:: block_error_percentiles=bool
2717 If set, record errors in trim block-sized units from writes and trims and
2718 output a histogram of how many trims it took to get to errors, and what kind
2719 of error was encountered.
2721 .. option:: bwavgtime=int
2723 Average the calculated bandwidth over the given time. Value is specified in
2724 milliseconds. If the job also does bandwidth logging through
2725 :option:`write_bw_log`, then the minimum of this option and
2726 :option:`log_avg_msec` will be used. Default: 500ms.
2728 .. option:: iopsavgtime=int
2730 Average the calculated IOPS over the given time. Value is specified in
2731 milliseconds. If the job also does IOPS logging through
2732 :option:`write_iops_log`, then the minimum of this option and
2733 :option:`log_avg_msec` will be used. Default: 500ms.
2735 .. option:: disk_util=bool
2737 Generate disk utilization statistics, if the platform supports it.
2740 .. option:: disable_lat=bool
2742 Disable measurements of total latency numbers. Useful only for cutting back
2743 the number of calls to :manpage:`gettimeofday(2)`, as that does impact
2744 performance at really high IOPS rates. Note that to really get rid of a
2745 large amount of these calls, this option must be used with
2746 :option:`disable_slat` and :option:`disable_bw_measurement` as well.
2748 .. option:: disable_clat=bool
2750 Disable measurements of completion latency numbers. See
2751 :option:`disable_lat`.
2753 .. option:: disable_slat=bool
2755 Disable measurements of submission latency numbers. See
2756 :option:`disable_slat`.
2758 .. option:: disable_bw_measurement=bool, disable_bw=bool
2760 Disable measurements of throughput/bandwidth numbers. See
2761 :option:`disable_lat`.
2763 .. option:: clat_percentiles=bool
2765 Enable the reporting of percentiles of completion latencies.
2767 .. option:: percentile_list=float_list
2769 Overwrite the default list of percentiles for completion latencies and the
2770 block error histogram. Each number is a floating number in the range
2771 (0,100], and the maximum length of the list is 20. Use ``:`` to separate the
2772 numbers, and list the numbers in ascending order. For example,
2773 ``--percentile_list=99.5:99.9`` will cause fio to report the values of
2774 completion latency below which 99.5% and 99.9% of the observed latencies
2781 .. option:: exitall_on_error
2783 When one job finishes in error, terminate the rest. The default is to wait
2784 for each job to finish.
2786 .. option:: continue_on_error=str
2788 Normally fio will exit the job on the first observed failure. If this option
2789 is set, fio will continue the job when there is a 'non-fatal error' (EIO or
2790 EILSEQ) until the runtime is exceeded or the I/O size specified is
2791 completed. If this option is used, there are two more stats that are
2792 appended, the total error count and the first error. The error field given
2793 in the stats is the first error that was hit during the run.
2795 The allowed values are:
2798 Exit on any I/O or verify errors.
2801 Continue on read errors, exit on all others.
2804 Continue on write errors, exit on all others.
2807 Continue on any I/O error, exit on all others.
2810 Continue on verify errors, exit on all others.
2813 Continue on all errors.
2816 Backward-compatible alias for 'none'.
2819 Backward-compatible alias for 'all'.
2821 .. option:: ignore_error=str
2823 Sometimes you want to ignore some errors during test in that case you can
2824 specify error list for each error type.
2825 ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for
2826 given error type is separated with ':'. Error may be symbol ('ENOSPC',
2827 'ENOMEM') or integer. Example::
2829 ignore_error=EAGAIN,ENOSPC:122
2831 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from
2834 .. option:: error_dump=bool
2836 If set dump every error even if it is non fatal, true by default. If
2837 disabled only fatal error will be dumped.
2839 Running predefined workloads
2840 ----------------------------
2842 Fio includes predefined profiles that mimic the I/O workloads generated by
2845 .. option:: profile=str
2847 The predefined workload to run. Current profiles are:
2850 Threaded I/O bench (tiotest/tiobench) like workload.
2853 Aerospike Certification Tool (ACT) like workload.
2855 To view a profile's additional options use :option:`--cmdhelp` after specifying
2856 the profile. For example::
2858 $ fio --profile=act --cmdhelp
2863 .. option:: device-names=str
2868 .. option:: load=int
2871 ACT load multiplier. Default: 1.
2873 .. option:: test-duration=time
2876 How long the entire test takes to run. Default: 24h.
2878 .. option:: threads-per-queue=int
2881 Number of read IO threads per device. Default: 8.
2883 .. option:: read-req-num-512-blocks=int
2886 Number of 512B blocks to read at the time. Default: 3.
2888 .. option:: large-block-op-kbytes=int
2891 Size of large block ops in KiB (writes). Default: 131072.
2896 Set to run ACT prep phase.
2898 Tiobench profile options
2899 ~~~~~~~~~~~~~~~~~~~~~~~~
2901 .. option:: size=str
2906 .. option:: block=int
2909 Block size in bytes. Default: 4096.
2911 .. option:: numruns=int
2921 .. option:: threads=int
2926 Interpreting the output
2927 -----------------------
2929 Fio spits out a lot of output. While running, fio will display the status of the
2930 jobs created. An example of that would be::
2932 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]
2934 The characters inside the square brackets denote the current status of each
2935 thread. The possible values (in typical life cycle order) are:
2937 +------+-----+-----------------------------------------------------------+
2939 +======+=====+===========================================================+
2940 | P | | Thread setup, but not started. |
2941 +------+-----+-----------------------------------------------------------+
2942 | C | | Thread created. |
2943 +------+-----+-----------------------------------------------------------+
2944 | I | | Thread initialized, waiting or generating necessary data. |
2945 +------+-----+-----------------------------------------------------------+
2946 | | p | Thread running pre-reading file(s). |
2947 +------+-----+-----------------------------------------------------------+
2948 | | R | Running, doing sequential reads. |
2949 +------+-----+-----------------------------------------------------------+
2950 | | r | Running, doing random reads. |
2951 +------+-----+-----------------------------------------------------------+
2952 | | W | Running, doing sequential writes. |
2953 +------+-----+-----------------------------------------------------------+
2954 | | w | Running, doing random writes. |
2955 +------+-----+-----------------------------------------------------------+
2956 | | M | Running, doing mixed sequential reads/writes. |
2957 +------+-----+-----------------------------------------------------------+
2958 | | m | Running, doing mixed random reads/writes. |
2959 +------+-----+-----------------------------------------------------------+
2960 | | F | Running, currently waiting for :manpage:`fsync(2)` |
2961 +------+-----+-----------------------------------------------------------+
2962 | | V | Running, doing verification of written data. |
2963 +------+-----+-----------------------------------------------------------+
2964 | E | | Thread exited, not reaped by main thread yet. |
2965 +------+-----+-----------------------------------------------------------+
2966 | _ | | Thread reaped, or |
2967 +------+-----+-----------------------------------------------------------+
2968 | X | | Thread reaped, exited with an error. |
2969 +------+-----+-----------------------------------------------------------+
2970 | K | | Thread reaped, exited due to signal. |
2971 +------+-----+-----------------------------------------------------------+
2973 Fio will condense the thread string as not to take up more space on the command
2974 line as is needed. For instance, if you have 10 readers and 10 writers running,
2975 the output would look like this::
2977 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]
2979 Fio will still maintain the ordering, though. So the above means that jobs 1..10
2980 are readers, and 11..20 are writers.
2982 The other values are fairly self explanatory -- number of threads currently
2983 running and doing I/O, the number of currently open files (f=), the rate of I/O
2984 since last check (read speed listed first, then write speed and optionally trim
2985 speed), and the estimated completion percentage and time for the current
2986 running group. It's impossible to estimate runtime of the following groups (if
2987 any). Note that the string is displayed in order, so it's possible to tell which
2988 of the jobs are currently doing what. The first character is the first job
2989 defined in the job file, and so forth.
2991 When fio is done (or interrupted by :kbd:`ctrl-c`), it will show the data for
2992 each thread, group of threads, and disks in that order. For each data direction,
2993 the output looks like::
2995 Client1 (g=0): err= 0:
2996 write: io= 32MiB, bw= 666KiB/s, iops=89 , runt= 50320msec
2997 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
2998 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
2999 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
3000 cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17
3001 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
3002 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3003 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3004 issued r/w: total=0/32768, short=0/0
3005 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
3006 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
3008 The client number is printed, along with the group id and error of that
3009 thread. Below is the I/O statistics, here for writes. In the order listed, they
3013 Number of megabytes I/O performed.
3016 Average bandwidth rate.
3019 Average I/Os performed per second.
3022 The runtime of that thread.
3025 Submission latency (avg being the average, stdev being the standard
3026 deviation). This is the time it took to submit the I/O. For sync I/O,
3027 the slat is really the completion latency, since queue/complete is one
3028 operation there. This value can be in milliseconds or microseconds, fio
3029 will choose the most appropriate base and print that. In the example
3030 above, milliseconds is the best scale. Note: in :option:`--minimal` mode
3031 latencies are always expressed in microseconds.
3034 Completion latency. Same names as slat, this denotes the time from
3035 submission to completion of the I/O pieces. For sync I/O, clat will
3036 usually be equal (or very close) to 0, as the time from submit to
3037 complete is basically just CPU time (I/O has already been done, see slat
3041 Bandwidth. Same names as the xlat stats, but also includes an
3042 approximate percentage of total aggregate bandwidth this thread received
3043 in this group. This last value is only really useful if the threads in
3044 this group are on the same disk, since they are then competing for disk
3048 CPU usage. User and system time, along with the number of context
3049 switches this thread went through, usage of system and user time, and
3050 finally the number of major and minor page faults. The CPU utilization
3051 numbers are averages for the jobs in that reporting group, while the
3052 context and fault counters are summed.
3055 The distribution of I/O depths over the job life time. The numbers are
3056 divided into powers of 2, so for example the 16= entries includes depths
3057 up to that value but higher than the previous entry. In other words, it
3058 covers the range from 16 to 31.
3061 How many pieces of I/O were submitting in a single submit call. Each
3062 entry denotes that amount and below, until the previous entry -- e.g.,
3063 8=100% mean that we submitted anywhere in between 5-8 I/Os per submit
3067 Like the above submit number, but for completions instead.
3070 The number of read/write requests issued, and how many of them were
3074 The distribution of I/O completion latencies. This is the time from when
3075 I/O leaves fio and when it gets completed. The numbers follow the same
3076 pattern as the I/O depths, meaning that 2=1.6% means that 1.6% of the
3077 I/O completed within 2 msecs, 20=12.8% means that 12.8% of the I/O took
3078 more than 10 msecs, but less than (or equal to) 20 msecs.
3080 After each client has been listed, the group statistics are printed. They
3081 will look like this::
3083 Run status group 0 (all jobs):
3084 READ: io=64MB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
3085 WRITE: io=64MB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
3087 For each data direction, it prints:
3090 Number of megabytes I/O performed.
3092 Aggregate bandwidth of threads in this group.
3094 The minimum average bandwidth a thread saw.
3096 The maximum average bandwidth a thread saw.
3098 The smallest runtime of the threads in that group.
3100 The longest runtime of the threads in that group.
3102 And finally, the disk statistics are printed. They will look like this::
3104 Disk stats (read/write):
3105 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
3107 Each value is printed for both reads and writes, with reads first. The
3111 Number of I/Os performed by all groups.
3113 Number of merges I/O the I/O scheduler.
3115 Number of ticks we kept the disk busy.
3117 Total time spent in the disk queue.
3119 The disk utilization. A value of 100% means we kept the disk
3120 busy constantly, 50% would be a disk idling half of the time.
3122 It is also possible to get fio to dump the current output while it is running,
3123 without terminating the job. To do that, send fio the **USR1** signal. You can
3124 also get regularly timed dumps by using the :option:`--status-interval`
3125 parameter, or by creating a file in :file:`/tmp` named
3126 :file:`fio-dump-status`. If fio sees this file, it will unlink it and dump the
3127 current output status.
3133 For scripted usage where you typically want to generate tables or graphs of the
3134 results, fio can output the results in a semicolon separated format. The format
3135 is one long line of values, such as::
3137 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%
3138 A description of this job goes here.
3140 The job description (if provided) follows on a second line.
3142 To enable terse output, use the :option:`--minimal` command line option. The
3143 first value is the version of the terse output format. If the output has to be
3144 changed for some reason, this number will be incremented by 1 to signify that
3147 Split up, the format is as follows:
3151 terse version, fio version, jobname, groupid, error
3155 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3156 Submission latency: min, max, mean, stdev (usec)
3157 Completion latency: min, max, mean, stdev (usec)
3158 Completion latency percentiles: 20 fields (see below)
3159 Total latency: min, max, mean, stdev (usec)
3160 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev
3166 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3167 Submission latency: min, max, mean, stdev (usec)
3168 Completion latency: min, max, mean, stdev(usec)
3169 Completion latency percentiles: 20 fields (see below)
3170 Total latency: min, max, mean, stdev (usec)
3171 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev
3175 user, system, context switches, major faults, minor faults
3179 <=1, 2, 4, 8, 16, 32, >=64
3181 I/O latencies microseconds::
3183 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
3185 I/O latencies milliseconds::
3187 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
3191 Disk name, Read ios, write ios,
3192 Read merges, write merges,
3193 Read ticks, write ticks,
3194 Time spent in queue, disk utilization percentage
3196 Additional Info (dependent on continue_on_error, default off)::
3198 total # errors, first error code
3200 Additional Info (dependent on description being set)::
3204 Completion latency percentiles can be a grouping of up to 20 sets, so for the
3205 terse output fio writes all of them. Each field will look like this::
3209 which is the Xth percentile, and the `usec` latency associated with it.
3211 For disk utilization, all disks used by fio are shown. So for each disk there
3212 will be a disk utilization section.
3218 There are two trace file format that you can encounter. The older (v1) format is
3219 unsupported since version 1.20-rc3 (March 2008). It will still be described
3220 below in case that you get an old trace and want to understand it.
3222 In any case the trace is a simple text file with a single action per line.
3225 Trace file format v1
3226 ~~~~~~~~~~~~~~~~~~~~
3228 Each line represents a single I/O action in the following format::
3232 where `rw=0/1` for read/write, and the offset and length entries being in bytes.
3234 This format is not supported in fio versions => 1.20-rc3.
3237 Trace file format v2
3238 ~~~~~~~~~~~~~~~~~~~~
3240 The second version of the trace file format was added in fio version 1.17. It
3241 allows to access more then one file per trace and has a bigger set of possible
3244 The first line of the trace file has to be::
3248 Following this can be lines in two different formats, which are described below.
3250 The file management format::
3254 The filename is given as an absolute path. The action can be one of these:
3257 Add the given filename to the trace.
3259 Open the file with the given filename. The filename has to have
3260 been added with the **add** action before.
3262 Close the file with the given filename. The file has to have been
3266 The file I/O action format::
3268 filename action offset length
3270 The `filename` is given as an absolute path, and has to have been added and
3271 opened before it can be used with this format. The `offset` and `length` are
3272 given in bytes. The `action` can be one of these:
3275 Wait for `offset` microseconds. Everything below 100 is discarded.
3276 The time is relative to the previous `wait` statement.
3278 Read `length` bytes beginning from `offset`.
3280 Write `length` bytes beginning from `offset`.
3282 :manpage:`fsync(2)` the file.
3284 :manpage:`fdatasync(2)` the file.
3286 Trim the given file from the given `offset` for `length` bytes.
3288 CPU idleness profiling
3289 ----------------------
3291 In some cases, we want to understand CPU overhead in a test. For example, we
3292 test patches for the specific goodness of whether they reduce CPU usage.
3293 Fio implements a balloon approach to create a thread per CPU that runs at idle
3294 priority, meaning that it only runs when nobody else needs the cpu.
3295 By measuring the amount of work completed by the thread, idleness of each CPU
3296 can be derived accordingly.
3298 An unit work is defined as touching a full page of unsigned characters. Mean and
3299 standard deviation of time to complete an unit work is reported in "unit work"
3300 section. Options can be chosen to report detailed percpu idleness or overall
3301 system idleness by aggregating percpu stats.
3304 Verification and triggers
3305 -------------------------
3307 Fio is usually run in one of two ways, when data verification is done. The first
3308 is a normal write job of some sort with verify enabled. When the write phase has
3309 completed, fio switches to reads and verifies everything it wrote. The second
3310 model is running just the write phase, and then later on running the same job
3311 (but with reads instead of writes) to repeat the same I/O patterns and verify
3312 the contents. Both of these methods depend on the write phase being completed,
3313 as fio otherwise has no idea how much data was written.
3315 With verification triggers, fio supports dumping the current write state to
3316 local files. Then a subsequent read verify workload can load this state and know
3317 exactly where to stop. This is useful for testing cases where power is cut to a
3318 server in a managed fashion, for instance.
3320 A verification trigger consists of two things:
3322 1) Storing the write state of each job.
3323 2) Executing a trigger command.
3325 The write state is relatively small, on the order of hundreds of bytes to single
3326 kilobytes. It contains information on the number of completions done, the last X
3329 A trigger is invoked either through creation ('touch') of a specified file in
3330 the system, or through a timeout setting. If fio is run with
3331 :option:`--trigger-file` = :file:`/tmp/trigger-file`, then it will continually
3332 check for the existence of :file:`/tmp/trigger-file`. When it sees this file, it
3333 will fire off the trigger (thus saving state, and executing the trigger
3336 For client/server runs, there's both a local and remote trigger. If fio is
3337 running as a server backend, it will send the job states back to the client for
3338 safe storage, then execute the remote trigger, if specified. If a local trigger
3339 is specified, the server will still send back the write state, but the client
3340 will then execute the trigger.
3342 Verification trigger example
3343 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3345 Lets say we want to run a powercut test on the remote machine 'server'. Our
3346 write workload is in :file:`write-test.fio`. We want to cut power to 'server' at
3347 some point during the run, and we'll run this test from the safety or our local
3348 machine, 'localbox'. On the server, we'll start the fio backend normally::
3350 server# fio --server
3352 and on the client, we'll fire off the workload::
3354 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger-remote="bash -c \"echo b > /proc/sysrq-triger\""
3356 We set :file:`/tmp/my-trigger` as the trigger file, and we tell fio to execute::
3358 echo b > /proc/sysrq-trigger
3360 on the server once it has received the trigger and sent us the write state. This
3361 will work, but it's not **really** cutting power to the server, it's merely
3362 abruptly rebooting it. If we have a remote way of cutting power to the server
3363 through IPMI or similar, we could do that through a local trigger command
3364 instead. Lets assume we have a script that does IPMI reboot of a given hostname,
3365 ipmi-reboot. On localbox, we could then have run fio with a local trigger
3368 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger="ipmi-reboot server"
3370 For this case, fio would wait for the server to send us the write state, then
3371 execute ``ipmi-reboot server`` when that happened.
3373 Loading verify state
3374 ~~~~~~~~~~~~~~~~~~~~
3376 To load store write state, read verification job file must contain the
3377 :option:`verify_state_load` option. If that is set, fio will load the previously
3378 stored state. For a local fio run this is done by loading the files directly,
3379 and on a client/server run, the server backend will ask the client to send the
3380 files over and load them from there.
3386 Fio supports a variety of log file formats, for logging latencies, bandwidth,
3387 and IOPS. The logs share a common format, which looks like this:
3389 *time* (`msec`), *value*, *data direction*, *offset*
3391 Time for the log entry is always in milliseconds. The *value* logged depends
3392 on the type of log, it will be one of the following:
3395 Value is latency in usecs
3401 *Data direction* is one of the following:
3410 The *offset* is the offset, in bytes, from the start of the file, for that
3411 particular I/O. The logging of the offset can be toggled with
3412 :option:`log_offset`.
3414 If windowed logging is enabled through :option:`log_avg_msec` then fio doesn't
3415 log individual I/Os. Instead of logs the average values over the specified period
3416 of time. Since 'data direction' and 'offset' are per-I/O values, they aren't
3417 applicable if windowed logging is enabled. If windowed logging is enabled and
3418 :option:`log_max_value` is set, then fio logs maximum values in that window
3419 instead of averages.
3425 Normally fio is invoked as a stand-alone application on the machine where the
3426 I/O workload should be generated. However, the frontend and backend of fio can
3427 be run separately. Ie the fio server can generate an I/O workload on the "Device
3428 Under Test" while being controlled from another machine.
3430 Start the server on the machine which has access to the storage DUT::
3434 where args defines what fio listens to. The arguments are of the form
3435 ``type,hostname`` or ``IP,port``. *type* is either ``ip`` (or ip4) for TCP/IP
3436 v4, ``ip6`` for TCP/IP v6, or ``sock`` for a local unix domain socket.
3437 *hostname* is either a hostname or IP address, and *port* is the port to listen
3438 to (only valid for TCP/IP, not a local socket). Some examples:
3442 Start a fio server, listening on all interfaces on the default port (8765).
3444 2) ``fio --server=ip:hostname,4444``
3446 Start a fio server, listening on IP belonging to hostname and on port 4444.
3448 3) ``fio --server=ip6:::1,4444``
3450 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
3452 4) ``fio --server=,4444``
3454 Start a fio server, listening on all interfaces on port 4444.
3456 5) ``fio --server=1.2.3.4``
3458 Start a fio server, listening on IP 1.2.3.4 on the default port.
3460 6) ``fio --server=sock:/tmp/fio.sock``
3462 Start a fio server, listening on the local socket /tmp/fio.sock.
3464 Once a server is running, a "client" can connect to the fio server with::
3466 fio <local-args> --client=<server> <remote-args> <job file(s)>
3468 where `local-args` are arguments for the client where it is running, `server`
3469 is the connect string, and `remote-args` and `job file(s)` are sent to the
3470 server. The `server` string follows the same format as it does on the server
3471 side, to allow IP/hostname/socket and port strings.
3473 Fio can connect to multiple servers this way::
3475 fio --client=<server1> <job file(s)> --client=<server2> <job file(s)>
3477 If the job file is located on the fio server, then you can tell the server to
3478 load a local file as well. This is done by using :option:`--remote-config` ::
3480 fio --client=server --remote-config /path/to/file.fio
3482 Then fio will open this local (to the server) job file instead of being passed
3483 one from the client.
3485 If you have many servers (example: 100 VMs/containers), you can input a pathname
3486 of a file containing host IPs/names as the parameter value for the
3487 :option:`--client` option. For example, here is an example :file:`host.list`
3488 file containing 2 hostnames::
3490 host1.your.dns.domain
3491 host2.your.dns.domain
3493 The fio command would then be::
3495 fio --client=host.list <job file(s)>
3497 In this mode, you cannot input server-specific parameters or job files -- all
3498 servers receive the same job file.
3500 In order to let ``fio --client`` runs use a shared filesystem from multiple
3501 hosts, ``fio --client`` now prepends the IP address of the server to the
3502 filename. For example, if fio is using directory :file:`/mnt/nfs/fio` and is
3503 writing filename :file:`fileio.tmp`, with a :option:`--client` `hostfile`
3504 containing two hostnames ``h1`` and ``h2`` with IP addresses 192.168.10.120 and
3505 192.168.10.121, then fio will create two files::
3507 /mnt/nfs/fio/192.168.10.120.fileio.tmp
3508 /mnt/nfs/fio/192.168.10.121.fileio.tmp