blob: 22c32fd0f59c1deeba9b96bd3f563064485c7b52 [file] [log] [blame]
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001.TH fio 1 "September 2007" "User Manual"
2.SH NAME
3fio \- flexible I/O tester
4.SH SYNOPSIS
5.B fio
6[\fIoptions\fR] [\fIjobfile\fR]...
7.SH DESCRIPTION
8.B fio
9is a tool that will spawn a number of threads or processes doing a
10particular type of I/O action as specified by the user.
11The typical use of fio is to write a job file matching the I/O load
12one wants to simulate.
13.SH OPTIONS
14.TP
Jens Axboe49da1242011-10-13 20:17:02 +020015.BI \-\-debug \fR=\fPtype
16Enable verbose tracing of various fio actions. May be `all' for all types
17or individual types separated by a comma (eg \-\-debug=io,file). `help' will
18list all available tracing options.
19.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +020020.BI \-\-output \fR=\fPfilename
21Write output to \fIfilename\fR.
22.TP
liang xieb2cecdc2012-08-31 08:22:42 -070023.BI \-\-runtime \fR=\fPruntime
24Limit run time to \fIruntime\fR seconds.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020025.TP
26.B \-\-latency\-log
27Generate per-job latency logs.
28.TP
29.B \-\-bandwidth\-log
30Generate per-job bandwidth logs.
31.TP
32.B \-\-minimal
Aaron Carrolld1429b52007-09-18 08:10:57 +020033Print statistics in a terse, semicolon-delimited format.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020034.TP
Jens Axboe49da1242011-10-13 20:17:02 +020035.B \-\-version
36Display version information and exit.
37.TP
Jens Axboe065248b2011-10-13 20:51:05 +020038.BI \-\-terse\-version \fR=\fPversion
Jens Axboe4d658652011-10-17 15:05:47 +020039Set terse version output format (Current version 3, or older version 2).
Jens Axboe49da1242011-10-13 20:17:02 +020040.TP
41.B \-\-help
42Display usage information and exit.
43.TP
44.BI \-\-cmdhelp \fR=\fPcommand
45Print help information for \fIcommand\fR. May be `all' for all commands.
46.TP
Steven Langde890a12011-11-09 14:03:34 +010047.BI \-\-enghelp \fR=\fPioengine[,command]
48List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR.
49.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +020050.BI \-\-showcmd \fR=\fPjobfile
51Convert \fIjobfile\fR to a set of command-line options.
52.TP
53.B \-\-readonly
54Enable read-only safety checks.
55.TP
56.BI \-\-eta \fR=\fPwhen
57Specifies when real-time ETA estimate should be printed. \fIwhen\fR may
58be one of `always', `never' or `auto'.
59.TP
Jens Axboe49da1242011-10-13 20:17:02 +020060.BI \-\-readonly
61Turn on safety read-only checks, preventing any attempted write.
62.TP
Aaron Carrollc0a5d352008-02-26 23:10:39 +010063.BI \-\-section \fR=\fPsec
Jens Axboe49da1242011-10-13 20:17:02 +020064Only run section \fIsec\fR from job file. Multiple of these options can be given, adding more sections to run.
Aaron Carrollc0a5d352008-02-26 23:10:39 +010065.TP
Jens Axboe49da1242011-10-13 20:17:02 +020066.BI \-\-alloc\-size \fR=\fPkb
67Set the internal smalloc pool size to \fIkb\fP kilobytes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020068.TP
Jens Axboe49da1242011-10-13 20:17:02 +020069.BI \-\-warnings\-fatal
70All fio parser warnings are fatal, causing fio to exit with an error.
Jens Axboe91837882008-02-05 12:02:07 +010071.TP
Jens Axboe49da1242011-10-13 20:17:02 +020072.BI \-\-max\-jobs \fR=\fPnr
Martin Steigerwald57e118a2012-05-07 17:06:13 +020073Set the maximum allowed number of jobs (threads/processes) to support.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020074.TP
Jens Axboe49da1242011-10-13 20:17:02 +020075.BI \-\-server \fR=\fPargs
76Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section.
Jens Axboef57a9c52011-09-09 21:01:37 +020077.TP
Jens Axboe49da1242011-10-13 20:17:02 +020078.BI \-\-daemonize \fR=\fPpidfile
79Background a fio server, writing the pid to the given pid file.
80.TP
81.BI \-\-client \fR=\fPhost
82Instead of running the jobs locally, send and run them on the given host.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020083.SH "JOB FILE FORMAT"
84Job files are in `ini' format. They consist of one or more
85job definitions, which begin with a job name in square brackets and
86extend to the next job name. The job name can be any ASCII string
87except `global', which has a special meaning. Following the job name is
88a sequence of zero or more parameters, one per line, that define the
89behavior of the job. Any line starting with a `;' or `#' character is
Aaron Carrolld1429b52007-09-18 08:10:57 +020090considered a comment and ignored.
Aaron Carrolld9956b62007-11-16 12:12:45 +010091.P
92If \fIjobfile\fR is specified as `-', the job file will be read from
93standard input.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020094.SS "Global Section"
95The global section contains default parameters for jobs specified in the
96job file. A job is only affected by global sections residing above it,
97and there may be any number of global sections. Specific job definitions
98may override any parameter set in global sections.
99.SH "JOB PARAMETERS"
100.SS Types
101Some parameters may take arguments of a specific type. The types used are:
102.TP
103.I str
104String: a sequence of alphanumeric characters.
105.TP
106.I int
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200107SI integer: a whole number, possibly containing a suffix denoting the base unit
Jens Axboeb09da8f2009-07-17 23:16:17 +0200108of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
109kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
110respectively. The suffix is not case sensitive. If prefixed with '0x', the
Martin Steigerwald5982a922011-06-27 16:07:24 +0200111value is assumed to be base 16 (hexadecimal). A suffix may include a trailing 'b',
112for instance 'kb' is identical to 'k'. You can specify a base 10 value
Jens Axboe57fc29f2010-06-23 22:24:07 +0200113by using 'KiB', 'MiB', 'GiB', etc. This is useful for disk drives where
114values are often given in base 10 values. Specifying '30GiB' will get you
11530*1000^3 bytes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200116.TP
117.I bool
118Boolean: a true or false value. `0' denotes false, `1' denotes true.
119.TP
120.I irange
121Integer range: a range of integers specified in the format
Aaron Carrolld1429b52007-09-18 08:10:57 +0200122\fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
123\fIupper\fR may contain a suffix as described above. If an option allows two
124sets of ranges, they are separated with a `,' or `/' character. For example:
125`8\-8k/8M\-4G'.
Yu-ju Hong83349192011-08-13 00:53:44 +0200126.TP
127.I float_list
128List of floating numbers: A list of floating numbers, separated by
129a ':' charcater.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200130.SS "Parameter List"
131.TP
132.BI name \fR=\fPstr
Aaron Carrolld9956b62007-11-16 12:12:45 +0100133May be used to override the job name. On the command line, this parameter
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200134has the special purpose of signalling the start of a new job.
135.TP
136.BI description \fR=\fPstr
137Human-readable description of the job. It is printed when the job is run, but
138otherwise has no special purpose.
139.TP
140.BI directory \fR=\fPstr
141Prefix filenames with this directory. Used to place files in a location other
142than `./'.
143.TP
144.BI filename \fR=\fPstr
145.B fio
146normally makes up a file name based on the job name, thread number, and file
Aaron Carrolld1429b52007-09-18 08:10:57 +0200147number. If you want to share files between threads in a job or several jobs,
Steven Langde890a12011-11-09 14:03:34 +0100148specify a \fIfilename\fR for each of them to override the default.
149If the I/O engine is file-based, you can specify
Aaron Carrolld1429b52007-09-18 08:10:57 +0200150a number of files by separating the names with a `:' character. `\-' is a
151reserved name, meaning stdin or stdout, depending on the read/write direction
152set.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200153.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200154.BI lockfile \fR=\fPstr
155Fio defaults to not locking any files before it does IO to them. If a file or
156file descriptor is shared, fio can serialize IO to that file to make the end
157result consistent. This is usual for emulating real workloads that share files.
158The lock modes are:
159.RS
160.RS
161.TP
162.B none
163No locking. This is the default.
164.TP
165.B exclusive
166Only one thread or process may do IO at the time, excluding all others.
167.TP
168.B readwrite
169Read-write locking on the file. Many readers may access the file at the same
170time, but writes get exclusive access.
171.RE
172.P
173The option may be post-fixed with a lock batch number. If set, then each
174thread/process may do that amount of IOs to the file before giving up the lock.
175Since lock acquisition is expensive, batching the lock/unlocks will speed up IO.
176.RE
177.P
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200178.BI opendir \fR=\fPstr
179Recursively open any files below directory \fIstr\fR.
180.TP
181.BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
182Type of I/O pattern. Accepted values are:
183.RS
184.RS
185.TP
186.B read
Aaron Carrolld1429b52007-09-18 08:10:57 +0200187Sequential reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200188.TP
189.B write
Aaron Carrolld1429b52007-09-18 08:10:57 +0200190Sequential writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200191.TP
192.B randread
Aaron Carrolld1429b52007-09-18 08:10:57 +0200193Random reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200194.TP
195.B randwrite
Aaron Carrolld1429b52007-09-18 08:10:57 +0200196Random writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200197.TP
Jens Axboe10b023d2012-03-23 13:40:06 +0100198.B rw, readwrite
Aaron Carrolld1429b52007-09-18 08:10:57 +0200199Mixed sequential reads and writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200200.TP
201.B randrw
Aaron Carrolld1429b52007-09-18 08:10:57 +0200202Mixed random reads and writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200203.RE
204.P
Jens Axboe38dad622010-07-20 14:46:00 -0600205For mixed I/O, the default split is 50/50. For certain types of io the result
206may still be skewed a bit, since the speed may be different. It is possible to
Jens Axboe3b7fa9e2012-04-26 19:39:47 +0200207specify a number of IO's to do before getting a new offset, this is done by
Jens Axboe38dad622010-07-20 14:46:00 -0600208appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
209would look like \fBrw=randread:8\fR for passing in an offset modifier with a
Jens Axboe059b0802011-08-25 09:09:37 +0200210value of 8. If the postfix is used with a sequential IO pattern, then the value
211specified will be added to the generated offset for each IO. For instance,
212using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
213into sequential IO with holes. See the \fBrw_sequencer\fR option.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200214.RE
215.TP
Jens Axboe38dad622010-07-20 14:46:00 -0600216.BI rw_sequencer \fR=\fPstr
217If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
218then this option controls how that number modifies the IO offset being
219generated. Accepted values are:
220.RS
221.RS
222.TP
223.B sequential
224Generate sequential offset
225.TP
226.B identical
227Generate the same offset
228.RE
229.P
230\fBsequential\fR is only useful for random IO, where fio would normally
231generate a new random offset for every IO. If you append eg 8 to randread, you
232would get a new random offset for every 8 IO's. The result would be a seek for
233only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
234that. As sequential IO is already sequential, setting \fBsequential\fR for that
235would not result in any differences. \fBidentical\fR behaves in a similar
236fashion, except it sends the same offset 8 number of times before generating a
237new offset.
238.RE
239.P
240.TP
Jens Axboe90fef2d2009-07-17 22:33:32 +0200241.BI kb_base \fR=\fPint
242The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
243manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
244reasons. Allow values are 1024 or 1000, with 1024 being the default.
245.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200246.BI randrepeat \fR=\fPbool
247Seed the random number generator in a predictable way so results are repeatable
Aaron Carrolld1429b52007-09-18 08:10:57 +0200248across runs. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200249.TP
Jens Axboe2615cc42011-03-28 09:35:09 +0200250.BI use_os_rand \fR=\fPbool
251Fio can either use the random generator supplied by the OS to generator random
252offsets, or it can use it's own internal generator (based on Tausworthe).
253Default is to use the internal generator, which is often of better quality and
254faster. Default: false.
255.TP
Eric Gourioua596f042011-06-17 09:11:45 +0200256.BI fallocate \fR=\fPstr
257Whether pre-allocation is performed when laying down files. Accepted values
258are:
259.RS
260.RS
261.TP
262.B none
263Do not pre-allocate space.
264.TP
265.B posix
266Pre-allocate via posix_fallocate().
267.TP
268.B keep
269Pre-allocate via fallocate() with FALLOC_FL_KEEP_SIZE set.
270.TP
271.B 0
272Backward-compatible alias for 'none'.
273.TP
274.B 1
275Backward-compatible alias for 'posix'.
276.RE
277.P
278May not be available on all supported platforms. 'keep' is only
279available on Linux. If using ZFS on Solaris this must be set to 'none'
280because ZFS doesn't support it. Default: 'posix'.
281.RE
Jens Axboe7bc8c2c2010-01-28 11:31:31 +0100282.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200283.BI fadvise_hint \fR=\fPbool
Zhu Yanhai23a7b042012-01-02 14:32:43 +0100284Use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
Aaron Carrolld1429b52007-09-18 08:10:57 +0200285are likely to be issued. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200286.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100287.BI size \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200288Total size of I/O for this job. \fBfio\fR will run until this many bytes have
289been transfered, unless limited by other options (\fBruntime\fR, for instance).
Jens Axboed7c8be02010-11-25 08:21:39 +0100290Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
Jens Axboed6667262010-06-25 11:32:48 +0200291divided between the available files for the job. If not set, fio will use the
292full size of the given files or devices. If the the files do not exist, size
Jens Axboe7bb59102011-07-12 19:47:03 +0200293must be given. It is also possible to give size as a percentage between 1 and
294100. If size=20% is given, fio will use 20% of the full size of the given files
295or devices.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200296.TP
Jens Axboe74586c12011-01-20 10:16:03 -0700297.BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200298Sets size to something really large and waits for ENOSPC (no space left on
299device) as the terminating condition. Only makes sense with sequential write.
300For a read workload, the mount point will be filled first then IO started on
Jens Axboe4f124322011-01-19 15:35:26 -0700301the result. This option doesn't make sense if operating on a raw device node,
302since the size of that is already known by the file system. Additionally,
303writing beyond end-of-device will not return ENOSPC there.
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200304.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200305.BI filesize \fR=\fPirange
306Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
Aaron Carrolld1429b52007-09-18 08:10:57 +0200307for files at random within the given range, limited to \fBsize\fR in total (if
308that is given). If \fBfilesize\fR is not specified, each created file is the
309same size.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200310.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100311.BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200312Block size for I/O units. Default: 4k. Values for reads and writes can be
Jens Axboe656ebab2010-04-13 10:39:14 +0200313specified separately in the format \fIread\fR,\fIwrite\fR, either of
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200314which may be empty to leave that value at its default.
315.TP
Jens Axboe91837882008-02-05 12:02:07 +0100316.BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
Aaron Carrolld1429b52007-09-18 08:10:57 +0200317Specify a range of I/O block sizes. The issued I/O unit will always be a
318multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
Jens Axboe91837882008-02-05 12:02:07 +0100319to both reads and writes if only one range is given, but can be specified
Jens Axboe656ebab2010-04-13 10:39:14 +0200320separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
Jens Axboe91837882008-02-05 12:02:07 +0100321Also (see \fBblocksize\fR).
322.TP
323.BI bssplit \fR=\fPstr
324This option allows even finer grained control of the block sizes issued,
325not just even splits between them. With this option, you can weight various
326block sizes for exact control of the issued IO for a job that has mixed
327block sizes. The format of the option is bssplit=blocksize/percentage,
Martin Steigerwald5982a922011-06-27 16:07:24 +0200328optionally adding as many definitions as needed separated by a colon.
Jens Axboe91837882008-02-05 12:02:07 +0100329Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
Jens Axboec83cdd32009-04-24 14:23:59 +0200330blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
331splits to reads and writes. The format is identical to what the
332\fBbs\fR option accepts, the read and write parts are separated with a
333comma.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200334.TP
335.B blocksize_unaligned\fR,\fP bs_unaligned
Aaron Carrolld1429b52007-09-18 08:10:57 +0200336If set, any size in \fBblocksize_range\fR may be used. This typically won't
337work with direct I/O, as that normally requires sector alignment.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200338.TP
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100339.BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
Martin Steigerwald639ce0f2009-05-20 11:33:49 +0200340At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
341the minimum blocksize given. Minimum alignment is typically 512b
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100342for using direct IO, though it usually depends on the hardware block size.
343This option is mutually exclusive with using a random map for files, so it
344will turn off that option.
Jens Axboe43602662009-03-14 20:08:47 +0100345.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200346.B zero_buffers
347Initialise buffers with all zeros. Default: fill buffers with random data.
348.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100349.B refill_buffers
350If this option is given, fio will refill the IO buffers on every submit. The
351default is to only fill it at init time and reuse that data. Only makes sense
352if zero_buffers isn't specified, naturally. If data verification is enabled,
353refill_buffers is also automatically enabled.
354.TP
Jens Axboefd684182011-09-19 09:24:44 +0200355.BI scramble_buffers \fR=\fPbool
356If \fBrefill_buffers\fR is too costly and the target is using data
357deduplication, then setting this option will slightly modify the IO buffer
358contents to defeat normal de-dupe attempts. This is not enough to defeat
359more clever block compression attempts, but it will stop naive dedupe
360of blocks. Default: true.
361.TP
Jens Axboec5751c62012-03-15 15:02:56 +0100362.BI buffer_compress_percentage \fR=\fPint
363If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
364that compress to the specified level. Fio does this by providing a mix of
365random data and zeroes. Note that this is per block size unit, for file/disk
366wide compression level that matches this setting, you'll also want to set
367\fBrefill_buffers\fR.
368.TP
369.BI buffer_compress_chunk \fR=\fPint
370See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
371big the ranges of random data and zeroed data is. Without this set, fio will
372provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
373the remaining zeroed. With this set to some chunk size smaller than the block
374size, fio can alternate random and zeroed data throughout the IO buffer.
375.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200376.BI nrfiles \fR=\fPint
377Number of files to use for this job. Default: 1.
378.TP
379.BI openfiles \fR=\fPint
380Number of files to keep open at the same time. Default: \fBnrfiles\fR.
381.TP
382.BI file_service_type \fR=\fPstr
383Defines how files to service are selected. The following types are defined:
384.RS
385.RS
386.TP
387.B random
388Choose a file at random
389.TP
390.B roundrobin
391Round robin over open files (default).
Jens Axboe6b7f6852009-03-09 14:22:56 +0100392.B sequential
393Do each file in the set sequentially.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200394.RE
395.P
396The number of I/Os to issue before switching a new file can be specified by
397appending `:\fIint\fR' to the service type.
398.RE
399.TP
400.BI ioengine \fR=\fPstr
401Defines how the job issues I/O. The following types are defined:
402.RS
403.RS
404.TP
405.B sync
406Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
407position the I/O location.
408.TP
gurudas paia31041e2007-10-23 15:12:30 +0200409.B psync
410Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
411.TP
Jens Axboe91837882008-02-05 12:02:07 +0100412.B vsync
413Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
414coalescing adjacents IOs into a single submission.
415.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200416.B libaio
Steven Langde890a12011-11-09 14:03:34 +0100417Linux native asynchronous I/O. This ioengine defines engine specific options.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200418.TP
419.B posixaio
Bruce Cran03e20d62011-01-02 20:14:54 +0100420POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
421.TP
422.B solarisaio
423Solaris native asynchronous I/O.
424.TP
425.B windowsaio
426Windows native asynchronous I/O.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200427.TP
428.B mmap
Aaron Carrolld1429b52007-09-18 08:10:57 +0200429File is memory mapped with \fImmap\fR\|(2) and data copied using
430\fImemcpy\fR\|(3).
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200431.TP
432.B splice
Aaron Carrolld1429b52007-09-18 08:10:57 +0200433\fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
434transfer data from user-space to the kernel.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200435.TP
436.B syslet-rw
437Use the syslet system calls to make regular read/write asynchronous.
438.TP
439.B sg
440SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
Aaron Carrolld1429b52007-09-18 08:10:57 +0200441the target is an sg character device, we use \fIread\fR\|(2) and
442\fIwrite\fR\|(2) for asynchronous I/O.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200443.TP
444.B null
445Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
446itself and for debugging and testing purposes.
447.TP
448.B net
Steven Langde890a12011-11-09 14:03:34 +0100449Transfer over the network. The protocol to be used can be defined with the
450\fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
451\fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
452This ioengine defines engine specific options.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200453.TP
454.B netsplice
455Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
Steven Langde890a12011-11-09 14:03:34 +0100456and send/receive. This ioengine defines engine specific options.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200457.TP
gurudas pai53aec0a2007-10-05 13:20:18 +0200458.B cpuio
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200459Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
460\fBcpucycles\fR parameters.
461.TP
462.B guasi
463The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
464approach to asycnronous I/O.
Aaron Carrolld1429b52007-09-18 08:10:57 +0200465.br
466See <http://www.xmailserver.org/guasi\-lib.html>.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200467.TP
ren yufei21b8aee2011-08-01 10:01:57 +0200468.B rdma
Bart Van Assche85286c52011-08-07 21:50:51 +0200469The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
470and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
ren yufei21b8aee2011-08-01 10:01:57 +0200471.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200472.B external
473Loads an external I/O engine object file. Append the engine filename as
474`:\fIenginepath\fR'.
Dmitry Monakhovd54fce82012-09-20 15:37:17 +0400475.TP
476.B falloc
477 IO engine that does regular linux native fallocate callt to simulate data
478transfer as fio ioengine
479.br
480 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
481.br
Jens Axboe0981fd72012-09-20 19:23:02 +0200482 DIR_WRITE does fallocate(,mode = 0)
Dmitry Monakhovd54fce82012-09-20 15:37:17 +0400483.br
484 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
485.TP
486.B e4defrag
487IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
488request to DDIR_WRITE event
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200489.RE
Jens Axboe595e1732012-12-05 21:15:01 +0100490.P
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200491.RE
492.TP
493.BI iodepth \fR=\fPint
Sebastian Kayser8489dae2010-12-01 22:28:47 +0100494Number of I/O units to keep in flight against the file. Note that increasing
495iodepth beyond 1 will not affect synchronous ioengines (except for small
Jens Axboeee72ca02010-12-02 20:05:37 +0100496degress when verify_async is in use). Even async engines my impose OS
497restrictions causing the desired depth not to be achieved. This may happen on
498Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
499not async on that OS. Keep an eye on the IO depth distribution in the
500fio output to verify that the achieved depth is as expected. Default: 1.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200501.TP
502.BI iodepth_batch \fR=\fPint
503Number of I/Os to submit at once. Default: \fBiodepth\fR.
504.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200505.BI iodepth_batch_complete \fR=\fPint
506This defines how many pieces of IO to retrieve at once. It defaults to 1 which
507 means that we'll ask for a minimum of 1 IO in the retrieval process from the
508kernel. The IO retrieval will go on until we hit the limit set by
509\fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
510completed events before queuing more IO. This helps reduce IO latency, at the
511cost of more retrieval system calls.
512.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200513.BI iodepth_low \fR=\fPint
514Low watermark indicating when to start filling the queue again. Default:
515\fBiodepth\fR.
516.TP
517.BI direct \fR=\fPbool
518If true, use non-buffered I/O (usually O_DIRECT). Default: false.
519.TP
520.BI buffered \fR=\fPbool
521If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
522Default: true.
523.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100524.BI offset \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200525Offset in the file to start I/O. Data before the offset will not be touched.
526.TP
Jens Axboe591e9e02012-03-15 14:50:58 +0100527.BI offset_increment \fR=\fPint
528If this is provided, then the real offset becomes the
529offset + offset_increment * thread_number, where the thread number is a counter
530that starts at 0 and is incremented for each job. This option is useful if
531there are several jobs which are intended to operate on a file in parallel in
532disjoint segments, with even spacing between the starting points.
533.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200534.BI fsync \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200535How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
5360, don't sync. Default: 0.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200537.TP
Jens Axboe5f9099e2009-06-16 22:40:26 +0200538.BI fdatasync \fR=\fPint
539Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
540data parts of the file. Default: 0.
541.TP
Jens Axboee76b1da2010-03-09 20:49:54 +0100542.BI sync_file_range \fR=\fPstr:int
543Use sync_file_range() for every \fRval\fP number of write operations. Fio will
544track range of writes that have happened since the last sync_file_range() call.
545\fRstr\fP can currently be one or more of:
546.RS
547.TP
548.B wait_before
549SYNC_FILE_RANGE_WAIT_BEFORE
550.TP
551.B write
552SYNC_FILE_RANGE_WRITE
553.TP
554.B wait_after
555SYNC_FILE_RANGE_WRITE
556.TP
557.RE
558.P
559So if you do sync_file_range=wait_before,write:8, fio would use
560\fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
561Also see the sync_file_range(2) man page. This option is Linux specific.
562.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200563.BI overwrite \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200564If writing, setup the file first and do overwrites. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200565.TP
566.BI end_fsync \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200567Sync file contents when job exits. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200568.TP
569.BI fsync_on_close \fR=\fPbool
570If true, sync file contents on close. This differs from \fBend_fsync\fR in that
Aaron Carrolld1429b52007-09-18 08:10:57 +0200571it will happen on every close, not just at the end of the job. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200572.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200573.BI rwmixread \fR=\fPint
574Percentage of a mixed workload that should be reads. Default: 50.
575.TP
576.BI rwmixwrite \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200577Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
Jens Axboec35dd7a2009-06-10 08:39:16 +0200578\fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
579overrides the first. This may interfere with a given rate setting, if fio is
580asked to limit reads or writes to a certain rate. If that is the case, then
581the distribution may be skewed. Default: 50.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200582.TP
Jens Axboe92d42d62012-11-15 15:38:32 -0700583.BI random_distribution \fR=\fPstr:float
584By default, fio will use a completely uniform random distribution when asked
585to perform random IO. Sometimes it is useful to skew the distribution in
586specific ways, ensuring that some parts of the data is more hot than others.
587Fio includes the following distribution models:
588.RS
589.TP
590.B random
591Uniform random distribution
592.TP
593.B zipf
594Zipf distribution
595.TP
596.B pareto
597Pareto distribution
598.TP
599.RE
600.P
601When using a zipf or pareto distribution, an input value is also needed to
602define the access pattern. For zipf, this is the zipf theta. For pareto,
603it's the pareto power. Fio includes a test program, genzipf, that can be
604used visualize what the given input values will yield in terms of hit rates.
605If you wanted to use zipf with a theta of 1.2, you would use
606random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
607fio will disable use of the random map.
608.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200609.B norandommap
610Normally \fBfio\fR will cover every block of the file when doing random I/O. If
611this parameter is given, a new offset will be chosen without looking at past
612I/O history. This parameter is mutually exclusive with \fBverify\fR.
613.TP
Jens Axboe744492c2011-08-08 09:47:13 +0200614.BI softrandommap \fR=\fPbool
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200615See \fBnorandommap\fR. If fio runs with the random block map enabled and it
616fails to allocate the map, if this option is set it will continue without a
617random block map. As coverage will not be as complete as with random maps, this
618option is disabled by default.
619.TP
Jens Axboee8b19612012-12-05 10:28:08 +0100620.BI random_generator \fR=\fPstr
621Fio supports the following engines for generating IO offsets for random IO:
622.RS
623.TP
624.B tausworthe
625Strong 2^88 cycle random number generator
626.TP
627.B lfsr
628Linear feedback shift register generator
629.TP
630.RE
631.P
632Tausworthe is a strong random number generator, but it requires tracking on the
633side if we want to ensure that blocks are only read or written once. LFSR
634guarantees that we never generate the same offset twice, and it's also less
635computationally expensive. It's not a true random generator, however, though
636for IO purposes it's typically good enough. LFSR only works with single block
637sizes, not with workloads that use multiple block sizes. If used with such a
638workload, fio may read or write some blocks multiple times.
639.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200640.BI nice \fR=\fPint
641Run job with given nice value. See \fInice\fR\|(2).
642.TP
643.BI prio \fR=\fPint
644Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
645\fIionice\fR\|(1).
646.TP
647.BI prioclass \fR=\fPint
648Set I/O priority class. See \fIionice\fR\|(1).
649.TP
650.BI thinktime \fR=\fPint
651Stall job for given number of microseconds between issuing I/Os.
652.TP
653.BI thinktime_spin \fR=\fPint
654Pretend to spend CPU time for given number of microseconds, sleeping the rest
655of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
656.TP
657.BI thinktime_blocks \fR=\fPint
658Number of blocks to issue before waiting \fBthinktime\fR microseconds.
659Default: 1.
660.TP
661.BI rate \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200662Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
663rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
664or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
665limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
666can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
667limit writes (to 500KB/sec), the latter will only limit reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200668.TP
669.BI ratemin \fR=\fPint
670Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
Jens Axboec35dd7a2009-06-10 08:39:16 +0200671Failing to meet this requirement will cause the job to exit. The same format
672as \fBrate\fR is used for read vs write separation.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200673.TP
674.BI rate_iops \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200675Cap the bandwidth to this number of IOPS. Basically the same as rate, just
676specified independently of bandwidth. The same format as \fBrate\fR is used for
677read vs write seperation. If \fBblocksize\fR is a range, the smallest block
678size is used as the metric.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200679.TP
680.BI rate_iops_min \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200681If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
682is used for read vs write seperation.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200683.TP
684.BI ratecycle \fR=\fPint
685Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
686milliseconds. Default: 1000ms.
687.TP
Jens Axboe15501532012-10-24 16:37:45 +0200688.BI max_latency \fR=\fPint
689If set, fio will exit the job if it exceeds this maximum latency. It will exit
690with an ETIME error.
691.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200692.BI cpumask \fR=\fPint
693Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
694may run on. See \fBsched_setaffinity\fR\|(2).
695.TP
696.BI cpus_allowed \fR=\fPstr
697Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
698.TP
Yufei Rend0b937e2012-10-19 23:11:52 -0400699.BI numa_cpu_nodes \fR=\fPstr
700Set this job running on spcified NUMA nodes' CPUs. The arguments allow
701comma delimited list of cpu numbers, A-B ranges, or 'all'.
702.TP
703.BI numa_mem_policy \fR=\fPstr
704Set this job's memory policy and corresponding NUMA nodes. Format of
705the argements:
706.RS
707.TP
708.B <mode>[:<nodelist>]
709.TP
710.B mode
711is one of the following memory policy:
712.TP
713.B default, prefer, bind, interleave, local
714.TP
715.RE
716For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
717needed to be specified. For \fBprefer\fR, only one node is
718allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
719comma delimited list of numbers, A-B ranges, or 'all'.
720.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200721.BI startdelay \fR=\fPint
722Delay start of job for the specified number of seconds.
723.TP
724.BI runtime \fR=\fPint
725Terminate processing after the specified number of seconds.
726.TP
727.B time_based
728If given, run for the specified \fBruntime\fR duration even if the files are
729completely read or written. The same workload will be repeated as many times
730as \fBruntime\fR allows.
731.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100732.BI ramp_time \fR=\fPint
733If set, fio will run the specified workload for this amount of time before
734logging any performance numbers. Useful for letting performance settle before
735logging results, thus minimizing the runtime required for stable results. Note
Jens Axboec35dd7a2009-06-10 08:39:16 +0200736that the \fBramp_time\fR is considered lead in time for a job, thus it will
737increase the total runtime if a special timeout or runtime is specified.
Jens Axboe901bb992009-03-14 20:17:36 +0100738.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200739.BI invalidate \fR=\fPbool
740Invalidate buffer-cache for the file prior to starting I/O. Default: true.
741.TP
742.BI sync \fR=\fPbool
743Use synchronous I/O for buffered writes. For the majority of I/O engines,
Aaron Carrolld1429b52007-09-18 08:10:57 +0200744this means using O_SYNC. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200745.TP
746.BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
747Allocation method for I/O unit buffer. Allowed values are:
748.RS
749.RS
750.TP
751.B malloc
752Allocate memory with \fImalloc\fR\|(3).
753.TP
754.B shm
755Use shared memory buffers allocated through \fIshmget\fR\|(2).
756.TP
757.B shmhuge
758Same as \fBshm\fR, but use huge pages as backing.
759.TP
760.B mmap
761Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
762is given after the option in the format `:\fIfile\fR'.
763.TP
764.B mmaphuge
765Same as \fBmmap\fR, but use huge files as backing.
766.RE
767.P
768The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
769job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
770the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
Jens Axboe2e266ba2009-09-14 08:56:53 +0200771have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
772huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
773and the documentation for that. Normally you just need to echo an appropriate
774number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
775use.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200776.RE
777.TP
Jens Axboed3923652011-08-03 12:38:39 +0200778.BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
Jens Axboed529ee12009-07-01 10:33:03 +0200779This indiciates the memory alignment of the IO memory buffers. Note that the
780given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
781the alignment of the following buffers are given by the \fBbs\fR used. In
782other words, if using a \fBbs\fR that is a multiple of the page sized in the
783system, all buffers will be aligned to this value. If using a \fBbs\fR that
784is not page aligned, the alignment of subsequent IO memory buffers is the
785sum of the \fBiomem_align\fR and \fBbs\fR used.
786.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100787.BI hugepage\-size \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200788Defines the size of a huge page. Must be at least equal to the system setting.
Jens Axboeb22989b2009-07-17 22:29:23 +0200789Should be a multiple of 1MB. Default: 4MB.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200790.TP
791.B exitall
792Terminate all jobs when one finishes. Default: wait for each job to finish.
793.TP
794.BI bwavgtime \fR=\fPint
795Average bandwidth calculations over the given time in milliseconds. Default:
796500ms.
797.TP
Jens Axboec8eeb9d2011-10-05 14:02:22 +0200798.BI iopsavgtime \fR=\fPint
799Average IOPS calculations over the given time in milliseconds. Default:
800500ms.
801.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200802.BI create_serialize \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200803If true, serialize file creation for the jobs. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200804.TP
805.BI create_fsync \fR=\fPbool
806\fIfsync\fR\|(2) data file after creation. Default: true.
807.TP
Jens Axboe6b7f6852009-03-09 14:22:56 +0100808.BI create_on_open \fR=\fPbool
809If true, the files are not created until they are opened for IO by the job.
810.TP
Jens Axboe25460cf2012-05-02 13:58:02 +0200811.BI create_only \fR=\fPbool
812If true, fio will only run the setup phase of the job. If files need to be
813laid out or updated on disk, only that will be done. The actual job contents
814are not executed.
815.TP
Jens Axboee9f48472009-06-03 12:14:08 +0200816.BI pre_read \fR=\fPbool
817If this is given, files will be pre-read into memory before starting the given
818IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
Jens Axboe9c0d2242009-07-01 12:26:28 +0200819pointless to pre-read and then drop the cache. This will only work for IO
820engines that are seekable, since they allow you to read the same data
821multiple times. Thus it will not work on eg network or splice IO.
Jens Axboee9f48472009-06-03 12:14:08 +0200822.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200823.BI unlink \fR=\fPbool
824Unlink job files when done. Default: false.
825.TP
826.BI loops \fR=\fPint
827Specifies the number of iterations (runs of the same workload) of this job.
828Default: 1.
829.TP
830.BI do_verify \fR=\fPbool
831Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
832Default: true.
833.TP
834.BI verify \fR=\fPstr
835Method of verifying file contents after each iteration of the job. Allowed
836values are:
837.RS
838.RS
839.TP
Jens Axboeb892dc02009-09-05 20:37:35 +0200840.B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
Jens Axboe0539d752010-06-21 15:22:56 +0200841Store appropriate checksum in the header of each block. crc32c-intel is
842hardware accelerated SSE4.2 driven, falls back to regular crc32c if
843not supported by the system.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200844.TP
845.B meta
846Write extra information about each I/O (timestamp, block number, etc.). The
Jens Axboe996093b2010-06-24 08:37:13 +0200847block number is verified. See \fBverify_pattern\fR as well.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200848.TP
849.B null
850Pretend to verify. Used for testing internals.
851.RE
Jens Axboeb892dc02009-09-05 20:37:35 +0200852
853This option can be used for repeated burn-in tests of a system to make sure
854that the written data is also correctly read back. If the data direction given
855is a read or random read, fio will assume that it should verify a previously
856written file. If the data direction includes any form of write, the verify will
857be of the newly written data.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200858.RE
859.TP
860.BI verify_sort \fR=\fPbool
861If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
862read them back in a sorted manner. Default: true.
863.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100864.BI verify_offset \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200865Swap the verification header with data somewhere else in the block before
Aaron Carrolld1429b52007-09-18 08:10:57 +0200866writing. It is swapped back before verifying.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200867.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100868.BI verify_interval \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200869Write the verification header for this number of bytes, which should divide
870\fBblocksize\fR. Default: \fBblocksize\fR.
871.TP
Jens Axboe996093b2010-06-24 08:37:13 +0200872.BI verify_pattern \fR=\fPstr
873If set, fio will fill the io buffers with this pattern. Fio defaults to filling
874with totally random bytes, but sometimes it's interesting to fill with a known
875pattern for io verification purposes. Depending on the width of the pattern,
876fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
877decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
878has to be a hex number that starts with either "0x" or "0X". Use with
879\fBverify\fP=meta.
880.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200881.BI verify_fatal \fR=\fPbool
882If true, exit the job on the first observed verification failure. Default:
883false.
884.TP
Jens Axboeb463e932011-01-12 09:03:23 +0100885.BI verify_dump \fR=\fPbool
886If set, dump the contents of both the original data block and the data block we
887read off disk to files. This allows later analysis to inspect just what kind of
Jens Axboeef71e312011-10-25 22:43:36 +0200888data corruption occurred. Off by default.
Jens Axboeb463e932011-01-12 09:03:23 +0100889.TP
Jens Axboee8462bd2009-07-06 12:59:04 +0200890.BI verify_async \fR=\fPint
891Fio will normally verify IO inline from the submitting thread. This option
892takes an integer describing how many async offload threads to create for IO
893verification instead, causing fio to offload the duty of verifying IO contents
Jens Axboec85c3242009-07-06 14:12:57 +0200894to one or more separate threads. If using this offload option, even sync IO
895engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
896allows them to have IO in flight while verifies are running.
Jens Axboee8462bd2009-07-06 12:59:04 +0200897.TP
898.BI verify_async_cpus \fR=\fPstr
899Tell fio to set the given CPU affinity on the async IO verification threads.
900See \fBcpus_allowed\fP for the format used.
901.TP
Jens Axboe6f874182010-06-21 12:53:26 +0200902.BI verify_backlog \fR=\fPint
903Fio will normally verify the written contents of a job that utilizes verify
904once that job has completed. In other words, everything is written then
905everything is read back and verified. You may want to verify continually
906instead for a variety of reasons. Fio stores the meta data associated with an
907IO block in memory, so for large verify workloads, quite a bit of memory would
David Nellans092f7072010-10-26 08:08:42 -0600908be used up holding this meta data. If this option is enabled, fio will write
909only N blocks before verifying these blocks.
Jens Axboe6f874182010-06-21 12:53:26 +0200910.TP
911.BI verify_backlog_batch \fR=\fPint
912Control how many blocks fio will verify if verify_backlog is set. If not set,
913will default to the value of \fBverify_backlog\fR (meaning the entire queue is
David Nellans092f7072010-10-26 08:08:42 -0600914read back and verified). If \fBverify_backlog_batch\fR is less than
915\fBverify_backlog\fR then not all blocks will be verified, if
916\fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
917will be verified more than once.
Jens Axboe6f874182010-06-21 12:53:26 +0200918.TP
Jens Axboed3923652011-08-03 12:38:39 +0200919.B stonewall "\fR,\fP wait_for_previous"
Martin Steigerwald5982a922011-06-27 16:07:24 +0200920Wait for preceding jobs in the job file to exit before starting this one.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200921\fBstonewall\fR implies \fBnew_group\fR.
922.TP
923.B new_group
924Start a new reporting group. If not given, all jobs in a file will be part
925of the same reporting group, unless separated by a stonewall.
926.TP
927.BI numjobs \fR=\fPint
928Number of clones (processes/threads performing the same workload) of this job.
929Default: 1.
930.TP
931.B group_reporting
932If set, display per-group reports instead of per-job when \fBnumjobs\fR is
933specified.
934.TP
935.B thread
936Use threads created with \fBpthread_create\fR\|(3) instead of processes created
937with \fBfork\fR\|(2).
938.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100939.BI zonesize \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200940Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
941.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100942.BI zoneskip \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200943Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200944read.
945.TP
946.BI write_iolog \fR=\fPstr
Stefan Hajnoczi5b42a482011-01-08 20:28:41 +0100947Write the issued I/O patterns to the specified file. Specify a separate file
948for each job, otherwise the iologs will be interspersed and the file may be
949corrupt.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200950.TP
951.BI read_iolog \fR=\fPstr
952Replay the I/O patterns contained in the specified file generated by
953\fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
954.TP
David Nellans64bbb862010-08-24 22:13:30 +0200955.BI replay_no_stall \fR=\fPint
956While replaying I/O patterns using \fBread_iolog\fR the default behavior
957attempts to respect timing information between I/Os. Enabling
958\fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
959still respecting ordering.
960.TP
David Nellansd1c46c02010-08-31 21:20:47 +0200961.BI replay_redirect \fR=\fPstr
962While replaying I/O patterns using \fBread_iolog\fR the default behavior
963is to replay the IOPS onto the major/minor device that each IOP was recorded
964from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
965single specified device regardless of the device it was recorded from.
966.TP
Steven Noonan836bad52011-09-14 09:21:33 +0200967.BI write_bw_log \fR=\fPstr
Jens Axboe901bb992009-03-14 20:17:36 +0100968If given, write a bandwidth log of the jobs in this job file. Can be used to
969store data of the bandwidth of the jobs in their lifetime. The included
970fio_generate_plots script uses gnuplot to turn these text files into nice
971graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
972option, the postfix is _bw.log.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200973.TP
Steven Noonan836bad52011-09-14 09:21:33 +0200974.BI write_lat_log \fR=\fPstr
Jens Axboe901bb992009-03-14 20:17:36 +0100975Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
976filename is given with this option, the default filename of "jobname_type.log"
977is used. Even if the filename is given, fio will still append the type of log.
978.TP
Jens Axboec8eeb9d2011-10-05 14:02:22 +0200979.BI write_iops_log \fR=\fPstr
980Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
981option, the default filename of "jobname_type.log" is used. Even if the
982filename is given, fio will still append the type of log.
983.TP
Jens Axboeb8bc8cb2011-12-01 09:04:31 +0100984.BI log_avg_msec \fR=\fPint
985By default, fio will log an entry in the iops, latency, or bw log for every
986IO that completes. When writing to the disk log, that can quickly grow to a
987very large size. Setting this option makes fio average the each log entry
988over the specified period of time, reducing the resolution of the log.
989Defaults to 0.
990.TP
Steven Noonan836bad52011-09-14 09:21:33 +0200991.BI disable_lat \fR=\fPbool
Jens Axboe02af0982010-06-24 09:59:34 +0200992Disable measurements of total latency numbers. Useful only for cutting
Jens Axboe901bb992009-03-14 20:17:36 +0100993back the number of calls to gettimeofday, as that does impact performance at
994really high IOPS rates. Note that to really get rid of a large amount of these
995calls, this option must be used with disable_slat and disable_bw as well.
996.TP
Steven Noonan836bad52011-09-14 09:21:33 +0200997.BI disable_clat \fR=\fPbool
Steven Noonanc95f9da2011-06-22 09:47:09 +0200998Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
Jens Axboe02af0982010-06-24 09:59:34 +0200999.TP
Steven Noonan836bad52011-09-14 09:21:33 +02001000.BI disable_slat \fR=\fPbool
Jens Axboe02af0982010-06-24 09:59:34 +02001001Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
Jens Axboe901bb992009-03-14 20:17:36 +01001002.TP
Steven Noonan836bad52011-09-14 09:21:33 +02001003.BI disable_bw_measurement \fR=\fPbool
Jens Axboe02af0982010-06-24 09:59:34 +02001004Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001005.TP
Jens Axboef7fa2652009-03-09 14:20:20 +01001006.BI lockmem \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001007Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1008simulate a smaller amount of memory.
1009.TP
1010.BI exec_prerun \fR=\fPstr
1011Before running the job, execute the specified command with \fBsystem\fR\|(3).
1012.TP
1013.BI exec_postrun \fR=\fPstr
1014Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1015.TP
1016.BI ioscheduler \fR=\fPstr
1017Attempt to switch the device hosting the file to the specified I/O scheduler.
1018.TP
1019.BI cpuload \fR=\fPint
1020If the job is a CPU cycle-eater, attempt to use the specified percentage of
1021CPU cycles.
1022.TP
1023.BI cpuchunks \fR=\fPint
1024If the job is a CPU cycle-eater, split the load into cycles of the
1025given time in milliseconds.
1026.TP
1027.BI disk_util \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +02001028Generate disk utilization statistics if the platform supports it. Default: true.
Jens Axboe901bb992009-03-14 20:17:36 +01001029.TP
Jens Axboe23893642012-12-17 14:44:08 +01001030.BI clocksource \fR=\fPstr
1031Use the given clocksource as the base of timing. The supported options are:
1032.RS
1033.TP
1034.B gettimeofday
1035gettimeofday(2)
1036.TP
1037.B clock_gettime
1038clock_gettime(2)
1039.TP
1040.B cpu
1041Internal CPU clock source
1042.TP
1043.RE
1044.P
1045\fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1046(and fio is heavy on time calls). Fio will automatically use this clocksource
1047if it's supported and considered reliable on the system it is running on,
1048unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1049means supporting TSC Invariant.
1050.TP
Jens Axboe901bb992009-03-14 20:17:36 +01001051.BI gtod_reduce \fR=\fPbool
1052Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
1053disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1054gettimeofday() call count. With this option enabled, we only do about 0.4% of
1055the gtod() calls we would have done if all time keeping was enabled.
1056.TP
1057.BI gtod_cpu \fR=\fPint
1058Sometimes it's cheaper to dedicate a single thread of execution to just getting
1059the current time. Fio (and databases, for instance) are very intensive on
1060gettimeofday() calls. With this option, you can set one CPU aside for doing
1061nothing but logging current time to a shared memory location. Then the other
1062threads/processes that run IO workloads need only copy that segment, instead of
1063entering the kernel with a gettimeofday() call. The CPU set aside for doing
1064these time calls will be excluded from other uses. Fio will manually clear it
1065from the CPU mask of other jobs.
Radha Ramachandranf2bba182009-06-15 08:40:16 +02001066.TP
Dmitry Monakhov8b28bd42012-09-23 15:46:09 +04001067.BI ignore_error \fR=\fPstr
1068Sometimes you want to ignore some errors during test in that case you can specify
1069error list for each error type.
1070.br
1071ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1072.br
1073errors for given error type is separated with ':'.
1074Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1075.br
1076Example: ignore_error=EAGAIN,ENOSPC:122 .
1077.br
1078This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1079.TP
1080.BI error_dump \fR=\fPbool
1081If set dump every error even if it is non fatal, true by default. If disabled
1082only fatal error will be dumped
1083.TP
Jens Axboea696fa22009-12-04 10:05:02 +01001084.BI cgroup \fR=\fPstr
1085Add job to this control group. If it doesn't exist, it will be created.
Jens Axboe6adb38a2009-12-07 08:01:26 +01001086The system must have a mounted cgroup blkio mount point for this to work. If
1087your system doesn't have it mounted, you can do so with:
1088
Martin Steigerwald5982a922011-06-27 16:07:24 +02001089# mount \-t cgroup \-o blkio none /cgroup
Jens Axboea696fa22009-12-04 10:05:02 +01001090.TP
1091.BI cgroup_weight \fR=\fPint
1092Set the weight of the cgroup to this value. See the documentation that comes
1093with the kernel, allowed values are in the range of 100..1000.
Jens Axboee0b0d892009-12-08 10:10:14 +01001094.TP
Vivek Goyal7de87092010-03-31 22:55:15 +02001095.BI cgroup_nodelete \fR=\fPbool
1096Normally fio will delete the cgroups it has created after the job completion.
1097To override this behavior and to leave cgroups around after the job completion,
1098set cgroup_nodelete=1. This can be useful if one wants to inspect various
1099cgroup files after job completion. Default: false
1100.TP
Jens Axboee0b0d892009-12-08 10:10:14 +01001101.BI uid \fR=\fPint
1102Instead of running as the invoking user, set the user ID to this value before
1103the thread/process does any work.
1104.TP
1105.BI gid \fR=\fPint
1106Set group ID, see \fBuid\fR.
Yu-ju Hong83349192011-08-13 00:53:44 +02001107.TP
Dan Ehrenberg9e684a42012-02-20 11:05:14 +01001108.BI flow_id \fR=\fPint
1109The ID of the flow. If not specified, it defaults to being a global flow. See
1110\fBflow\fR.
1111.TP
1112.BI flow \fR=\fPint
1113Weight in token-based flow control. If this value is used, then there is a
1114\fBflow counter\fR which is used to regulate the proportion of activity between
1115two or more jobs. fio attempts to keep this flow counter near zero. The
1116\fBflow\fR parameter stands for how much should be added or subtracted to the
1117flow counter on each iteration of the main I/O loop. That is, if one job has
1118\fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
11191:8 ratio in how much one runs vs the other.
1120.TP
1121.BI flow_watermark \fR=\fPint
1122The maximum value that the absolute value of the flow counter is allowed to
1123reach before the job must wait for a lower value of the counter.
1124.TP
1125.BI flow_sleep \fR=\fPint
1126The period of time, in microseconds, to wait after the flow watermark has been
1127exceeded before retrying operations
1128.TP
Yu-ju Hong83349192011-08-13 00:53:44 +02001129.BI clat_percentiles \fR=\fPbool
1130Enable the reporting of percentiles of completion latencies.
1131.TP
1132.BI percentile_list \fR=\fPfloat_list
1133Overwrite the default list of percentiles for completion
1134latencies. Each number is a floating number in the range (0,100], and
1135the maximum length of the list is 20. Use ':' to separate the
Martin Steigerwald3eb07282011-10-05 11:41:54 +02001136numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
Yu-ju Hong83349192011-08-13 00:53:44 +02001137report the values of completion latency below which 99.5% and 99.9% of
1138the observed latencies fell, respectively.
Steven Langde890a12011-11-09 14:03:34 +01001139.SS "Ioengine Parameters List"
1140Some parameters are only valid when a specific ioengine is in use. These are
1141used identically to normal parameters, with the caveat that when used on the
1142command line, the must come after the ioengine that defines them is selected.
1143.TP
1144.BI (libaio)userspace_reap
1145Normally, with the libaio engine in use, fio will use
1146the io_getevents system call to reap newly returned events.
1147With this flag turned on, the AIO ring will be read directly
1148from user-space to reap events. The reaping mode is only
1149enabled when polling for a minimum of 0 events (eg when
1150iodepth_batch_complete=0).
1151.TP
1152.BI (net,netsplice)hostname \fR=\fPstr
1153The host name or IP address to use for TCP or UDP based IO.
1154If the job is a TCP listener or UDP reader, the hostname is not
1155used and must be omitted.
1156.TP
1157.BI (net,netsplice)port \fR=\fPint
1158The TCP or UDP port to bind to or connect to.
1159.TP
1160.BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1161The network protocol to use. Accepted values are:
1162.RS
1163.RS
1164.TP
1165.B tcp
1166Transmission control protocol
1167.TP
1168.B udp
Bruce Cranf5cc3d02012-10-10 08:17:44 -06001169User datagram protocol
Steven Langde890a12011-11-09 14:03:34 +01001170.TP
1171.B unix
1172UNIX domain socket
1173.RE
1174.P
1175When the protocol is TCP or UDP, the port must also be given,
1176as well as the hostname if the job is a TCP listener or UDP
1177reader. For unix sockets, the normal filename option should be
1178used and the port is invalid.
1179.RE
1180.TP
1181.BI (net,netsplice)listen
1182For TCP network connections, tell fio to listen for incoming
1183connections rather than initiating an outgoing connection. The
1184hostname must be omitted if this option is used.
Dmitry Monakhovd54fce82012-09-20 15:37:17 +04001185.TP
Jens Axboe7aeb1e92012-12-06 20:53:57 +01001186.BI (net, pingpong) \fR=\fPbool
1187Normal a network writer will just continue writing data, and a network reader
1188will just consume packages. If pingpong=1 is set, a writer will send its normal
1189payload to the reader, then wait for the reader to send the same payload back.
1190This allows fio to measure network latencies. The submission and completion
1191latencies then measure local time spent sending or receiving, and the
1192completion latency measures how long it took for the other end to receive and
1193send back.
1194.TP
Dmitry Monakhovd54fce82012-09-20 15:37:17 +04001195.BI (e4defrag,donorname) \fR=\fPstr
1196File will be used as a block donor (swap extents between files)
1197.TP
1198.BI (e4defrag,inplace) \fR=\fPint
1199Configure donor file block allocation strategy
1200.RS
1201.BI 0(default) :
1202Preallocate donor's file on init
1203.TP
1204.BI 1:
1205allocate space immidietly inside defragment event, and free right after event
1206.RE
1207.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001208.SH OUTPUT
Aaron Carrolld1429b52007-09-18 08:10:57 +02001209While running, \fBfio\fR will display the status of the created jobs. For
1210example:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001211.RS
Aaron Carrolld1429b52007-09-18 08:10:57 +02001212.P
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001213Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1214.RE
1215.P
Aaron Carrolld1429b52007-09-18 08:10:57 +02001216The characters in the first set of brackets denote the current status of each
1217threads. The possible values are:
1218.P
1219.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001220.RS
1221.TP
1222.B P
1223Setup but not started.
1224.TP
1225.B C
1226Thread created.
1227.TP
1228.B I
1229Initialized, waiting.
1230.TP
1231.B R
1232Running, doing sequential reads.
1233.TP
1234.B r
1235Running, doing random reads.
1236.TP
1237.B W
1238Running, doing sequential writes.
1239.TP
1240.B w
1241Running, doing random writes.
1242.TP
1243.B M
1244Running, doing mixed sequential reads/writes.
1245.TP
1246.B m
1247Running, doing mixed random reads/writes.
1248.TP
1249.B F
1250Running, currently waiting for \fBfsync\fR\|(2).
1251.TP
1252.B V
1253Running, verifying written data.
1254.TP
1255.B E
1256Exited, not reaped by main thread.
1257.TP
1258.B \-
1259Exited, thread reaped.
1260.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +02001261.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001262.P
1263The second set of brackets shows the estimated completion percentage of
1264the current group. The third set shows the read and write I/O rate,
1265respectively. Finally, the estimated run time of the job is displayed.
1266.P
1267When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1268for each thread, each group of threads, and each disk, in that order.
1269.P
1270Per-thread statistics first show the threads client number, group-id, and
1271error code. The remaining figures are as follows:
1272.RS
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001273.TP
1274.B io
1275Number of megabytes of I/O performed.
1276.TP
1277.B bw
1278Average data rate (bandwidth).
1279.TP
1280.B runt
1281Threads run time.
1282.TP
1283.B slat
1284Submission latency minimum, maximum, average and standard deviation. This is
1285the time it took to submit the I/O.
1286.TP
1287.B clat
1288Completion latency minimum, maximum, average and standard deviation. This
1289is the time between submission and completion.
1290.TP
1291.B bw
1292Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1293and standard deviation.
1294.TP
1295.B cpu
1296CPU usage statistics. Includes user and system time, number of context switches
1297this thread went through and number of major and minor page faults.
1298.TP
1299.B IO depths
1300Distribution of I/O depths. Each depth includes everything less than (or equal)
1301to it, but greater than the previous depth.
1302.TP
1303.B IO issued
1304Number of read/write requests issued, and number of short read/write requests.
1305.TP
1306.B IO latencies
1307Distribution of I/O completion latencies. The numbers follow the same pattern
1308as \fBIO depths\fR.
1309.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001310.P
1311The group statistics show:
Aaron Carrolld1429b52007-09-18 08:10:57 +02001312.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001313.RS
1314.TP
1315.B io
1316Number of megabytes I/O performed.
1317.TP
1318.B aggrb
1319Aggregate bandwidth of threads in the group.
1320.TP
1321.B minb
1322Minimum average bandwidth a thread saw.
1323.TP
1324.B maxb
1325Maximum average bandwidth a thread saw.
1326.TP
1327.B mint
Aaron Carrolld1429b52007-09-18 08:10:57 +02001328Shortest runtime of threads in the group.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001329.TP
1330.B maxt
1331Longest runtime of threads in the group.
1332.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +02001333.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001334.P
1335Finally, disk statistics are printed with reads first:
Aaron Carrolld1429b52007-09-18 08:10:57 +02001336.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001337.RS
1338.TP
1339.B ios
1340Number of I/Os performed by all groups.
1341.TP
1342.B merge
1343Number of merges in the I/O scheduler.
1344.TP
1345.B ticks
1346Number of ticks we kept the disk busy.
1347.TP
1348.B io_queue
1349Total time spent in the disk queue.
1350.TP
1351.B util
1352Disk utilization.
1353.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +02001354.PD
Jens Axboe8423bd12012-04-12 09:18:38 +02001355.P
1356It is also possible to get fio to dump the current output while it is
1357running, without terminating the job. To do that, send fio the \fBUSR1\fR
1358signal.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001359.SH TERSE OUTPUT
1360If the \fB\-\-minimal\fR option is given, the results will be printed in a
David Nellans562c2d22010-09-23 08:38:17 +02001361semicolon-delimited format suitable for scripted use - a job description
1362(if provided) follows on a new line. Note that the first
Jens Axboe525c2bf2010-06-30 15:22:21 +02001363number in the line is the version number. If the output has to be changed
1364for some reason, this number will be incremented by 1 to signify that
1365change. The fields are:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001366.P
1367.RS
Jens Axboe5e726d02011-10-14 08:08:10 +02001368.B terse version, fio version, jobname, groupid, error
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001369.P
1370Read status:
1371.RS
Jens Axboe312b4af2011-10-13 13:11:42 +02001372.B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001373.P
1374Submission latency:
1375.RS
1376.B min, max, mean, standard deviation
1377.RE
1378Completion latency:
1379.RS
1380.B min, max, mean, standard deviation
1381.RE
Jens Axboe1db92cb2011-10-13 13:43:36 +02001382Completion latency percentiles (20 fields):
1383.RS
1384.B Xth percentile=usec
1385.RE
Jens Axboe525c2bf2010-06-30 15:22:21 +02001386Total latency:
1387.RS
1388.B min, max, mean, standard deviation
1389.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001390Bandwidth:
1391.RS
1392.B min, max, aggregate percentage of total, mean, standard deviation
1393.RE
1394.RE
1395.P
1396Write status:
1397.RS
Jens Axboe312b4af2011-10-13 13:11:42 +02001398.B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001399.P
1400Submission latency:
1401.RS
1402.B min, max, mean, standard deviation
1403.RE
1404Completion latency:
1405.RS
1406.B min, max, mean, standard deviation
1407.RE
Jens Axboe1db92cb2011-10-13 13:43:36 +02001408Completion latency percentiles (20 fields):
1409.RS
1410.B Xth percentile=usec
1411.RE
Jens Axboe525c2bf2010-06-30 15:22:21 +02001412Total latency:
1413.RS
1414.B min, max, mean, standard deviation
1415.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001416Bandwidth:
1417.RS
1418.B min, max, aggregate percentage of total, mean, standard deviation
1419.RE
1420.RE
1421.P
Aaron Carrolld1429b52007-09-18 08:10:57 +02001422CPU usage:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001423.RS
Carl Henrik Lundebd2626f2008-06-12 09:17:46 +02001424.B user, system, context switches, major page faults, minor page faults
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001425.RE
1426.P
1427IO depth distribution:
1428.RS
1429.B <=1, 2, 4, 8, 16, 32, >=64
1430.RE
1431.P
David Nellans562c2d22010-09-23 08:38:17 +02001432IO latency distribution:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001433.RS
David Nellans562c2d22010-09-23 08:38:17 +02001434Microseconds:
1435.RS
1436.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1437.RE
1438Milliseconds:
1439.RS
1440.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1441.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001442.RE
1443.P
Jens Axboef2f788d2011-10-13 14:03:52 +02001444Disk utilization (1 for each disk used):
1445.RS
1446.B name, read ios, write ios, read merges, write merges, read ticks, write ticks, read in-queue time, write in-queue time, disk utilization percentage
1447.RE
1448.P
Martin Steigerwald5982a922011-06-27 16:07:24 +02001449Error Info (dependent on continue_on_error, default off):
David Nellans562c2d22010-09-23 08:38:17 +02001450.RS
1451.B total # errors, first error code
1452.RE
1453.P
1454.B text description (if provided in config - appears on newline)
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001455.RE
Jens Axboe49da1242011-10-13 20:17:02 +02001456.SH CLIENT / SERVER
1457Normally you would run fio as a stand-alone application on the machine
1458where the IO workload should be generated. However, it is also possible to
1459run the frontend and backend of fio separately. This makes it possible to
1460have a fio server running on the machine(s) where the IO workload should
1461be running, while controlling it from another machine.
1462
1463To start the server, you would do:
1464
1465\fBfio \-\-server=args\fR
1466
1467on that machine, where args defines what fio listens to. The arguments
Jens Axboe811826b2011-10-24 09:11:50 +02001468are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
Martin Steigerwald20c67f12012-05-07 17:06:26 +02001469for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1470socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
Jens Axboe811826b2011-10-24 09:11:50 +02001471listen to (only valid for TCP/IP, not a local socket). Some examples:
Jens Axboe49da1242011-10-13 20:17:02 +02001472
Martin Steigerwalde01e9742012-05-07 17:06:54 +020014731) fio \-\-server
Jens Axboe49da1242011-10-13 20:17:02 +02001474
1475 Start a fio server, listening on all interfaces on the default port (8765).
1476
Martin Steigerwalde01e9742012-05-07 17:06:54 +020014772) fio \-\-server=ip:hostname,4444
Jens Axboe49da1242011-10-13 20:17:02 +02001478
1479 Start a fio server, listening on IP belonging to hostname and on port 4444.
1480
Martin Steigerwalde01e9742012-05-07 17:06:54 +020014813) fio \-\-server=ip6:::1,4444
Jens Axboe811826b2011-10-24 09:11:50 +02001482
1483 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1484
Martin Steigerwalde01e9742012-05-07 17:06:54 +020014854) fio \-\-server=,4444
Jens Axboe49da1242011-10-13 20:17:02 +02001486
1487 Start a fio server, listening on all interfaces on port 4444.
1488
Martin Steigerwalde01e9742012-05-07 17:06:54 +020014895) fio \-\-server=1.2.3.4
Jens Axboe49da1242011-10-13 20:17:02 +02001490
1491 Start a fio server, listening on IP 1.2.3.4 on the default port.
1492
Martin Steigerwalde01e9742012-05-07 17:06:54 +020014936) fio \-\-server=sock:/tmp/fio.sock
Jens Axboe49da1242011-10-13 20:17:02 +02001494
1495 Start a fio server, listening on the local socket /tmp/fio.sock.
1496
1497When a server is running, you can connect to it from a client. The client
1498is run with:
1499
Martin Steigerwalde01e9742012-05-07 17:06:54 +02001500fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
Jens Axboe49da1242011-10-13 20:17:02 +02001501
Martin Steigerwalde01e9742012-05-07 17:06:54 +02001502where \-\-local-args are arguments that are local to the client where it is
1503running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
Jens Axboe49da1242011-10-13 20:17:02 +02001504are sent to the server. The 'server' string follows the same format as it
1505does on the server side, to allow IP/hostname/socket and port strings.
1506You can connect to multiple clients as well, to do that you could run:
1507
Martin Steigerwalde01e9742012-05-07 17:06:54 +02001508fio \-\-client=server2 \-\-client=server2 <job file(s)>
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001509.SH AUTHORS
Jens Axboe49da1242011-10-13 20:17:02 +02001510
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001511.B fio
Jens Axboeaa58d252010-06-09 09:49:38 +02001512was written by Jens Axboe <jens.axboe@oracle.com>,
1513now Jens Axboe <jaxboe@fusionio.com>.
Aaron Carrolld1429b52007-09-18 08:10:57 +02001514.br
1515This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001516on documentation by Jens Axboe.
1517.SH "REPORTING BUGS"
Jens Axboe482900c2009-06-02 12:15:51 +02001518Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
Aaron Carrolld1429b52007-09-18 08:10:57 +02001519See \fBREADME\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001520.SH "SEE ALSO"
Aaron Carrolld1429b52007-09-18 08:10:57 +02001521For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1522.br
1523Sample jobfiles are available in the \fBexamples\fR directory.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001524