blob: d372727b69e7311d095ced198c2855f66091ec3d [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
15.BI \-\-output \fR=\fPfilename
16Write output to \fIfilename\fR.
17.TP
18.BI \-\-timeout \fR=\fPtimeout
19Limit run time to \fItimeout\fR seconds.
20.TP
21.B \-\-latency\-log
22Generate per-job latency logs.
23.TP
24.B \-\-bandwidth\-log
25Generate per-job bandwidth logs.
26.TP
27.B \-\-minimal
Aaron Carrolld1429b52007-09-18 08:10:57 +020028Print statistics in a terse, semicolon-delimited format.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020029.TP
30.BI \-\-showcmd \fR=\fPjobfile
31Convert \fIjobfile\fR to a set of command-line options.
32.TP
33.B \-\-readonly
34Enable read-only safety checks.
35.TP
36.BI \-\-eta \fR=\fPwhen
37Specifies when real-time ETA estimate should be printed. \fIwhen\fR may
38be one of `always', `never' or `auto'.
39.TP
Aaron Carrollc0a5d352008-02-26 23:10:39 +010040.BI \-\-section \fR=\fPsec
41Only run section \fIsec\fR from job file.
42.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +020043.BI \-\-cmdhelp \fR=\fPcommand
44Print help information for \fIcommand\fR. May be `all' for all commands.
45.TP
Jens Axboe91837882008-02-05 12:02:07 +010046.BI \-\-debug \fR=\fPtype
47Enable verbose tracing of various fio actions. May be `all' for all types
Michael Prokopc6e13ea2009-08-12 17:24:41 +020048or individual types seperated by a comma (eg \-\-debug=io,file). `help' will
Jens Axboe91837882008-02-05 12:02:07 +010049list all available tracing options.
50.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +020051.B \-\-help
52Display usage information and exit.
53.TP
54.B \-\-version
55Display version information and exit.
56.SH "JOB FILE FORMAT"
57Job files are in `ini' format. They consist of one or more
58job definitions, which begin with a job name in square brackets and
59extend to the next job name. The job name can be any ASCII string
60except `global', which has a special meaning. Following the job name is
61a sequence of zero or more parameters, one per line, that define the
62behavior of the job. Any line starting with a `;' or `#' character is
Aaron Carrolld1429b52007-09-18 08:10:57 +020063considered a comment and ignored.
Aaron Carrolld9956b62007-11-16 12:12:45 +010064.P
65If \fIjobfile\fR is specified as `-', the job file will be read from
66standard input.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020067.SS "Global Section"
68The global section contains default parameters for jobs specified in the
69job file. A job is only affected by global sections residing above it,
70and there may be any number of global sections. Specific job definitions
71may override any parameter set in global sections.
72.SH "JOB PARAMETERS"
73.SS Types
74Some parameters may take arguments of a specific type. The types used are:
75.TP
76.I str
77String: a sequence of alphanumeric characters.
78.TP
79.I int
Aaron Carrolld60e92d2007-09-17 10:32:59 +020080SI integer: a whole number, possibly containing a suffix denoting the base unit
Jens Axboeb09da8f2009-07-17 23:16:17 +020081of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
82kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
83respectively. The suffix is not case sensitive. If prefixed with '0x', the
Jens Axboe57fc29f2010-06-23 22:24:07 +020084value is assumed to be base 16 (hexadecimal). A suffix may include a trailing
85'b', for instance 'kb' is identical to 'k'. You can specify a base 10 value
86by using 'KiB', 'MiB', 'GiB', etc. This is useful for disk drives where
87values are often given in base 10 values. Specifying '30GiB' will get you
8830*1000^3 bytes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020089.TP
90.I bool
91Boolean: a true or false value. `0' denotes false, `1' denotes true.
92.TP
93.I irange
94Integer range: a range of integers specified in the format
Aaron Carrolld1429b52007-09-18 08:10:57 +020095\fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
96\fIupper\fR may contain a suffix as described above. If an option allows two
97sets of ranges, they are separated with a `,' or `/' character. For example:
98`8\-8k/8M\-4G'.
Aaron Carrolld60e92d2007-09-17 10:32:59 +020099.SS "Parameter List"
100.TP
101.BI name \fR=\fPstr
Aaron Carrolld9956b62007-11-16 12:12:45 +0100102May be used to override the job name. On the command line, this parameter
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200103has the special purpose of signalling the start of a new job.
104.TP
105.BI description \fR=\fPstr
106Human-readable description of the job. It is printed when the job is run, but
107otherwise has no special purpose.
108.TP
109.BI directory \fR=\fPstr
110Prefix filenames with this directory. Used to place files in a location other
111than `./'.
112.TP
113.BI filename \fR=\fPstr
114.B fio
115normally makes up a file name based on the job name, thread number, and file
Aaron Carrolld1429b52007-09-18 08:10:57 +0200116number. If you want to share files between threads in a job or several jobs,
117specify a \fIfilename\fR for each of them to override the default. If the I/O
118engine used is `net', \fIfilename\fR is the host and port to connect to in the
119format \fIhost\fR/\fIport\fR. If the I/O engine is file-based, you can specify
120a number of files by separating the names with a `:' character. `\-' is a
121reserved name, meaning stdin or stdout, depending on the read/write direction
122set.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200123.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200124.BI lockfile \fR=\fPstr
125Fio defaults to not locking any files before it does IO to them. If a file or
126file descriptor is shared, fio can serialize IO to that file to make the end
127result consistent. This is usual for emulating real workloads that share files.
128The lock modes are:
129.RS
130.RS
131.TP
132.B none
133No locking. This is the default.
134.TP
135.B exclusive
136Only one thread or process may do IO at the time, excluding all others.
137.TP
138.B readwrite
139Read-write locking on the file. Many readers may access the file at the same
140time, but writes get exclusive access.
141.RE
142.P
143The option may be post-fixed with a lock batch number. If set, then each
144thread/process may do that amount of IOs to the file before giving up the lock.
145Since lock acquisition is expensive, batching the lock/unlocks will speed up IO.
146.RE
147.P
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200148.BI opendir \fR=\fPstr
149Recursively open any files below directory \fIstr\fR.
150.TP
151.BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
152Type of I/O pattern. Accepted values are:
153.RS
154.RS
155.TP
156.B read
Aaron Carrolld1429b52007-09-18 08:10:57 +0200157Sequential reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200158.TP
159.B write
Aaron Carrolld1429b52007-09-18 08:10:57 +0200160Sequential writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200161.TP
162.B randread
Aaron Carrolld1429b52007-09-18 08:10:57 +0200163Random reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200164.TP
165.B randwrite
Aaron Carrolld1429b52007-09-18 08:10:57 +0200166Random writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200167.TP
168.B rw
Aaron Carrolld1429b52007-09-18 08:10:57 +0200169Mixed sequential reads and writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200170.TP
171.B randrw
Aaron Carrolld1429b52007-09-18 08:10:57 +0200172Mixed random reads and writes.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200173.RE
174.P
Jens Axboe38dad622010-07-20 14:46:00 -0600175For mixed I/O, the default split is 50/50. For certain types of io the result
176may still be skewed a bit, since the speed may be different. It is possible to
177specify a number of IO's to do before getting a new offset, this is one by
178appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
179would look like \fBrw=randread:8\fR for passing in an offset modifier with a
180value of 8. See the \fBrw_sequencer\fR option.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200181.RE
182.TP
Jens Axboe38dad622010-07-20 14:46:00 -0600183.BI rw_sequencer \fR=\fPstr
184If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
185then this option controls how that number modifies the IO offset being
186generated. Accepted values are:
187.RS
188.RS
189.TP
190.B sequential
191Generate sequential offset
192.TP
193.B identical
194Generate the same offset
195.RE
196.P
197\fBsequential\fR is only useful for random IO, where fio would normally
198generate a new random offset for every IO. If you append eg 8 to randread, you
199would get a new random offset for every 8 IO's. The result would be a seek for
200only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
201that. As sequential IO is already sequential, setting \fBsequential\fR for that
202would not result in any differences. \fBidentical\fR behaves in a similar
203fashion, except it sends the same offset 8 number of times before generating a
204new offset.
205.RE
206.P
207.TP
Jens Axboe90fef2d2009-07-17 22:33:32 +0200208.BI kb_base \fR=\fPint
209The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
210manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
211reasons. Allow values are 1024 or 1000, with 1024 being the default.
212.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200213.BI randrepeat \fR=\fPbool
214Seed the random number generator in a predictable way so results are repeatable
Aaron Carrolld1429b52007-09-18 08:10:57 +0200215across runs. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200216.TP
Jens Axboe7bc8c2c2010-01-28 11:31:31 +0100217.BI fallocate \fR=\fPbool
218By default, fio will use fallocate() to advise the system of the size of the
219file we are going to write. This can be turned off with fallocate=0. May not
220be available on all supported platforms.
221.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200222.BI fadvise_hint \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200223Disable use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
224are likely to be issued. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200225.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100226.BI size \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200227Total size of I/O for this job. \fBfio\fR will run until this many bytes have
228been transfered, unless limited by other options (\fBruntime\fR, for instance).
Jens Axboed7c8be02010-11-25 08:21:39 +0100229Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
Jens Axboed6667262010-06-25 11:32:48 +0200230divided between the available files for the job. If not set, fio will use the
231full size of the given files or devices. If the the files do not exist, size
232must be given.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200233.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200234.BI fill_device \fR=\fPbool
235Sets size to something really large and waits for ENOSPC (no space left on
236device) as the terminating condition. Only makes sense with sequential write.
237For a read workload, the mount point will be filled first then IO started on
238the result.
239.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200240.BI filesize \fR=\fPirange
241Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
Aaron Carrolld1429b52007-09-18 08:10:57 +0200242for files at random within the given range, limited to \fBsize\fR in total (if
243that is given). If \fBfilesize\fR is not specified, each created file is the
244same size.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200245.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100246.BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200247Block size for I/O units. Default: 4k. Values for reads and writes can be
Jens Axboe656ebab2010-04-13 10:39:14 +0200248specified separately in the format \fIread\fR,\fIwrite\fR, either of
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200249which may be empty to leave that value at its default.
250.TP
Jens Axboe91837882008-02-05 12:02:07 +0100251.BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
Aaron Carrolld1429b52007-09-18 08:10:57 +0200252Specify a range of I/O block sizes. The issued I/O unit will always be a
253multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
Jens Axboe91837882008-02-05 12:02:07 +0100254to both reads and writes if only one range is given, but can be specified
Jens Axboe656ebab2010-04-13 10:39:14 +0200255separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
Jens Axboe91837882008-02-05 12:02:07 +0100256Also (see \fBblocksize\fR).
257.TP
258.BI bssplit \fR=\fPstr
259This option allows even finer grained control of the block sizes issued,
260not just even splits between them. With this option, you can weight various
261block sizes for exact control of the issued IO for a job that has mixed
262block sizes. The format of the option is bssplit=blocksize/percentage,
263optionally adding as many definitions as needed seperated by a colon.
264Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
Jens Axboec83cdd32009-04-24 14:23:59 +0200265blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
266splits to reads and writes. The format is identical to what the
267\fBbs\fR option accepts, the read and write parts are separated with a
268comma.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200269.TP
270.B blocksize_unaligned\fR,\fP bs_unaligned
Aaron Carrolld1429b52007-09-18 08:10:57 +0200271If set, any size in \fBblocksize_range\fR may be used. This typically won't
272work with direct I/O, as that normally requires sector alignment.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200273.TP
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100274.BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
Martin Steigerwald639ce0f2009-05-20 11:33:49 +0200275At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
276the minimum blocksize given. Minimum alignment is typically 512b
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100277for using direct IO, though it usually depends on the hardware block size.
278This option is mutually exclusive with using a random map for files, so it
279will turn off that option.
Jens Axboe43602662009-03-14 20:08:47 +0100280.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200281.B zero_buffers
282Initialise buffers with all zeros. Default: fill buffers with random data.
283.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100284.B refill_buffers
285If this option is given, fio will refill the IO buffers on every submit. The
286default is to only fill it at init time and reuse that data. Only makes sense
287if zero_buffers isn't specified, naturally. If data verification is enabled,
288refill_buffers is also automatically enabled.
289.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200290.BI nrfiles \fR=\fPint
291Number of files to use for this job. Default: 1.
292.TP
293.BI openfiles \fR=\fPint
294Number of files to keep open at the same time. Default: \fBnrfiles\fR.
295.TP
296.BI file_service_type \fR=\fPstr
297Defines how files to service are selected. The following types are defined:
298.RS
299.RS
300.TP
301.B random
302Choose a file at random
303.TP
304.B roundrobin
305Round robin over open files (default).
Jens Axboe6b7f6852009-03-09 14:22:56 +0100306.B sequential
307Do each file in the set sequentially.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200308.RE
309.P
310The number of I/Os to issue before switching a new file can be specified by
311appending `:\fIint\fR' to the service type.
312.RE
313.TP
314.BI ioengine \fR=\fPstr
315Defines how the job issues I/O. The following types are defined:
316.RS
317.RS
318.TP
319.B sync
320Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
321position the I/O location.
322.TP
gurudas paia31041e2007-10-23 15:12:30 +0200323.B psync
324Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
325.TP
Jens Axboe91837882008-02-05 12:02:07 +0100326.B vsync
327Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
328coalescing adjacents IOs into a single submission.
329.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200330.B libaio
331Linux native asynchronous I/O.
332.TP
333.B posixaio
Bruce Cran03e20d62011-01-02 20:14:54 +0100334POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
335.TP
336.B solarisaio
337Solaris native asynchronous I/O.
338.TP
339.B windowsaio
340Windows native asynchronous I/O.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200341.TP
342.B mmap
Aaron Carrolld1429b52007-09-18 08:10:57 +0200343File is memory mapped with \fImmap\fR\|(2) and data copied using
344\fImemcpy\fR\|(3).
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200345.TP
346.B splice
Aaron Carrolld1429b52007-09-18 08:10:57 +0200347\fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
348transfer data from user-space to the kernel.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200349.TP
350.B syslet-rw
351Use the syslet system calls to make regular read/write asynchronous.
352.TP
353.B sg
354SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
Aaron Carrolld1429b52007-09-18 08:10:57 +0200355the target is an sg character device, we use \fIread\fR\|(2) and
356\fIwrite\fR\|(2) for asynchronous I/O.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200357.TP
358.B null
359Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
360itself and for debugging and testing purposes.
361.TP
362.B net
363Transfer over the network. \fBfilename\fR must be set appropriately to
364`\fIhost\fR/\fIport\fR' regardless of data direction. If receiving, only the
365\fIport\fR argument is used.
366.TP
367.B netsplice
368Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
369and send/receive.
370.TP
gurudas pai53aec0a2007-10-05 13:20:18 +0200371.B cpuio
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200372Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
373\fBcpucycles\fR parameters.
374.TP
375.B guasi
376The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
377approach to asycnronous I/O.
Aaron Carrolld1429b52007-09-18 08:10:57 +0200378.br
379See <http://www.xmailserver.org/guasi\-lib.html>.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200380.TP
381.B external
382Loads an external I/O engine object file. Append the engine filename as
383`:\fIenginepath\fR'.
384.RE
385.RE
386.TP
387.BI iodepth \fR=\fPint
Sebastian Kayser8489dae2010-12-01 22:28:47 +0100388Number of I/O units to keep in flight against the file. Note that increasing
389iodepth beyond 1 will not affect synchronous ioengines (except for small
Jens Axboeee72ca02010-12-02 20:05:37 +0100390degress when verify_async is in use). Even async engines my impose OS
391restrictions causing the desired depth not to be achieved. This may happen on
392Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
393not async on that OS. Keep an eye on the IO depth distribution in the
394fio output to verify that the achieved depth is as expected. Default: 1.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200395.TP
396.BI iodepth_batch \fR=\fPint
397Number of I/Os to submit at once. Default: \fBiodepth\fR.
398.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200399.BI iodepth_batch_complete \fR=\fPint
400This defines how many pieces of IO to retrieve at once. It defaults to 1 which
401 means that we'll ask for a minimum of 1 IO in the retrieval process from the
402kernel. The IO retrieval will go on until we hit the limit set by
403\fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
404completed events before queuing more IO. This helps reduce IO latency, at the
405cost of more retrieval system calls.
406.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200407.BI iodepth_low \fR=\fPint
408Low watermark indicating when to start filling the queue again. Default:
409\fBiodepth\fR.
410.TP
411.BI direct \fR=\fPbool
412If true, use non-buffered I/O (usually O_DIRECT). Default: false.
413.TP
414.BI buffered \fR=\fPbool
415If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
416Default: true.
417.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100418.BI offset \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200419Offset in the file to start I/O. Data before the offset will not be touched.
420.TP
421.BI fsync \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200422How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
4230, don't sync. Default: 0.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200424.TP
Jens Axboe5f9099e2009-06-16 22:40:26 +0200425.BI fdatasync \fR=\fPint
426Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
427data parts of the file. Default: 0.
428.TP
Jens Axboee76b1da2010-03-09 20:49:54 +0100429.BI sync_file_range \fR=\fPstr:int
430Use sync_file_range() for every \fRval\fP number of write operations. Fio will
431track range of writes that have happened since the last sync_file_range() call.
432\fRstr\fP can currently be one or more of:
433.RS
434.TP
435.B wait_before
436SYNC_FILE_RANGE_WAIT_BEFORE
437.TP
438.B write
439SYNC_FILE_RANGE_WRITE
440.TP
441.B wait_after
442SYNC_FILE_RANGE_WRITE
443.TP
444.RE
445.P
446So if you do sync_file_range=wait_before,write:8, fio would use
447\fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
448Also see the sync_file_range(2) man page. This option is Linux specific.
449.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200450.BI overwrite \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200451If writing, setup the file first and do overwrites. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200452.TP
453.BI end_fsync \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200454Sync file contents when job exits. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200455.TP
456.BI fsync_on_close \fR=\fPbool
457If true, sync file contents on close. This differs from \fBend_fsync\fR in that
Aaron Carrolld1429b52007-09-18 08:10:57 +0200458it will happen on every close, not just at the end of the job. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200459.TP
460.BI rwmixcycle \fR=\fPint
461How many milliseconds before switching between reads and writes for a mixed
462workload. Default: 500ms.
463.TP
464.BI rwmixread \fR=\fPint
465Percentage of a mixed workload that should be reads. Default: 50.
466.TP
467.BI rwmixwrite \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200468Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
Jens Axboec35dd7a2009-06-10 08:39:16 +0200469\fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
470overrides the first. This may interfere with a given rate setting, if fio is
471asked to limit reads or writes to a certain rate. If that is the case, then
472the distribution may be skewed. Default: 50.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200473.TP
474.B norandommap
475Normally \fBfio\fR will cover every block of the file when doing random I/O. If
476this parameter is given, a new offset will be chosen without looking at past
477I/O history. This parameter is mutually exclusive with \fBverify\fR.
478.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200479.B softrandommap
480See \fBnorandommap\fR. If fio runs with the random block map enabled and it
481fails to allocate the map, if this option is set it will continue without a
482random block map. As coverage will not be as complete as with random maps, this
483option is disabled by default.
484.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200485.BI nice \fR=\fPint
486Run job with given nice value. See \fInice\fR\|(2).
487.TP
488.BI prio \fR=\fPint
489Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
490\fIionice\fR\|(1).
491.TP
492.BI prioclass \fR=\fPint
493Set I/O priority class. See \fIionice\fR\|(1).
494.TP
495.BI thinktime \fR=\fPint
496Stall job for given number of microseconds between issuing I/Os.
497.TP
498.BI thinktime_spin \fR=\fPint
499Pretend to spend CPU time for given number of microseconds, sleeping the rest
500of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
501.TP
502.BI thinktime_blocks \fR=\fPint
503Number of blocks to issue before waiting \fBthinktime\fR microseconds.
504Default: 1.
505.TP
506.BI rate \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200507Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
508rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
509or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
510limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
511can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
512limit writes (to 500KB/sec), the latter will only limit reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200513.TP
514.BI ratemin \fR=\fPint
515Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
Jens Axboec35dd7a2009-06-10 08:39:16 +0200516Failing to meet this requirement will cause the job to exit. The same format
517as \fBrate\fR is used for read vs write separation.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200518.TP
519.BI rate_iops \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200520Cap the bandwidth to this number of IOPS. Basically the same as rate, just
521specified independently of bandwidth. The same format as \fBrate\fR is used for
522read vs write seperation. If \fBblocksize\fR is a range, the smallest block
523size is used as the metric.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200524.TP
525.BI rate_iops_min \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200526If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
527is used for read vs write seperation.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200528.TP
529.BI ratecycle \fR=\fPint
530Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
531milliseconds. Default: 1000ms.
532.TP
533.BI cpumask \fR=\fPint
534Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
535may run on. See \fBsched_setaffinity\fR\|(2).
536.TP
537.BI cpus_allowed \fR=\fPstr
538Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
539.TP
540.BI startdelay \fR=\fPint
541Delay start of job for the specified number of seconds.
542.TP
543.BI runtime \fR=\fPint
544Terminate processing after the specified number of seconds.
545.TP
546.B time_based
547If given, run for the specified \fBruntime\fR duration even if the files are
548completely read or written. The same workload will be repeated as many times
549as \fBruntime\fR allows.
550.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100551.BI ramp_time \fR=\fPint
552If set, fio will run the specified workload for this amount of time before
553logging any performance numbers. Useful for letting performance settle before
554logging results, thus minimizing the runtime required for stable results. Note
Jens Axboec35dd7a2009-06-10 08:39:16 +0200555that the \fBramp_time\fR is considered lead in time for a job, thus it will
556increase the total runtime if a special timeout or runtime is specified.
Jens Axboe901bb992009-03-14 20:17:36 +0100557.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200558.BI invalidate \fR=\fPbool
559Invalidate buffer-cache for the file prior to starting I/O. Default: true.
560.TP
561.BI sync \fR=\fPbool
562Use synchronous I/O for buffered writes. For the majority of I/O engines,
Aaron Carrolld1429b52007-09-18 08:10:57 +0200563this means using O_SYNC. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200564.TP
565.BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
566Allocation method for I/O unit buffer. Allowed values are:
567.RS
568.RS
569.TP
570.B malloc
571Allocate memory with \fImalloc\fR\|(3).
572.TP
573.B shm
574Use shared memory buffers allocated through \fIshmget\fR\|(2).
575.TP
576.B shmhuge
577Same as \fBshm\fR, but use huge pages as backing.
578.TP
579.B mmap
580Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
581is given after the option in the format `:\fIfile\fR'.
582.TP
583.B mmaphuge
584Same as \fBmmap\fR, but use huge files as backing.
585.RE
586.P
587The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
588job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
589the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
Jens Axboe2e266ba2009-09-14 08:56:53 +0200590have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
591huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
592and the documentation for that. Normally you just need to echo an appropriate
593number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
594use.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200595.RE
596.TP
Jens Axboed529ee12009-07-01 10:33:03 +0200597.BI iomem_align \fR=\fPint
598This indiciates the memory alignment of the IO memory buffers. Note that the
599given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
600the alignment of the following buffers are given by the \fBbs\fR used. In
601other words, if using a \fBbs\fR that is a multiple of the page sized in the
602system, all buffers will be aligned to this value. If using a \fBbs\fR that
603is not page aligned, the alignment of subsequent IO memory buffers is the
604sum of the \fBiomem_align\fR and \fBbs\fR used.
605.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100606.BI hugepage\-size \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200607Defines the size of a huge page. Must be at least equal to the system setting.
Jens Axboeb22989b2009-07-17 22:29:23 +0200608Should be a multiple of 1MB. Default: 4MB.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200609.TP
610.B exitall
611Terminate all jobs when one finishes. Default: wait for each job to finish.
612.TP
613.BI bwavgtime \fR=\fPint
614Average bandwidth calculations over the given time in milliseconds. Default:
615500ms.
616.TP
617.BI create_serialize \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200618If true, serialize file creation for the jobs. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200619.TP
620.BI create_fsync \fR=\fPbool
621\fIfsync\fR\|(2) data file after creation. Default: true.
622.TP
Jens Axboe6b7f6852009-03-09 14:22:56 +0100623.BI create_on_open \fR=\fPbool
624If true, the files are not created until they are opened for IO by the job.
625.TP
Jens Axboee9f48472009-06-03 12:14:08 +0200626.BI pre_read \fR=\fPbool
627If this is given, files will be pre-read into memory before starting the given
628IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
Jens Axboe9c0d2242009-07-01 12:26:28 +0200629pointless to pre-read and then drop the cache. This will only work for IO
630engines that are seekable, since they allow you to read the same data
631multiple times. Thus it will not work on eg network or splice IO.
Jens Axboee9f48472009-06-03 12:14:08 +0200632.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200633.BI unlink \fR=\fPbool
634Unlink job files when done. Default: false.
635.TP
636.BI loops \fR=\fPint
637Specifies the number of iterations (runs of the same workload) of this job.
638Default: 1.
639.TP
640.BI do_verify \fR=\fPbool
641Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
642Default: true.
643.TP
644.BI verify \fR=\fPstr
645Method of verifying file contents after each iteration of the job. Allowed
646values are:
647.RS
648.RS
649.TP
Jens Axboeb892dc02009-09-05 20:37:35 +0200650.B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
Jens Axboe0539d752010-06-21 15:22:56 +0200651Store appropriate checksum in the header of each block. crc32c-intel is
652hardware accelerated SSE4.2 driven, falls back to regular crc32c if
653not supported by the system.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200654.TP
655.B meta
656Write extra information about each I/O (timestamp, block number, etc.). The
Jens Axboe996093b2010-06-24 08:37:13 +0200657block number is verified. See \fBverify_pattern\fR as well.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200658.TP
659.B null
660Pretend to verify. Used for testing internals.
661.RE
Jens Axboeb892dc02009-09-05 20:37:35 +0200662
663This option can be used for repeated burn-in tests of a system to make sure
664that the written data is also correctly read back. If the data direction given
665is a read or random read, fio will assume that it should verify a previously
666written file. If the data direction includes any form of write, the verify will
667be of the newly written data.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200668.RE
669.TP
670.BI verify_sort \fR=\fPbool
671If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
672read them back in a sorted manner. Default: true.
673.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100674.BI verify_offset \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200675Swap the verification header with data somewhere else in the block before
Aaron Carrolld1429b52007-09-18 08:10:57 +0200676writing. It is swapped back before verifying.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200677.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100678.BI verify_interval \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200679Write the verification header for this number of bytes, which should divide
680\fBblocksize\fR. Default: \fBblocksize\fR.
681.TP
Jens Axboe996093b2010-06-24 08:37:13 +0200682.BI verify_pattern \fR=\fPstr
683If set, fio will fill the io buffers with this pattern. Fio defaults to filling
684with totally random bytes, but sometimes it's interesting to fill with a known
685pattern for io verification purposes. Depending on the width of the pattern,
686fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
687decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
688has to be a hex number that starts with either "0x" or "0X". Use with
689\fBverify\fP=meta.
690.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200691.BI verify_fatal \fR=\fPbool
692If true, exit the job on the first observed verification failure. Default:
693false.
694.TP
Jens Axboeb463e932011-01-12 09:03:23 +0100695.BI verify_dump \fR=\fPbool
696If set, dump the contents of both the original data block and the data block we
697read off disk to files. This allows later analysis to inspect just what kind of
698data corruption occurred. On by default.
699.TP
Jens Axboee8462bd2009-07-06 12:59:04 +0200700.BI verify_async \fR=\fPint
701Fio will normally verify IO inline from the submitting thread. This option
702takes an integer describing how many async offload threads to create for IO
703verification instead, causing fio to offload the duty of verifying IO contents
Jens Axboec85c3242009-07-06 14:12:57 +0200704to one or more separate threads. If using this offload option, even sync IO
705engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
706allows them to have IO in flight while verifies are running.
Jens Axboee8462bd2009-07-06 12:59:04 +0200707.TP
708.BI verify_async_cpus \fR=\fPstr
709Tell fio to set the given CPU affinity on the async IO verification threads.
710See \fBcpus_allowed\fP for the format used.
711.TP
Jens Axboe6f874182010-06-21 12:53:26 +0200712.BI verify_backlog \fR=\fPint
713Fio will normally verify the written contents of a job that utilizes verify
714once that job has completed. In other words, everything is written then
715everything is read back and verified. You may want to verify continually
716instead for a variety of reasons. Fio stores the meta data associated with an
717IO block in memory, so for large verify workloads, quite a bit of memory would
David Nellans092f7072010-10-26 08:08:42 -0600718be used up holding this meta data. If this option is enabled, fio will write
719only N blocks before verifying these blocks.
Jens Axboe6f874182010-06-21 12:53:26 +0200720.TP
721.BI verify_backlog_batch \fR=\fPint
722Control how many blocks fio will verify if verify_backlog is set. If not set,
723will default to the value of \fBverify_backlog\fR (meaning the entire queue is
David Nellans092f7072010-10-26 08:08:42 -0600724read back and verified). If \fBverify_backlog_batch\fR is less than
725\fBverify_backlog\fR then not all blocks will be verified, if
726\fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
727will be verified more than once.
Jens Axboe6f874182010-06-21 12:53:26 +0200728.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200729.B stonewall
Aaron Carrolld1429b52007-09-18 08:10:57 +0200730Wait for preceeding jobs in the job file to exit before starting this one.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200731\fBstonewall\fR implies \fBnew_group\fR.
732.TP
733.B new_group
734Start a new reporting group. If not given, all jobs in a file will be part
735of the same reporting group, unless separated by a stonewall.
736.TP
737.BI numjobs \fR=\fPint
738Number of clones (processes/threads performing the same workload) of this job.
739Default: 1.
740.TP
741.B group_reporting
742If set, display per-group reports instead of per-job when \fBnumjobs\fR is
743specified.
744.TP
745.B thread
746Use threads created with \fBpthread_create\fR\|(3) instead of processes created
747with \fBfork\fR\|(2).
748.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100749.BI zonesize \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200750Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
751.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100752.BI zoneskip \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200753Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200754read.
755.TP
756.BI write_iolog \fR=\fPstr
Stefan Hajnoczi5b42a482011-01-08 20:28:41 +0100757Write the issued I/O patterns to the specified file. Specify a separate file
758for each job, otherwise the iologs will be interspersed and the file may be
759corrupt.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200760.TP
761.BI read_iolog \fR=\fPstr
762Replay the I/O patterns contained in the specified file generated by
763\fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
764.TP
David Nellans64bbb862010-08-24 22:13:30 +0200765.BI replay_no_stall \fR=\fPint
766While replaying I/O patterns using \fBread_iolog\fR the default behavior
767attempts to respect timing information between I/Os. Enabling
768\fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
769still respecting ordering.
770.TP
David Nellansd1c46c02010-08-31 21:20:47 +0200771.BI replay_redirect \fR=\fPstr
772While replaying I/O patterns using \fBread_iolog\fR the default behavior
773is to replay the IOPS onto the major/minor device that each IOP was recorded
774from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
775single specified device regardless of the device it was recorded from.
776.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100777.B write_bw_log \fR=\fPstr
778If given, write a bandwidth log of the jobs in this job file. Can be used to
779store data of the bandwidth of the jobs in their lifetime. The included
780fio_generate_plots script uses gnuplot to turn these text files into nice
781graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
782option, the postfix is _bw.log.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200783.TP
David Nellans64bbb862010-08-24 22:13:30 +0200784.B write_lat_log \fR=\fPstr
Jens Axboe901bb992009-03-14 20:17:36 +0100785Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
786filename is given with this option, the default filename of "jobname_type.log"
787is used. Even if the filename is given, fio will still append the type of log.
788.TP
Jens Axboe02af0982010-06-24 09:59:34 +0200789.B disable_lat \fR=\fPbool
790Disable measurements of total latency numbers. Useful only for cutting
Jens Axboe901bb992009-03-14 20:17:36 +0100791back the number of calls to gettimeofday, as that does impact performance at
792really high IOPS rates. Note that to really get rid of a large amount of these
793calls, this option must be used with disable_slat and disable_bw as well.
794.TP
Jens Axboe02af0982010-06-24 09:59:34 +0200795.B disable_clat \fR=\fPbool
796Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
797.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100798.B disable_slat \fR=\fPbool
Jens Axboe02af0982010-06-24 09:59:34 +0200799Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
Jens Axboe901bb992009-03-14 20:17:36 +0100800.TP
801.B disable_bw_measurement \fR=\fPbool
Jens Axboe02af0982010-06-24 09:59:34 +0200802Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200803.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100804.BI lockmem \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200805Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
806simulate a smaller amount of memory.
807.TP
808.BI exec_prerun \fR=\fPstr
809Before running the job, execute the specified command with \fBsystem\fR\|(3).
810.TP
811.BI exec_postrun \fR=\fPstr
812Same as \fBexec_prerun\fR, but the command is executed after the job completes.
813.TP
814.BI ioscheduler \fR=\fPstr
815Attempt to switch the device hosting the file to the specified I/O scheduler.
816.TP
817.BI cpuload \fR=\fPint
818If the job is a CPU cycle-eater, attempt to use the specified percentage of
819CPU cycles.
820.TP
821.BI cpuchunks \fR=\fPint
822If the job is a CPU cycle-eater, split the load into cycles of the
823given time in milliseconds.
824.TP
825.BI disk_util \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200826Generate disk utilization statistics if the platform supports it. Default: true.
Jens Axboe901bb992009-03-14 20:17:36 +0100827.TP
828.BI gtod_reduce \fR=\fPbool
829Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
830disable_bw) plus reduce precision of the timeout somewhat to really shrink the
831gettimeofday() call count. With this option enabled, we only do about 0.4% of
832the gtod() calls we would have done if all time keeping was enabled.
833.TP
834.BI gtod_cpu \fR=\fPint
835Sometimes it's cheaper to dedicate a single thread of execution to just getting
836the current time. Fio (and databases, for instance) are very intensive on
837gettimeofday() calls. With this option, you can set one CPU aside for doing
838nothing but logging current time to a shared memory location. Then the other
839threads/processes that run IO workloads need only copy that segment, instead of
840entering the kernel with a gettimeofday() call. The CPU set aside for doing
841these time calls will be excluded from other uses. Fio will manually clear it
842from the CPU mask of other jobs.
Radha Ramachandranf2bba182009-06-15 08:40:16 +0200843.TP
Jens Axboea696fa22009-12-04 10:05:02 +0100844.BI cgroup \fR=\fPstr
845Add job to this control group. If it doesn't exist, it will be created.
Jens Axboe6adb38a2009-12-07 08:01:26 +0100846The system must have a mounted cgroup blkio mount point for this to work. If
847your system doesn't have it mounted, you can do so with:
848
849# mount -t cgroup -o blkio none /cgroup
Jens Axboea696fa22009-12-04 10:05:02 +0100850.TP
851.BI cgroup_weight \fR=\fPint
852Set the weight of the cgroup to this value. See the documentation that comes
853with the kernel, allowed values are in the range of 100..1000.
Jens Axboee0b0d892009-12-08 10:10:14 +0100854.TP
Vivek Goyal7de87092010-03-31 22:55:15 +0200855.BI cgroup_nodelete \fR=\fPbool
856Normally fio will delete the cgroups it has created after the job completion.
857To override this behavior and to leave cgroups around after the job completion,
858set cgroup_nodelete=1. This can be useful if one wants to inspect various
859cgroup files after job completion. Default: false
860.TP
Jens Axboee0b0d892009-12-08 10:10:14 +0100861.BI uid \fR=\fPint
862Instead of running as the invoking user, set the user ID to this value before
863the thread/process does any work.
864.TP
865.BI gid \fR=\fPint
866Set group ID, see \fBuid\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200867.SH OUTPUT
Aaron Carrolld1429b52007-09-18 08:10:57 +0200868While running, \fBfio\fR will display the status of the created jobs. For
869example:
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200870.RS
Aaron Carrolld1429b52007-09-18 08:10:57 +0200871.P
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200872Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
873.RE
874.P
Aaron Carrolld1429b52007-09-18 08:10:57 +0200875The characters in the first set of brackets denote the current status of each
876threads. The possible values are:
877.P
878.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200879.RS
880.TP
881.B P
882Setup but not started.
883.TP
884.B C
885Thread created.
886.TP
887.B I
888Initialized, waiting.
889.TP
890.B R
891Running, doing sequential reads.
892.TP
893.B r
894Running, doing random reads.
895.TP
896.B W
897Running, doing sequential writes.
898.TP
899.B w
900Running, doing random writes.
901.TP
902.B M
903Running, doing mixed sequential reads/writes.
904.TP
905.B m
906Running, doing mixed random reads/writes.
907.TP
908.B F
909Running, currently waiting for \fBfsync\fR\|(2).
910.TP
911.B V
912Running, verifying written data.
913.TP
914.B E
915Exited, not reaped by main thread.
916.TP
917.B \-
918Exited, thread reaped.
919.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +0200920.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200921.P
922The second set of brackets shows the estimated completion percentage of
923the current group. The third set shows the read and write I/O rate,
924respectively. Finally, the estimated run time of the job is displayed.
925.P
926When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
927for each thread, each group of threads, and each disk, in that order.
928.P
929Per-thread statistics first show the threads client number, group-id, and
930error code. The remaining figures are as follows:
931.RS
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200932.TP
933.B io
934Number of megabytes of I/O performed.
935.TP
936.B bw
937Average data rate (bandwidth).
938.TP
939.B runt
940Threads run time.
941.TP
942.B slat
943Submission latency minimum, maximum, average and standard deviation. This is
944the time it took to submit the I/O.
945.TP
946.B clat
947Completion latency minimum, maximum, average and standard deviation. This
948is the time between submission and completion.
949.TP
950.B bw
951Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
952and standard deviation.
953.TP
954.B cpu
955CPU usage statistics. Includes user and system time, number of context switches
956this thread went through and number of major and minor page faults.
957.TP
958.B IO depths
959Distribution of I/O depths. Each depth includes everything less than (or equal)
960to it, but greater than the previous depth.
961.TP
962.B IO issued
963Number of read/write requests issued, and number of short read/write requests.
964.TP
965.B IO latencies
966Distribution of I/O completion latencies. The numbers follow the same pattern
967as \fBIO depths\fR.
968.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200969.P
970The group statistics show:
Aaron Carrolld1429b52007-09-18 08:10:57 +0200971.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200972.RS
973.TP
974.B io
975Number of megabytes I/O performed.
976.TP
977.B aggrb
978Aggregate bandwidth of threads in the group.
979.TP
980.B minb
981Minimum average bandwidth a thread saw.
982.TP
983.B maxb
984Maximum average bandwidth a thread saw.
985.TP
986.B mint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200987Shortest runtime of threads in the group.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200988.TP
989.B maxt
990Longest runtime of threads in the group.
991.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +0200992.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200993.P
994Finally, disk statistics are printed with reads first:
Aaron Carrolld1429b52007-09-18 08:10:57 +0200995.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200996.RS
997.TP
998.B ios
999Number of I/Os performed by all groups.
1000.TP
1001.B merge
1002Number of merges in the I/O scheduler.
1003.TP
1004.B ticks
1005Number of ticks we kept the disk busy.
1006.TP
1007.B io_queue
1008Total time spent in the disk queue.
1009.TP
1010.B util
1011Disk utilization.
1012.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +02001013.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001014.SH TERSE OUTPUT
1015If the \fB\-\-minimal\fR option is given, the results will be printed in a
David Nellans562c2d22010-09-23 08:38:17 +02001016semicolon-delimited format suitable for scripted use - a job description
1017(if provided) follows on a new line. Note that the first
Jens Axboe525c2bf2010-06-30 15:22:21 +02001018number in the line is the version number. If the output has to be changed
1019for some reason, this number will be incremented by 1 to signify that
1020change. The fields are:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001021.P
1022.RS
Jens Axboe525c2bf2010-06-30 15:22:21 +02001023.B version, jobname, groupid, error
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001024.P
1025Read status:
1026.RS
Jens Axboeb22989b2009-07-17 22:29:23 +02001027.B KB I/O, bandwidth \fR(KB/s)\fP, runtime \fR(ms)\fP
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001028.P
1029Submission latency:
1030.RS
1031.B min, max, mean, standard deviation
1032.RE
1033Completion latency:
1034.RS
1035.B min, max, mean, standard deviation
1036.RE
Jens Axboe525c2bf2010-06-30 15:22:21 +02001037Total latency:
1038.RS
1039.B min, max, mean, standard deviation
1040.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001041Bandwidth:
1042.RS
1043.B min, max, aggregate percentage of total, mean, standard deviation
1044.RE
1045.RE
1046.P
1047Write status:
1048.RS
Jens Axboeb22989b2009-07-17 22:29:23 +02001049.B KB I/O, bandwidth \fR(KB/s)\fP, runtime \fR(ms)\fP
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001050.P
1051Submission latency:
1052.RS
1053.B min, max, mean, standard deviation
1054.RE
1055Completion latency:
1056.RS
1057.B min, max, mean, standard deviation
1058.RE
Jens Axboe525c2bf2010-06-30 15:22:21 +02001059Total latency:
1060.RS
1061.B min, max, mean, standard deviation
1062.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001063Bandwidth:
1064.RS
1065.B min, max, aggregate percentage of total, mean, standard deviation
1066.RE
1067.RE
1068.P
Aaron Carrolld1429b52007-09-18 08:10:57 +02001069CPU usage:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001070.RS
Carl Henrik Lundebd2626f2008-06-12 09:17:46 +02001071.B user, system, context switches, major page faults, minor page faults
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001072.RE
1073.P
1074IO depth distribution:
1075.RS
1076.B <=1, 2, 4, 8, 16, 32, >=64
1077.RE
1078.P
David Nellans562c2d22010-09-23 08:38:17 +02001079IO latency distribution:
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001080.RS
David Nellans562c2d22010-09-23 08:38:17 +02001081Microseconds:
1082.RS
1083.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1084.RE
1085Milliseconds:
1086.RS
1087.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1088.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001089.RE
1090.P
David Nellans562c2d22010-09-23 08:38:17 +02001091Error Info (dependant on continue_on_error, default off):
1092.RS
1093.B total # errors, first error code
1094.RE
1095.P
1096.B text description (if provided in config - appears on newline)
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001097.RE
1098.SH AUTHORS
1099.B fio
Jens Axboeaa58d252010-06-09 09:49:38 +02001100was written by Jens Axboe <jens.axboe@oracle.com>,
1101now Jens Axboe <jaxboe@fusionio.com>.
Aaron Carrolld1429b52007-09-18 08:10:57 +02001102.br
1103This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001104on documentation by Jens Axboe.
1105.SH "REPORTING BUGS"
Jens Axboe482900c2009-06-02 12:15:51 +02001106Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
Aaron Carrolld1429b52007-09-18 08:10:57 +02001107See \fBREADME\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001108.SH "SEE ALSO"
Aaron Carrolld1429b52007-09-18 08:10:57 +02001109For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1110.br
1111Sample jobfiles are available in the \fBexamples\fR directory.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001112