blob: 0e6095d6499b1f33e8cd1937a17ecbdd2fa548b5 [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
Aaron Carrolld1429b52007-09-18 08:10:57 +0200175For mixed I/O, the default split is 50/50. For random I/O, the number of I/Os
176to perform before getting a new offset can be specified by appending
177`:\fIint\fR' to the pattern type. The default is 1.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200178.RE
179.TP
Jens Axboe90fef2d2009-07-17 22:33:32 +0200180.BI kb_base \fR=\fPint
181The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
182manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
183reasons. Allow values are 1024 or 1000, with 1024 being the default.
184.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200185.BI randrepeat \fR=\fPbool
186Seed the random number generator in a predictable way so results are repeatable
Aaron Carrolld1429b52007-09-18 08:10:57 +0200187across runs. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200188.TP
Jens Axboe7bc8c2c2010-01-28 11:31:31 +0100189.BI fallocate \fR=\fPbool
190By default, fio will use fallocate() to advise the system of the size of the
191file we are going to write. This can be turned off with fallocate=0. May not
192be available on all supported platforms.
193.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200194.BI fadvise_hint \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200195Disable use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
196are likely to be issued. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200197.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100198.BI size \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200199Total size of I/O for this job. \fBfio\fR will run until this many bytes have
200been transfered, unless limited by other options (\fBruntime\fR, for instance).
201Unless \fBnr_files\fR and \fBfilesize\fR options are given, this amount will be
202divided between the available files for the job.
203.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200204.BI fill_device \fR=\fPbool
205Sets size to something really large and waits for ENOSPC (no space left on
206device) as the terminating condition. Only makes sense with sequential write.
207For a read workload, the mount point will be filled first then IO started on
208the result.
209.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200210.BI filesize \fR=\fPirange
211Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
Aaron Carrolld1429b52007-09-18 08:10:57 +0200212for files at random within the given range, limited to \fBsize\fR in total (if
213that is given). If \fBfilesize\fR is not specified, each created file is the
214same size.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200215.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100216.BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200217Block size for I/O units. Default: 4k. Values for reads and writes can be
Jens Axboe656ebab2010-04-13 10:39:14 +0200218specified separately in the format \fIread\fR,\fIwrite\fR, either of
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200219which may be empty to leave that value at its default.
220.TP
Jens Axboe91837882008-02-05 12:02:07 +0100221.BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
Aaron Carrolld1429b52007-09-18 08:10:57 +0200222Specify a range of I/O block sizes. The issued I/O unit will always be a
223multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
Jens Axboe91837882008-02-05 12:02:07 +0100224to both reads and writes if only one range is given, but can be specified
Jens Axboe656ebab2010-04-13 10:39:14 +0200225separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
Jens Axboe91837882008-02-05 12:02:07 +0100226Also (see \fBblocksize\fR).
227.TP
228.BI bssplit \fR=\fPstr
229This option allows even finer grained control of the block sizes issued,
230not just even splits between them. With this option, you can weight various
231block sizes for exact control of the issued IO for a job that has mixed
232block sizes. The format of the option is bssplit=blocksize/percentage,
233optionally adding as many definitions as needed seperated by a colon.
234Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
Jens Axboec83cdd32009-04-24 14:23:59 +0200235blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
236splits to reads and writes. The format is identical to what the
237\fBbs\fR option accepts, the read and write parts are separated with a
238comma.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200239.TP
240.B blocksize_unaligned\fR,\fP bs_unaligned
Aaron Carrolld1429b52007-09-18 08:10:57 +0200241If set, any size in \fBblocksize_range\fR may be used. This typically won't
242work with direct I/O, as that normally requires sector alignment.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200243.TP
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100244.BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
Martin Steigerwald639ce0f2009-05-20 11:33:49 +0200245At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
246the minimum blocksize given. Minimum alignment is typically 512b
Jens Axboe2b7a01d2009-03-11 11:00:13 +0100247for using direct IO, though it usually depends on the hardware block size.
248This option is mutually exclusive with using a random map for files, so it
249will turn off that option.
Jens Axboe43602662009-03-14 20:08:47 +0100250.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200251.B zero_buffers
252Initialise buffers with all zeros. Default: fill buffers with random data.
253.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100254.B refill_buffers
255If this option is given, fio will refill the IO buffers on every submit. The
256default is to only fill it at init time and reuse that data. Only makes sense
257if zero_buffers isn't specified, naturally. If data verification is enabled,
258refill_buffers is also automatically enabled.
259.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200260.BI nrfiles \fR=\fPint
261Number of files to use for this job. Default: 1.
262.TP
263.BI openfiles \fR=\fPint
264Number of files to keep open at the same time. Default: \fBnrfiles\fR.
265.TP
266.BI file_service_type \fR=\fPstr
267Defines how files to service are selected. The following types are defined:
268.RS
269.RS
270.TP
271.B random
272Choose a file at random
273.TP
274.B roundrobin
275Round robin over open files (default).
Jens Axboe6b7f6852009-03-09 14:22:56 +0100276.B sequential
277Do each file in the set sequentially.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200278.RE
279.P
280The number of I/Os to issue before switching a new file can be specified by
281appending `:\fIint\fR' to the service type.
282.RE
283.TP
284.BI ioengine \fR=\fPstr
285Defines how the job issues I/O. The following types are defined:
286.RS
287.RS
288.TP
289.B sync
290Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
291position the I/O location.
292.TP
gurudas paia31041e2007-10-23 15:12:30 +0200293.B psync
294Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
295.TP
Jens Axboe91837882008-02-05 12:02:07 +0100296.B vsync
297Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
298coalescing adjacents IOs into a single submission.
299.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200300.B libaio
301Linux native asynchronous I/O.
302.TP
303.B posixaio
304glibc POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
305.TP
306.B mmap
Aaron Carrolld1429b52007-09-18 08:10:57 +0200307File is memory mapped with \fImmap\fR\|(2) and data copied using
308\fImemcpy\fR\|(3).
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200309.TP
310.B splice
Aaron Carrolld1429b52007-09-18 08:10:57 +0200311\fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
312transfer data from user-space to the kernel.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200313.TP
314.B syslet-rw
315Use the syslet system calls to make regular read/write asynchronous.
316.TP
317.B sg
318SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
Aaron Carrolld1429b52007-09-18 08:10:57 +0200319the target is an sg character device, we use \fIread\fR\|(2) and
320\fIwrite\fR\|(2) for asynchronous I/O.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200321.TP
322.B null
323Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
324itself and for debugging and testing purposes.
325.TP
326.B net
327Transfer over the network. \fBfilename\fR must be set appropriately to
328`\fIhost\fR/\fIport\fR' regardless of data direction. If receiving, only the
329\fIport\fR argument is used.
330.TP
331.B netsplice
332Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
333and send/receive.
334.TP
gurudas pai53aec0a2007-10-05 13:20:18 +0200335.B cpuio
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200336Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
337\fBcpucycles\fR parameters.
338.TP
339.B guasi
340The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
341approach to asycnronous I/O.
Aaron Carrolld1429b52007-09-18 08:10:57 +0200342.br
343See <http://www.xmailserver.org/guasi\-lib.html>.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200344.TP
345.B external
346Loads an external I/O engine object file. Append the engine filename as
347`:\fIenginepath\fR'.
348.RE
349.RE
350.TP
351.BI iodepth \fR=\fPint
352Number of I/O units to keep in flight against the file. Default: 1.
353.TP
354.BI iodepth_batch \fR=\fPint
355Number of I/Os to submit at once. Default: \fBiodepth\fR.
356.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200357.BI iodepth_batch_complete \fR=\fPint
358This defines how many pieces of IO to retrieve at once. It defaults to 1 which
359 means that we'll ask for a minimum of 1 IO in the retrieval process from the
360kernel. The IO retrieval will go on until we hit the limit set by
361\fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
362completed events before queuing more IO. This helps reduce IO latency, at the
363cost of more retrieval system calls.
364.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200365.BI iodepth_low \fR=\fPint
366Low watermark indicating when to start filling the queue again. Default:
367\fBiodepth\fR.
368.TP
369.BI direct \fR=\fPbool
370If true, use non-buffered I/O (usually O_DIRECT). Default: false.
371.TP
372.BI buffered \fR=\fPbool
373If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
374Default: true.
375.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100376.BI offset \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200377Offset in the file to start I/O. Data before the offset will not be touched.
378.TP
379.BI fsync \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200380How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
3810, don't sync. Default: 0.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200382.TP
Jens Axboe5f9099e2009-06-16 22:40:26 +0200383.BI fdatasync \fR=\fPint
384Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
385data parts of the file. Default: 0.
386.TP
Jens Axboee76b1da2010-03-09 20:49:54 +0100387.BI sync_file_range \fR=\fPstr:int
388Use sync_file_range() for every \fRval\fP number of write operations. Fio will
389track range of writes that have happened since the last sync_file_range() call.
390\fRstr\fP can currently be one or more of:
391.RS
392.TP
393.B wait_before
394SYNC_FILE_RANGE_WAIT_BEFORE
395.TP
396.B write
397SYNC_FILE_RANGE_WRITE
398.TP
399.B wait_after
400SYNC_FILE_RANGE_WRITE
401.TP
402.RE
403.P
404So if you do sync_file_range=wait_before,write:8, fio would use
405\fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
406Also see the sync_file_range(2) man page. This option is Linux specific.
407.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200408.BI overwrite \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200409If writing, setup the file first and do overwrites. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200410.TP
411.BI end_fsync \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200412Sync file contents when job exits. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200413.TP
414.BI fsync_on_close \fR=\fPbool
415If true, sync file contents on close. This differs from \fBend_fsync\fR in that
Aaron Carrolld1429b52007-09-18 08:10:57 +0200416it will happen on every close, not just at the end of the job. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200417.TP
418.BI rwmixcycle \fR=\fPint
419How many milliseconds before switching between reads and writes for a mixed
420workload. Default: 500ms.
421.TP
422.BI rwmixread \fR=\fPint
423Percentage of a mixed workload that should be reads. Default: 50.
424.TP
425.BI rwmixwrite \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200426Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
Jens Axboec35dd7a2009-06-10 08:39:16 +0200427\fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
428overrides the first. This may interfere with a given rate setting, if fio is
429asked to limit reads or writes to a certain rate. If that is the case, then
430the distribution may be skewed. Default: 50.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200431.TP
432.B norandommap
433Normally \fBfio\fR will cover every block of the file when doing random I/O. If
434this parameter is given, a new offset will be chosen without looking at past
435I/O history. This parameter is mutually exclusive with \fBverify\fR.
436.TP
Jens Axboe3ce9dca2009-06-10 08:55:21 +0200437.B softrandommap
438See \fBnorandommap\fR. If fio runs with the random block map enabled and it
439fails to allocate the map, if this option is set it will continue without a
440random block map. As coverage will not be as complete as with random maps, this
441option is disabled by default.
442.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200443.BI nice \fR=\fPint
444Run job with given nice value. See \fInice\fR\|(2).
445.TP
446.BI prio \fR=\fPint
447Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
448\fIionice\fR\|(1).
449.TP
450.BI prioclass \fR=\fPint
451Set I/O priority class. See \fIionice\fR\|(1).
452.TP
453.BI thinktime \fR=\fPint
454Stall job for given number of microseconds between issuing I/Os.
455.TP
456.BI thinktime_spin \fR=\fPint
457Pretend to spend CPU time for given number of microseconds, sleeping the rest
458of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
459.TP
460.BI thinktime_blocks \fR=\fPint
461Number of blocks to issue before waiting \fBthinktime\fR microseconds.
462Default: 1.
463.TP
464.BI rate \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200465Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
466rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
467or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
468limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
469can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
470limit writes (to 500KB/sec), the latter will only limit reads.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200471.TP
472.BI ratemin \fR=\fPint
473Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
Jens Axboec35dd7a2009-06-10 08:39:16 +0200474Failing to meet this requirement will cause the job to exit. The same format
475as \fBrate\fR is used for read vs write separation.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200476.TP
477.BI rate_iops \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200478Cap the bandwidth to this number of IOPS. Basically the same as rate, just
479specified independently of bandwidth. The same format as \fBrate\fR is used for
480read vs write seperation. If \fBblocksize\fR is a range, the smallest block
481size is used as the metric.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200482.TP
483.BI rate_iops_min \fR=\fPint
Jens Axboec35dd7a2009-06-10 08:39:16 +0200484If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
485is used for read vs write seperation.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200486.TP
487.BI ratecycle \fR=\fPint
488Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
489milliseconds. Default: 1000ms.
490.TP
491.BI cpumask \fR=\fPint
492Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
493may run on. See \fBsched_setaffinity\fR\|(2).
494.TP
495.BI cpus_allowed \fR=\fPstr
496Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
497.TP
498.BI startdelay \fR=\fPint
499Delay start of job for the specified number of seconds.
500.TP
501.BI runtime \fR=\fPint
502Terminate processing after the specified number of seconds.
503.TP
504.B time_based
505If given, run for the specified \fBruntime\fR duration even if the files are
506completely read or written. The same workload will be repeated as many times
507as \fBruntime\fR allows.
508.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100509.BI ramp_time \fR=\fPint
510If set, fio will run the specified workload for this amount of time before
511logging any performance numbers. Useful for letting performance settle before
512logging results, thus minimizing the runtime required for stable results. Note
Jens Axboec35dd7a2009-06-10 08:39:16 +0200513that the \fBramp_time\fR is considered lead in time for a job, thus it will
514increase the total runtime if a special timeout or runtime is specified.
Jens Axboe901bb992009-03-14 20:17:36 +0100515.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200516.BI invalidate \fR=\fPbool
517Invalidate buffer-cache for the file prior to starting I/O. Default: true.
518.TP
519.BI sync \fR=\fPbool
520Use synchronous I/O for buffered writes. For the majority of I/O engines,
Aaron Carrolld1429b52007-09-18 08:10:57 +0200521this means using O_SYNC. Default: false.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200522.TP
523.BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
524Allocation method for I/O unit buffer. Allowed values are:
525.RS
526.RS
527.TP
528.B malloc
529Allocate memory with \fImalloc\fR\|(3).
530.TP
531.B shm
532Use shared memory buffers allocated through \fIshmget\fR\|(2).
533.TP
534.B shmhuge
535Same as \fBshm\fR, but use huge pages as backing.
536.TP
537.B mmap
538Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
539is given after the option in the format `:\fIfile\fR'.
540.TP
541.B mmaphuge
542Same as \fBmmap\fR, but use huge files as backing.
543.RE
544.P
545The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
546job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
547the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
Jens Axboe2e266ba2009-09-14 08:56:53 +0200548have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
549huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
550and the documentation for that. Normally you just need to echo an appropriate
551number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
552use.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200553.RE
554.TP
Jens Axboed529ee12009-07-01 10:33:03 +0200555.BI iomem_align \fR=\fPint
556This indiciates the memory alignment of the IO memory buffers. Note that the
557given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
558the alignment of the following buffers are given by the \fBbs\fR used. In
559other words, if using a \fBbs\fR that is a multiple of the page sized in the
560system, all buffers will be aligned to this value. If using a \fBbs\fR that
561is not page aligned, the alignment of subsequent IO memory buffers is the
562sum of the \fBiomem_align\fR and \fBbs\fR used.
563.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100564.BI hugepage\-size \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200565Defines the size of a huge page. Must be at least equal to the system setting.
Jens Axboeb22989b2009-07-17 22:29:23 +0200566Should be a multiple of 1MB. Default: 4MB.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200567.TP
568.B exitall
569Terminate all jobs when one finishes. Default: wait for each job to finish.
570.TP
571.BI bwavgtime \fR=\fPint
572Average bandwidth calculations over the given time in milliseconds. Default:
573500ms.
574.TP
575.BI create_serialize \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200576If true, serialize file creation for the jobs. Default: true.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200577.TP
578.BI create_fsync \fR=\fPbool
579\fIfsync\fR\|(2) data file after creation. Default: true.
580.TP
Jens Axboe6b7f6852009-03-09 14:22:56 +0100581.BI create_on_open \fR=\fPbool
582If true, the files are not created until they are opened for IO by the job.
583.TP
Jens Axboee9f48472009-06-03 12:14:08 +0200584.BI pre_read \fR=\fPbool
585If this is given, files will be pre-read into memory before starting the given
586IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
Jens Axboe9c0d2242009-07-01 12:26:28 +0200587pointless to pre-read and then drop the cache. This will only work for IO
588engines that are seekable, since they allow you to read the same data
589multiple times. Thus it will not work on eg network or splice IO.
Jens Axboee9f48472009-06-03 12:14:08 +0200590.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200591.BI unlink \fR=\fPbool
592Unlink job files when done. Default: false.
593.TP
594.BI loops \fR=\fPint
595Specifies the number of iterations (runs of the same workload) of this job.
596Default: 1.
597.TP
598.BI do_verify \fR=\fPbool
599Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
600Default: true.
601.TP
602.BI verify \fR=\fPstr
603Method of verifying file contents after each iteration of the job. Allowed
604values are:
605.RS
606.RS
607.TP
Jens Axboeb892dc02009-09-05 20:37:35 +0200608.B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
Jens Axboe0539d752010-06-21 15:22:56 +0200609Store appropriate checksum in the header of each block. crc32c-intel is
610hardware accelerated SSE4.2 driven, falls back to regular crc32c if
611not supported by the system.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200612.TP
613.B meta
614Write extra information about each I/O (timestamp, block number, etc.). The
Jens Axboe996093b2010-06-24 08:37:13 +0200615block number is verified. See \fBverify_pattern\fR as well.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200616.TP
617.B null
618Pretend to verify. Used for testing internals.
619.RE
Jens Axboeb892dc02009-09-05 20:37:35 +0200620
621This option can be used for repeated burn-in tests of a system to make sure
622that the written data is also correctly read back. If the data direction given
623is a read or random read, fio will assume that it should verify a previously
624written file. If the data direction includes any form of write, the verify will
625be of the newly written data.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200626.RE
627.TP
628.BI verify_sort \fR=\fPbool
629If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
630read them back in a sorted manner. Default: true.
631.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100632.BI verify_offset \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200633Swap the verification header with data somewhere else in the block before
Aaron Carrolld1429b52007-09-18 08:10:57 +0200634writing. It is swapped back before verifying.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200635.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100636.BI verify_interval \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200637Write the verification header for this number of bytes, which should divide
638\fBblocksize\fR. Default: \fBblocksize\fR.
639.TP
Jens Axboe996093b2010-06-24 08:37:13 +0200640.BI verify_pattern \fR=\fPstr
641If set, fio will fill the io buffers with this pattern. Fio defaults to filling
642with totally random bytes, but sometimes it's interesting to fill with a known
643pattern for io verification purposes. Depending on the width of the pattern,
644fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
645decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
646has to be a hex number that starts with either "0x" or "0X". Use with
647\fBverify\fP=meta.
648.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200649.BI verify_fatal \fR=\fPbool
650If true, exit the job on the first observed verification failure. Default:
651false.
652.TP
Jens Axboee8462bd2009-07-06 12:59:04 +0200653.BI verify_async \fR=\fPint
654Fio will normally verify IO inline from the submitting thread. This option
655takes an integer describing how many async offload threads to create for IO
656verification instead, causing fio to offload the duty of verifying IO contents
Jens Axboec85c3242009-07-06 14:12:57 +0200657to one or more separate threads. If using this offload option, even sync IO
658engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
659allows them to have IO in flight while verifies are running.
Jens Axboee8462bd2009-07-06 12:59:04 +0200660.TP
661.BI verify_async_cpus \fR=\fPstr
662Tell fio to set the given CPU affinity on the async IO verification threads.
663See \fBcpus_allowed\fP for the format used.
664.TP
Jens Axboe6f874182010-06-21 12:53:26 +0200665.BI verify_backlog \fR=\fPint
666Fio will normally verify the written contents of a job that utilizes verify
667once that job has completed. In other words, everything is written then
668everything is read back and verified. You may want to verify continually
669instead for a variety of reasons. Fio stores the meta data associated with an
670IO block in memory, so for large verify workloads, quite a bit of memory would
671be used up holding this meta data. If this option is enabled, fio will verify
672the previously written blocks before continuing to write new ones.
673.TP
674.BI verify_backlog_batch \fR=\fPint
675Control how many blocks fio will verify if verify_backlog is set. If not set,
676will default to the value of \fBverify_backlog\fR (meaning the entire queue is
677read back and verified).
678.TP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200679.B stonewall
Aaron Carrolld1429b52007-09-18 08:10:57 +0200680Wait for preceeding jobs in the job file to exit before starting this one.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200681\fBstonewall\fR implies \fBnew_group\fR.
682.TP
683.B new_group
684Start a new reporting group. If not given, all jobs in a file will be part
685of the same reporting group, unless separated by a stonewall.
686.TP
687.BI numjobs \fR=\fPint
688Number of clones (processes/threads performing the same workload) of this job.
689Default: 1.
690.TP
691.B group_reporting
692If set, display per-group reports instead of per-job when \fBnumjobs\fR is
693specified.
694.TP
695.B thread
696Use threads created with \fBpthread_create\fR\|(3) instead of processes created
697with \fBfork\fR\|(2).
698.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100699.BI zonesize \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200700Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
701.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100702.BI zoneskip \fR=\fPint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200703Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200704read.
705.TP
706.BI write_iolog \fR=\fPstr
707Write the issued I/O patterns to the specified file.
708.TP
709.BI read_iolog \fR=\fPstr
710Replay the I/O patterns contained in the specified file generated by
711\fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
712.TP
Jens Axboe901bb992009-03-14 20:17:36 +0100713.B write_bw_log \fR=\fPstr
714If given, write a bandwidth log of the jobs in this job file. Can be used to
715store data of the bandwidth of the jobs in their lifetime. The included
716fio_generate_plots script uses gnuplot to turn these text files into nice
717graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
718option, the postfix is _bw.log.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200719.TP
720.B write_lat_log
Jens Axboe901bb992009-03-14 20:17:36 +0100721Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
722filename is given with this option, the default filename of "jobname_type.log"
723is used. Even if the filename is given, fio will still append the type of log.
724.TP
725.B disable_clat \fR=\fPbool
726Disable measurements of completion latency numbers. Useful only for cutting
727back the number of calls to gettimeofday, as that does impact performance at
728really high IOPS rates. Note that to really get rid of a large amount of these
729calls, this option must be used with disable_slat and disable_bw as well.
730.TP
731.B disable_slat \fR=\fPbool
732Disable measurements of submission latency numbers. See \fBdisable_clat\fR.
733.TP
734.B disable_bw_measurement \fR=\fPbool
735Disable measurements of throughput/bandwidth numbers. See \fBdisable_clat\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200736.TP
Jens Axboef7fa2652009-03-09 14:20:20 +0100737.BI lockmem \fR=\fPint
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200738Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
739simulate a smaller amount of memory.
740.TP
741.BI exec_prerun \fR=\fPstr
742Before running the job, execute the specified command with \fBsystem\fR\|(3).
743.TP
744.BI exec_postrun \fR=\fPstr
745Same as \fBexec_prerun\fR, but the command is executed after the job completes.
746.TP
747.BI ioscheduler \fR=\fPstr
748Attempt to switch the device hosting the file to the specified I/O scheduler.
749.TP
750.BI cpuload \fR=\fPint
751If the job is a CPU cycle-eater, attempt to use the specified percentage of
752CPU cycles.
753.TP
754.BI cpuchunks \fR=\fPint
755If the job is a CPU cycle-eater, split the load into cycles of the
756given time in milliseconds.
757.TP
758.BI disk_util \fR=\fPbool
Aaron Carrolld1429b52007-09-18 08:10:57 +0200759Generate disk utilization statistics if the platform supports it. Default: true.
Jens Axboe901bb992009-03-14 20:17:36 +0100760.TP
761.BI gtod_reduce \fR=\fPbool
762Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
763disable_bw) plus reduce precision of the timeout somewhat to really shrink the
764gettimeofday() call count. With this option enabled, we only do about 0.4% of
765the gtod() calls we would have done if all time keeping was enabled.
766.TP
767.BI gtod_cpu \fR=\fPint
768Sometimes it's cheaper to dedicate a single thread of execution to just getting
769the current time. Fio (and databases, for instance) are very intensive on
770gettimeofday() calls. With this option, you can set one CPU aside for doing
771nothing but logging current time to a shared memory location. Then the other
772threads/processes that run IO workloads need only copy that segment, instead of
773entering the kernel with a gettimeofday() call. The CPU set aside for doing
774these time calls will be excluded from other uses. Fio will manually clear it
775from the CPU mask of other jobs.
Radha Ramachandranf2bba182009-06-15 08:40:16 +0200776.TP
Jens Axboea696fa22009-12-04 10:05:02 +0100777.BI cgroup \fR=\fPstr
778Add job to this control group. If it doesn't exist, it will be created.
Jens Axboe6adb38a2009-12-07 08:01:26 +0100779The system must have a mounted cgroup blkio mount point for this to work. If
780your system doesn't have it mounted, you can do so with:
781
782# mount -t cgroup -o blkio none /cgroup
Jens Axboea696fa22009-12-04 10:05:02 +0100783.TP
784.BI cgroup_weight \fR=\fPint
785Set the weight of the cgroup to this value. See the documentation that comes
786with the kernel, allowed values are in the range of 100..1000.
Jens Axboee0b0d892009-12-08 10:10:14 +0100787.TP
Vivek Goyal7de87092010-03-31 22:55:15 +0200788.BI cgroup_nodelete \fR=\fPbool
789Normally fio will delete the cgroups it has created after the job completion.
790To override this behavior and to leave cgroups around after the job completion,
791set cgroup_nodelete=1. This can be useful if one wants to inspect various
792cgroup files after job completion. Default: false
793.TP
Jens Axboee0b0d892009-12-08 10:10:14 +0100794.BI uid \fR=\fPint
795Instead of running as the invoking user, set the user ID to this value before
796the thread/process does any work.
797.TP
798.BI gid \fR=\fPint
799Set group ID, see \fBuid\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200800.SH OUTPUT
Aaron Carrolld1429b52007-09-18 08:10:57 +0200801While running, \fBfio\fR will display the status of the created jobs. For
802example:
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200803.RS
Aaron Carrolld1429b52007-09-18 08:10:57 +0200804.P
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200805Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
806.RE
807.P
Aaron Carrolld1429b52007-09-18 08:10:57 +0200808The characters in the first set of brackets denote the current status of each
809threads. The possible values are:
810.P
811.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200812.RS
813.TP
814.B P
815Setup but not started.
816.TP
817.B C
818Thread created.
819.TP
820.B I
821Initialized, waiting.
822.TP
823.B R
824Running, doing sequential reads.
825.TP
826.B r
827Running, doing random reads.
828.TP
829.B W
830Running, doing sequential writes.
831.TP
832.B w
833Running, doing random writes.
834.TP
835.B M
836Running, doing mixed sequential reads/writes.
837.TP
838.B m
839Running, doing mixed random reads/writes.
840.TP
841.B F
842Running, currently waiting for \fBfsync\fR\|(2).
843.TP
844.B V
845Running, verifying written data.
846.TP
847.B E
848Exited, not reaped by main thread.
849.TP
850.B \-
851Exited, thread reaped.
852.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +0200853.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200854.P
855The second set of brackets shows the estimated completion percentage of
856the current group. The third set shows the read and write I/O rate,
857respectively. Finally, the estimated run time of the job is displayed.
858.P
859When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
860for each thread, each group of threads, and each disk, in that order.
861.P
862Per-thread statistics first show the threads client number, group-id, and
863error code. The remaining figures are as follows:
864.RS
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200865.TP
866.B io
867Number of megabytes of I/O performed.
868.TP
869.B bw
870Average data rate (bandwidth).
871.TP
872.B runt
873Threads run time.
874.TP
875.B slat
876Submission latency minimum, maximum, average and standard deviation. This is
877the time it took to submit the I/O.
878.TP
879.B clat
880Completion latency minimum, maximum, average and standard deviation. This
881is the time between submission and completion.
882.TP
883.B bw
884Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
885and standard deviation.
886.TP
887.B cpu
888CPU usage statistics. Includes user and system time, number of context switches
889this thread went through and number of major and minor page faults.
890.TP
891.B IO depths
892Distribution of I/O depths. Each depth includes everything less than (or equal)
893to it, but greater than the previous depth.
894.TP
895.B IO issued
896Number of read/write requests issued, and number of short read/write requests.
897.TP
898.B IO latencies
899Distribution of I/O completion latencies. The numbers follow the same pattern
900as \fBIO depths\fR.
901.RE
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200902.P
903The group statistics show:
Aaron Carrolld1429b52007-09-18 08:10:57 +0200904.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200905.RS
906.TP
907.B io
908Number of megabytes I/O performed.
909.TP
910.B aggrb
911Aggregate bandwidth of threads in the group.
912.TP
913.B minb
914Minimum average bandwidth a thread saw.
915.TP
916.B maxb
917Maximum average bandwidth a thread saw.
918.TP
919.B mint
Aaron Carrolld1429b52007-09-18 08:10:57 +0200920Shortest runtime of threads in the group.
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200921.TP
922.B maxt
923Longest runtime of threads in the group.
924.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +0200925.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200926.P
927Finally, disk statistics are printed with reads first:
Aaron Carrolld1429b52007-09-18 08:10:57 +0200928.PD 0
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200929.RS
930.TP
931.B ios
932Number of I/Os performed by all groups.
933.TP
934.B merge
935Number of merges in the I/O scheduler.
936.TP
937.B ticks
938Number of ticks we kept the disk busy.
939.TP
940.B io_queue
941Total time spent in the disk queue.
942.TP
943.B util
944Disk utilization.
945.RE
Aaron Carrolld1429b52007-09-18 08:10:57 +0200946.PD
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200947.SH TERSE OUTPUT
948If the \fB\-\-minimal\fR option is given, the results will be printed in a
949semicolon-delimited format suitable for scripted use. The fields are:
950.P
951.RS
952.B jobname, groupid, error
953.P
954Read status:
955.RS
Jens Axboeb22989b2009-07-17 22:29:23 +0200956.B KB I/O, bandwidth \fR(KB/s)\fP, runtime \fR(ms)\fP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200957.P
958Submission latency:
959.RS
960.B min, max, mean, standard deviation
961.RE
962Completion latency:
963.RS
964.B min, max, mean, standard deviation
965.RE
966Bandwidth:
967.RS
968.B min, max, aggregate percentage of total, mean, standard deviation
969.RE
970.RE
971.P
972Write status:
973.RS
Jens Axboeb22989b2009-07-17 22:29:23 +0200974.B KB I/O, bandwidth \fR(KB/s)\fP, runtime \fR(ms)\fP
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200975.P
976Submission latency:
977.RS
978.B min, max, mean, standard deviation
979.RE
980Completion latency:
981.RS
982.B min, max, mean, standard deviation
983.RE
984Bandwidth:
985.RS
986.B min, max, aggregate percentage of total, mean, standard deviation
987.RE
988.RE
989.P
Aaron Carrolld1429b52007-09-18 08:10:57 +0200990CPU usage:
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200991.RS
Carl Henrik Lundebd2626f2008-06-12 09:17:46 +0200992.B user, system, context switches, major page faults, minor page faults
Aaron Carrolld60e92d2007-09-17 10:32:59 +0200993.RE
994.P
995IO depth distribution:
996.RS
997.B <=1, 2, 4, 8, 16, 32, >=64
998.RE
999.P
1000IO latency distribution (ms):
1001.RS
1002.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000
1003.RE
1004.P
1005.B text description
1006.RE
1007.SH AUTHORS
1008.B fio
Jens Axboeaa58d252010-06-09 09:49:38 +02001009was written by Jens Axboe <jens.axboe@oracle.com>,
1010now Jens Axboe <jaxboe@fusionio.com>.
Aaron Carrolld1429b52007-09-18 08:10:57 +02001011.br
1012This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001013on documentation by Jens Axboe.
1014.SH "REPORTING BUGS"
Jens Axboe482900c2009-06-02 12:15:51 +02001015Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
Aaron Carrolld1429b52007-09-18 08:10:57 +02001016See \fBREADME\fR.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001017.SH "SEE ALSO"
Aaron Carrolld1429b52007-09-18 08:10:57 +02001018For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1019.br
1020Sample jobfiles are available in the \fBexamples\fR directory.
Aaron Carrolld60e92d2007-09-17 10:32:59 +02001021