->getevents() should take unsigned args
[fio.git] / fio.1
CommitLineData
d60e92d1
AC
1.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
d1429b5c 28Print statistics in a terse, semicolon-delimited format.
d60e92d1
AC
29.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
40.BI \-\-cmdhelp \fR=\fPcommand
41Print help information for \fIcommand\fR. May be `all' for all commands.
42.TP
43.B \-\-help
44Display usage information and exit.
45.TP
46.B \-\-version
47Display version information and exit.
48.SH "JOB FILE FORMAT"
49Job files are in `ini' format. They consist of one or more
50job definitions, which begin with a job name in square brackets and
51extend to the next job name. The job name can be any ASCII string
52except `global', which has a special meaning. Following the job name is
53a sequence of zero or more parameters, one per line, that define the
54behavior of the job. Any line starting with a `;' or `#' character is
d1429b5c 55considered a comment and ignored.
d9956b64
AC
56.P
57If \fIjobfile\fR is specified as `-', the job file will be read from
58standard input.
d60e92d1
AC
59.SS "Global Section"
60The global section contains default parameters for jobs specified in the
61job file. A job is only affected by global sections residing above it,
62and there may be any number of global sections. Specific job definitions
63may override any parameter set in global sections.
64.SH "JOB PARAMETERS"
65.SS Types
66Some parameters may take arguments of a specific type. The types used are:
67.TP
68.I str
69String: a sequence of alphanumeric characters.
70.TP
71.I int
72Integer: a whole number, possibly negative. If prefixed with `0x', the value
73is assumed to be base 16 (hexadecimal).
74.TP
75.I siint
76SI integer: a whole number, possibly containing a suffix denoting the base unit
77of the value. Accepted suffixes are `k', 'M' and 'G', denoting kilo (1024),
78mega (1024*1024) and giga (1024*1024*1024) respectively.
79.TP
80.I bool
81Boolean: a true or false value. `0' denotes false, `1' denotes true.
82.TP
83.I irange
84Integer range: a range of integers specified in the format
d1429b5c
AC
85\fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
86\fIupper\fR may contain a suffix as described above. If an option allows two
87sets of ranges, they are separated with a `,' or `/' character. For example:
88`8\-8k/8M\-4G'.
d60e92d1
AC
89.SS "Parameter List"
90.TP
91.BI name \fR=\fPstr
d9956b64 92May be used to override the job name. On the command line, this parameter
d60e92d1
AC
93has the special purpose of signalling the start of a new job.
94.TP
95.BI description \fR=\fPstr
96Human-readable description of the job. It is printed when the job is run, but
97otherwise has no special purpose.
98.TP
99.BI directory \fR=\fPstr
100Prefix filenames with this directory. Used to place files in a location other
101than `./'.
102.TP
103.BI filename \fR=\fPstr
104.B fio
105normally makes up a file name based on the job name, thread number, and file
d1429b5c
AC
106number. If you want to share files between threads in a job or several jobs,
107specify a \fIfilename\fR for each of them to override the default. If the I/O
108engine used is `net', \fIfilename\fR is the host and port to connect to in the
109format \fIhost\fR/\fIport\fR. If the I/O engine is file-based, you can specify
110a number of files by separating the names with a `:' character. `\-' is a
111reserved name, meaning stdin or stdout, depending on the read/write direction
112set.
d60e92d1
AC
113.TP
114.BI opendir \fR=\fPstr
115Recursively open any files below directory \fIstr\fR.
116.TP
117.BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
118Type of I/O pattern. Accepted values are:
119.RS
120.RS
121.TP
122.B read
d1429b5c 123Sequential reads.
d60e92d1
AC
124.TP
125.B write
d1429b5c 126Sequential writes.
d60e92d1
AC
127.TP
128.B randread
d1429b5c 129Random reads.
d60e92d1
AC
130.TP
131.B randwrite
d1429b5c 132Random writes.
d60e92d1
AC
133.TP
134.B rw
d1429b5c 135Mixed sequential reads and writes.
d60e92d1
AC
136.TP
137.B randrw
d1429b5c 138Mixed random reads and writes.
d60e92d1
AC
139.RE
140.P
d1429b5c
AC
141For mixed I/O, the default split is 50/50. For random I/O, the number of I/Os
142to perform before getting a new offset can be specified by appending
143`:\fIint\fR' to the pattern type. The default is 1.
d60e92d1
AC
144.RE
145.TP
146.BI randrepeat \fR=\fPbool
147Seed the random number generator in a predictable way so results are repeatable
d1429b5c 148across runs. Default: true.
d60e92d1
AC
149.TP
150.BI fadvise_hint \fR=\fPbool
d1429b5c
AC
151Disable use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
152are likely to be issued. Default: true.
d60e92d1
AC
153.TP
154.BI size \fR=\fPsiint
155Total size of I/O for this job. \fBfio\fR will run until this many bytes have
156been transfered, unless limited by other options (\fBruntime\fR, for instance).
157Unless \fBnr_files\fR and \fBfilesize\fR options are given, this amount will be
158divided between the available files for the job.
159.TP
160.BI filesize \fR=\fPirange
161Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
d1429b5c
AC
162for files at random within the given range, limited to \fBsize\fR in total (if
163that is given). If \fBfilesize\fR is not specified, each created file is the
164same size.
d60e92d1
AC
165.TP
166.BI blocksize \fR=\fPsiint "\fR,\fB bs" \fR=\fPsiint
167Block size for I/O units. Default: 4k. Values for reads and writes can be
168specified seperately in the format \fIread\fR,\fIwrite\fR, either of
169which may be empty to leave that value at its default.
170.TP
171.BI blocksize_range \fR=\fPirange "\fR,\fB bsrange" \fR=\fPirange
d1429b5c
AC
172Specify a range of I/O block sizes. The issued I/O unit will always be a
173multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
174to both reads and writes, but can be specified seperately (see \fBblocksize\fR).
d60e92d1
AC
175.TP
176.B blocksize_unaligned\fR,\fP bs_unaligned
d1429b5c
AC
177If set, any size in \fBblocksize_range\fR may be used. This typically won't
178work with direct I/O, as that normally requires sector alignment.
d60e92d1
AC
179.TP
180.B zero_buffers
181Initialise buffers with all zeros. Default: fill buffers with random data.
182.TP
183.BI nrfiles \fR=\fPint
184Number of files to use for this job. Default: 1.
185.TP
186.BI openfiles \fR=\fPint
187Number of files to keep open at the same time. Default: \fBnrfiles\fR.
188.TP
189.BI file_service_type \fR=\fPstr
190Defines how files to service are selected. The following types are defined:
191.RS
192.RS
193.TP
194.B random
195Choose a file at random
196.TP
197.B roundrobin
198Round robin over open files (default).
199.RE
200.P
201The number of I/Os to issue before switching a new file can be specified by
202appending `:\fIint\fR' to the service type.
203.RE
204.TP
205.BI ioengine \fR=\fPstr
206Defines how the job issues I/O. The following types are defined:
207.RS
208.RS
209.TP
210.B sync
211Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
212position the I/O location.
213.TP
a31041ea 214.B psync
215Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
216.TP
d60e92d1
AC
217.B libaio
218Linux native asynchronous I/O.
219.TP
220.B posixaio
221glibc POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
222.TP
223.B mmap
d1429b5c
AC
224File is memory mapped with \fImmap\fR\|(2) and data copied using
225\fImemcpy\fR\|(3).
d60e92d1
AC
226.TP
227.B splice
d1429b5c
AC
228\fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
229transfer data from user-space to the kernel.
d60e92d1
AC
230.TP
231.B syslet-rw
232Use the syslet system calls to make regular read/write asynchronous.
233.TP
234.B sg
235SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
d1429b5c
AC
236the target is an sg character device, we use \fIread\fR\|(2) and
237\fIwrite\fR\|(2) for asynchronous I/O.
d60e92d1
AC
238.TP
239.B null
240Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
241itself and for debugging and testing purposes.
242.TP
243.B net
244Transfer over the network. \fBfilename\fR must be set appropriately to
245`\fIhost\fR/\fIport\fR' regardless of data direction. If receiving, only the
246\fIport\fR argument is used.
247.TP
248.B netsplice
249Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
250and send/receive.
251.TP
53aec0a4 252.B cpuio
d60e92d1
AC
253Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
254\fBcpucycles\fR parameters.
255.TP
256.B guasi
257The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
258approach to asycnronous I/O.
d1429b5c
AC
259.br
260See <http://www.xmailserver.org/guasi\-lib.html>.
d60e92d1
AC
261.TP
262.B external
263Loads an external I/O engine object file. Append the engine filename as
264`:\fIenginepath\fR'.
265.RE
266.RE
267.TP
268.BI iodepth \fR=\fPint
269Number of I/O units to keep in flight against the file. Default: 1.
270.TP
271.BI iodepth_batch \fR=\fPint
272Number of I/Os to submit at once. Default: \fBiodepth\fR.
273.TP
274.BI iodepth_low \fR=\fPint
275Low watermark indicating when to start filling the queue again. Default:
276\fBiodepth\fR.
277.TP
278.BI direct \fR=\fPbool
279If true, use non-buffered I/O (usually O_DIRECT). Default: false.
280.TP
281.BI buffered \fR=\fPbool
282If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
283Default: true.
284.TP
285.BI offset \fR=\fPsiint
286Offset in the file to start I/O. Data before the offset will not be touched.
287.TP
288.BI fsync \fR=\fPint
d1429b5c
AC
289How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
2900, don't sync. Default: 0.
d60e92d1
AC
291.TP
292.BI overwrite \fR=\fPbool
d1429b5c 293If writing, setup the file first and do overwrites. Default: false.
d60e92d1
AC
294.TP
295.BI end_fsync \fR=\fPbool
d1429b5c 296Sync file contents when job exits. Default: false.
d60e92d1
AC
297.TP
298.BI fsync_on_close \fR=\fPbool
299If true, sync file contents on close. This differs from \fBend_fsync\fR in that
d1429b5c 300it will happen on every close, not just at the end of the job. Default: false.
d60e92d1
AC
301.TP
302.BI rwmixcycle \fR=\fPint
303How many milliseconds before switching between reads and writes for a mixed
304workload. Default: 500ms.
305.TP
306.BI rwmixread \fR=\fPint
307Percentage of a mixed workload that should be reads. Default: 50.
308.TP
309.BI rwmixwrite \fR=\fPint
d1429b5c 310Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
d60e92d1 311\fBwrmixwrite\fR are given and do not sum to 100%, the latter of the two
d1429b5c 312overrides the first. Default: 50.
d60e92d1
AC
313.TP
314.B norandommap
315Normally \fBfio\fR will cover every block of the file when doing random I/O. If
316this parameter is given, a new offset will be chosen without looking at past
317I/O history. This parameter is mutually exclusive with \fBverify\fR.
318.TP
319.BI nice \fR=\fPint
320Run job with given nice value. See \fInice\fR\|(2).
321.TP
322.BI prio \fR=\fPint
323Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
324\fIionice\fR\|(1).
325.TP
326.BI prioclass \fR=\fPint
327Set I/O priority class. See \fIionice\fR\|(1).
328.TP
329.BI thinktime \fR=\fPint
330Stall job for given number of microseconds between issuing I/Os.
331.TP
332.BI thinktime_spin \fR=\fPint
333Pretend to spend CPU time for given number of microseconds, sleeping the rest
334of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
335.TP
336.BI thinktime_blocks \fR=\fPint
337Number of blocks to issue before waiting \fBthinktime\fR microseconds.
338Default: 1.
339.TP
340.BI rate \fR=\fPint
341Cap bandwidth used by this job to this number of KiB/s.
342.TP
343.BI ratemin \fR=\fPint
344Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
345Failing to meet this requirement will cause the job to exit.
346.TP
347.BI rate_iops \fR=\fPint
348Cap the bandwidth to this number of IOPS. If \fBblocksize\fR is a range, the
349smallest block size is used as the metric.
350.TP
351.BI rate_iops_min \fR=\fPint
352If this rate of I/O is not met, the job will exit.
353.TP
354.BI ratecycle \fR=\fPint
355Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
356milliseconds. Default: 1000ms.
357.TP
358.BI cpumask \fR=\fPint
359Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
360may run on. See \fBsched_setaffinity\fR\|(2).
361.TP
362.BI cpus_allowed \fR=\fPstr
363Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
364.TP
365.BI startdelay \fR=\fPint
366Delay start of job for the specified number of seconds.
367.TP
368.BI runtime \fR=\fPint
369Terminate processing after the specified number of seconds.
370.TP
371.B time_based
372If given, run for the specified \fBruntime\fR duration even if the files are
373completely read or written. The same workload will be repeated as many times
374as \fBruntime\fR allows.
375.TP
376.BI invalidate \fR=\fPbool
377Invalidate buffer-cache for the file prior to starting I/O. Default: true.
378.TP
379.BI sync \fR=\fPbool
380Use synchronous I/O for buffered writes. For the majority of I/O engines,
d1429b5c 381this means using O_SYNC. Default: false.
d60e92d1
AC
382.TP
383.BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
384Allocation method for I/O unit buffer. Allowed values are:
385.RS
386.RS
387.TP
388.B malloc
389Allocate memory with \fImalloc\fR\|(3).
390.TP
391.B shm
392Use shared memory buffers allocated through \fIshmget\fR\|(2).
393.TP
394.B shmhuge
395Same as \fBshm\fR, but use huge pages as backing.
396.TP
397.B mmap
398Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
399is given after the option in the format `:\fIfile\fR'.
400.TP
401.B mmaphuge
402Same as \fBmmap\fR, but use huge files as backing.
403.RE
404.P
405The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
406job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
407the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
408have hugetlbfs mounted, and \fIfile\fR must point there.
409.RE
410.TP
d1429b5c 411.BI hugepage\-size \fR=\fPsiint
d60e92d1
AC
412Defines the size of a huge page. Must be at least equal to the system setting.
413Should be a multiple of 1MiB. Default: 4MiB.
414.TP
415.B exitall
416Terminate all jobs when one finishes. Default: wait for each job to finish.
417.TP
418.BI bwavgtime \fR=\fPint
419Average bandwidth calculations over the given time in milliseconds. Default:
420500ms.
421.TP
422.BI create_serialize \fR=\fPbool
d1429b5c 423If true, serialize file creation for the jobs. Default: true.
d60e92d1
AC
424.TP
425.BI create_fsync \fR=\fPbool
426\fIfsync\fR\|(2) data file after creation. Default: true.
427.TP
428.BI unlink \fR=\fPbool
429Unlink job files when done. Default: false.
430.TP
431.BI loops \fR=\fPint
432Specifies the number of iterations (runs of the same workload) of this job.
433Default: 1.
434.TP
435.BI do_verify \fR=\fPbool
436Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
437Default: true.
438.TP
439.BI verify \fR=\fPstr
440Method of verifying file contents after each iteration of the job. Allowed
441values are:
442.RS
443.RS
444.TP
445.B md5 crc16 crc32 crc64 crc7 sha256 sha512
446Store appropriate checksum in the header of each block.
447.TP
448.B meta
449Write extra information about each I/O (timestamp, block number, etc.). The
450block number is verified.
451.TP
452.B pattern
453Fill I/O buffers with a specific pattern that is used to verify. The pattern is
454specified by appending `:\fIint\fR' to the parameter. \fIint\fR cannot be larger
455than 32-bits.
456.TP
457.B null
458Pretend to verify. Used for testing internals.
459.RE
460.RE
461.TP
462.BI verify_sort \fR=\fPbool
463If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
464read them back in a sorted manner. Default: true.
465.TP
466.BI verify_offset \fR=\fPsiint
467Swap the verification header with data somewhere else in the block before
d1429b5c 468writing. It is swapped back before verifying.
d60e92d1
AC
469.TP
470.BI verify_interval \fR=\fPsiint
471Write the verification header for this number of bytes, which should divide
472\fBblocksize\fR. Default: \fBblocksize\fR.
473.TP
474.BI verify_fatal \fR=\fPbool
475If true, exit the job on the first observed verification failure. Default:
476false.
477.TP
478.B stonewall
d1429b5c 479Wait for preceeding jobs in the job file to exit before starting this one.
d60e92d1
AC
480\fBstonewall\fR implies \fBnew_group\fR.
481.TP
482.B new_group
483Start a new reporting group. If not given, all jobs in a file will be part
484of the same reporting group, unless separated by a stonewall.
485.TP
486.BI numjobs \fR=\fPint
487Number of clones (processes/threads performing the same workload) of this job.
488Default: 1.
489.TP
490.B group_reporting
491If set, display per-group reports instead of per-job when \fBnumjobs\fR is
492specified.
493.TP
494.B thread
495Use threads created with \fBpthread_create\fR\|(3) instead of processes created
496with \fBfork\fR\|(2).
497.TP
498.BI zonesize \fR=\fPsiint
499Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
500.TP
501.BI zoneskip \fR=\fPsiint
d1429b5c 502Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
d60e92d1
AC
503read.
504.TP
505.BI write_iolog \fR=\fPstr
506Write the issued I/O patterns to the specified file.
507.TP
508.BI read_iolog \fR=\fPstr
509Replay the I/O patterns contained in the specified file generated by
510\fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
511.TP
512.B write_bw_log
513If given, write bandwidth logs of the jobs in this file.
514.TP
515.B write_lat_log
516Same as \fBwrite_bw_log\fR, but writes I/O completion latencies.
517.TP
518.BI lockmem \fR=\fPsiint
519Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
520simulate a smaller amount of memory.
521.TP
522.BI exec_prerun \fR=\fPstr
523Before running the job, execute the specified command with \fBsystem\fR\|(3).
524.TP
525.BI exec_postrun \fR=\fPstr
526Same as \fBexec_prerun\fR, but the command is executed after the job completes.
527.TP
528.BI ioscheduler \fR=\fPstr
529Attempt to switch the device hosting the file to the specified I/O scheduler.
530.TP
531.BI cpuload \fR=\fPint
532If the job is a CPU cycle-eater, attempt to use the specified percentage of
533CPU cycles.
534.TP
535.BI cpuchunks \fR=\fPint
536If the job is a CPU cycle-eater, split the load into cycles of the
537given time in milliseconds.
538.TP
539.BI disk_util \fR=\fPbool
d1429b5c 540Generate disk utilization statistics if the platform supports it. Default: true.
d60e92d1 541.SH OUTPUT
d1429b5c
AC
542While running, \fBfio\fR will display the status of the created jobs. For
543example:
d60e92d1 544.RS
d1429b5c 545.P
d60e92d1
AC
546Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
547.RE
548.P
d1429b5c
AC
549The characters in the first set of brackets denote the current status of each
550threads. The possible values are:
551.P
552.PD 0
d60e92d1
AC
553.RS
554.TP
555.B P
556Setup but not started.
557.TP
558.B C
559Thread created.
560.TP
561.B I
562Initialized, waiting.
563.TP
564.B R
565Running, doing sequential reads.
566.TP
567.B r
568Running, doing random reads.
569.TP
570.B W
571Running, doing sequential writes.
572.TP
573.B w
574Running, doing random writes.
575.TP
576.B M
577Running, doing mixed sequential reads/writes.
578.TP
579.B m
580Running, doing mixed random reads/writes.
581.TP
582.B F
583Running, currently waiting for \fBfsync\fR\|(2).
584.TP
585.B V
586Running, verifying written data.
587.TP
588.B E
589Exited, not reaped by main thread.
590.TP
591.B \-
592Exited, thread reaped.
593.RE
d1429b5c 594.PD
d60e92d1
AC
595.P
596The second set of brackets shows the estimated completion percentage of
597the current group. The third set shows the read and write I/O rate,
598respectively. Finally, the estimated run time of the job is displayed.
599.P
600When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
601for each thread, each group of threads, and each disk, in that order.
602.P
603Per-thread statistics first show the threads client number, group-id, and
604error code. The remaining figures are as follows:
605.RS
d60e92d1
AC
606.TP
607.B io
608Number of megabytes of I/O performed.
609.TP
610.B bw
611Average data rate (bandwidth).
612.TP
613.B runt
614Threads run time.
615.TP
616.B slat
617Submission latency minimum, maximum, average and standard deviation. This is
618the time it took to submit the I/O.
619.TP
620.B clat
621Completion latency minimum, maximum, average and standard deviation. This
622is the time between submission and completion.
623.TP
624.B bw
625Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
626and standard deviation.
627.TP
628.B cpu
629CPU usage statistics. Includes user and system time, number of context switches
630this thread went through and number of major and minor page faults.
631.TP
632.B IO depths
633Distribution of I/O depths. Each depth includes everything less than (or equal)
634to it, but greater than the previous depth.
635.TP
636.B IO issued
637Number of read/write requests issued, and number of short read/write requests.
638.TP
639.B IO latencies
640Distribution of I/O completion latencies. The numbers follow the same pattern
641as \fBIO depths\fR.
642.RE
d60e92d1
AC
643.P
644The group statistics show:
d1429b5c 645.PD 0
d60e92d1
AC
646.RS
647.TP
648.B io
649Number of megabytes I/O performed.
650.TP
651.B aggrb
652Aggregate bandwidth of threads in the group.
653.TP
654.B minb
655Minimum average bandwidth a thread saw.
656.TP
657.B maxb
658Maximum average bandwidth a thread saw.
659.TP
660.B mint
d1429b5c 661Shortest runtime of threads in the group.
d60e92d1
AC
662.TP
663.B maxt
664Longest runtime of threads in the group.
665.RE
d1429b5c 666.PD
d60e92d1
AC
667.P
668Finally, disk statistics are printed with reads first:
d1429b5c 669.PD 0
d60e92d1
AC
670.RS
671.TP
672.B ios
673Number of I/Os performed by all groups.
674.TP
675.B merge
676Number of merges in the I/O scheduler.
677.TP
678.B ticks
679Number of ticks we kept the disk busy.
680.TP
681.B io_queue
682Total time spent in the disk queue.
683.TP
684.B util
685Disk utilization.
686.RE
d1429b5c 687.PD
d60e92d1
AC
688.SH TERSE OUTPUT
689If the \fB\-\-minimal\fR option is given, the results will be printed in a
690semicolon-delimited format suitable for scripted use. The fields are:
691.P
692.RS
693.B jobname, groupid, error
694.P
695Read status:
696.RS
697.B KiB I/O, bandwidth \fR(KiB/s)\fP, runtime \fR(ms)\fP
698.P
699Submission latency:
700.RS
701.B min, max, mean, standard deviation
702.RE
703Completion latency:
704.RS
705.B min, max, mean, standard deviation
706.RE
707Bandwidth:
708.RS
709.B min, max, aggregate percentage of total, mean, standard deviation
710.RE
711.RE
712.P
713Write status:
714.RS
715.B KiB I/O, bandwidth \fR(KiB/s)\fP, runtime \fR(ms)\fP
716.P
717Submission latency:
718.RS
719.B min, max, mean, standard deviation
720.RE
721Completion latency:
722.RS
723.B min, max, mean, standard deviation
724.RE
725Bandwidth:
726.RS
727.B min, max, aggregate percentage of total, mean, standard deviation
728.RE
729.RE
730.P
d1429b5c 731CPU usage:
d60e92d1
AC
732.RS
733.B user, system, context switches
734.RE
735.P
736IO depth distribution:
737.RS
738.B <=1, 2, 4, 8, 16, 32, >=64
739.RE
740.P
741IO latency distribution (ms):
742.RS
743.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000
744.RE
745.P
746.B text description
747.RE
748.SH AUTHORS
749.B fio
750was written by Jens Axboe <jens.axboe@oracle.com>.
d1429b5c
AC
751.br
752This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
d60e92d1
AC
753on documentation by Jens Axboe.
754.SH "REPORTING BUGS"
d1429b5c
AC
755Report bugs to the \fBfio\fR mailing list <fio-devel@kernel.dk>.
756See \fBREADME\fR.
d60e92d1 757.SH "SEE ALSO"
d1429b5c
AC
758For further documentation see \fBHOWTO\fR and \fBREADME\fR.
759.br
760Sample jobfiles are available in the \fBexamples\fR directory.
d60e92d1 761