Add create_only option
[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
49da1240
JA
15.BI \-\-debug \fR=\fPtype
16Enable verbose tracing of various fio actions. May be `all' for all types
17or individual types separated by a comma (eg \-\-debug=io,file). `help' will
18list all available tracing options.
19.TP
d60e92d1
AC
20.BI \-\-output \fR=\fPfilename
21Write output to \fIfilename\fR.
22.TP
23.BI \-\-timeout \fR=\fPtimeout
24Limit run time to \fItimeout\fR seconds.
25.TP
26.B \-\-latency\-log
27Generate per-job latency logs.
28.TP
29.B \-\-bandwidth\-log
30Generate per-job bandwidth logs.
31.TP
32.B \-\-minimal
d1429b5c 33Print statistics in a terse, semicolon-delimited format.
d60e92d1 34.TP
49da1240
JA
35.B \-\-version
36Display version information and exit.
37.TP
065248bf 38.BI \-\-terse\-version \fR=\fPversion
4d658652 39Set terse version output format (Current version 3, or older version 2).
49da1240
JA
40.TP
41.B \-\-help
42Display usage information and exit.
43.TP
44.BI \-\-cmdhelp \fR=\fPcommand
45Print help information for \fIcommand\fR. May be `all' for all commands.
46.TP
de890a1e
SL
47.BI \-\-enghelp \fR=\fPioengine[,command]
48List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR.
49.TP
d60e92d1
AC
50.BI \-\-showcmd \fR=\fPjobfile
51Convert \fIjobfile\fR to a set of command-line options.
52.TP
53.B \-\-readonly
54Enable read-only safety checks.
55.TP
56.BI \-\-eta \fR=\fPwhen
57Specifies when real-time ETA estimate should be printed. \fIwhen\fR may
58be one of `always', `never' or `auto'.
59.TP
49da1240
JA
60.BI \-\-readonly
61Turn on safety read-only checks, preventing any attempted write.
62.TP
c0a5d35e 63.BI \-\-section \fR=\fPsec
49da1240 64Only run section \fIsec\fR from job file. Multiple of these options can be given, adding more sections to run.
c0a5d35e 65.TP
49da1240
JA
66.BI \-\-alloc\-size \fR=\fPkb
67Set the internal smalloc pool size to \fIkb\fP kilobytes.
d60e92d1 68.TP
49da1240
JA
69.BI \-\-warnings\-fatal
70All fio parser warnings are fatal, causing fio to exit with an error.
9183788d 71.TP
49da1240
JA
72.BI \-\-max\-jobs \fR=\fPnr
73Set the maximum allowed number of jobs (threads/processes) to suport.
d60e92d1 74.TP
49da1240
JA
75.BI \-\-server \fR=\fPargs
76Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section.
f57a9c59 77.TP
49da1240
JA
78.BI \-\-daemonize \fR=\fPpidfile
79Background a fio server, writing the pid to the given pid file.
80.TP
81.BI \-\-client \fR=\fPhost
82Instead of running the jobs locally, send and run them on the given host.
d60e92d1
AC
83.SH "JOB FILE FORMAT"
84Job files are in `ini' format. They consist of one or more
85job definitions, which begin with a job name in square brackets and
86extend to the next job name. The job name can be any ASCII string
87except `global', which has a special meaning. Following the job name is
88a sequence of zero or more parameters, one per line, that define the
89behavior of the job. Any line starting with a `;' or `#' character is
d1429b5c 90considered a comment and ignored.
d9956b64
AC
91.P
92If \fIjobfile\fR is specified as `-', the job file will be read from
93standard input.
d60e92d1
AC
94.SS "Global Section"
95The global section contains default parameters for jobs specified in the
96job file. A job is only affected by global sections residing above it,
97and there may be any number of global sections. Specific job definitions
98may override any parameter set in global sections.
99.SH "JOB PARAMETERS"
100.SS Types
101Some parameters may take arguments of a specific type. The types used are:
102.TP
103.I str
104String: a sequence of alphanumeric characters.
105.TP
106.I int
d60e92d1 107SI integer: a whole number, possibly containing a suffix denoting the base unit
b09da8fa
JA
108of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
109kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
110respectively. The suffix is not case sensitive. If prefixed with '0x', the
5982a925
MS
111value is assumed to be base 16 (hexadecimal). A suffix may include a trailing 'b',
112for instance 'kb' is identical to 'k'. You can specify a base 10 value
57fc29fa
JA
113by using 'KiB', 'MiB', 'GiB', etc. This is useful for disk drives where
114values are often given in base 10 values. Specifying '30GiB' will get you
11530*1000^3 bytes.
d60e92d1
AC
116.TP
117.I bool
118Boolean: a true or false value. `0' denotes false, `1' denotes true.
119.TP
120.I irange
121Integer range: a range of integers specified in the format
d1429b5c
AC
122\fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
123\fIupper\fR may contain a suffix as described above. If an option allows two
124sets of ranges, they are separated with a `,' or `/' character. For example:
125`8\-8k/8M\-4G'.
83349190
YH
126.TP
127.I float_list
128List of floating numbers: A list of floating numbers, separated by
129a ':' charcater.
d60e92d1
AC
130.SS "Parameter List"
131.TP
132.BI name \fR=\fPstr
d9956b64 133May be used to override the job name. On the command line, this parameter
d60e92d1
AC
134has the special purpose of signalling the start of a new job.
135.TP
136.BI description \fR=\fPstr
137Human-readable description of the job. It is printed when the job is run, but
138otherwise has no special purpose.
139.TP
140.BI directory \fR=\fPstr
141Prefix filenames with this directory. Used to place files in a location other
142than `./'.
143.TP
144.BI filename \fR=\fPstr
145.B fio
146normally makes up a file name based on the job name, thread number, and file
d1429b5c 147number. If you want to share files between threads in a job or several jobs,
de890a1e
SL
148specify a \fIfilename\fR for each of them to override the default.
149If the I/O engine is file-based, you can specify
d1429b5c
AC
150a number of files by separating the names with a `:' character. `\-' is a
151reserved name, meaning stdin or stdout, depending on the read/write direction
152set.
d60e92d1 153.TP
3ce9dcaf
JA
154.BI lockfile \fR=\fPstr
155Fio defaults to not locking any files before it does IO to them. If a file or
156file descriptor is shared, fio can serialize IO to that file to make the end
157result consistent. This is usual for emulating real workloads that share files.
158The lock modes are:
159.RS
160.RS
161.TP
162.B none
163No locking. This is the default.
164.TP
165.B exclusive
166Only one thread or process may do IO at the time, excluding all others.
167.TP
168.B readwrite
169Read-write locking on the file. Many readers may access the file at the same
170time, but writes get exclusive access.
171.RE
172.P
173The option may be post-fixed with a lock batch number. If set, then each
174thread/process may do that amount of IOs to the file before giving up the lock.
175Since lock acquisition is expensive, batching the lock/unlocks will speed up IO.
176.RE
177.P
d60e92d1
AC
178.BI opendir \fR=\fPstr
179Recursively open any files below directory \fIstr\fR.
180.TP
181.BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
182Type of I/O pattern. Accepted values are:
183.RS
184.RS
185.TP
186.B read
d1429b5c 187Sequential reads.
d60e92d1
AC
188.TP
189.B write
d1429b5c 190Sequential writes.
d60e92d1
AC
191.TP
192.B randread
d1429b5c 193Random reads.
d60e92d1
AC
194.TP
195.B randwrite
d1429b5c 196Random writes.
d60e92d1 197.TP
10b023db 198.B rw, readwrite
d1429b5c 199Mixed sequential reads and writes.
d60e92d1
AC
200.TP
201.B randrw
d1429b5c 202Mixed random reads and writes.
d60e92d1
AC
203.RE
204.P
38dad62d
JA
205For mixed I/O, the default split is 50/50. For certain types of io the result
206may still be skewed a bit, since the speed may be different. It is possible to
207specify a number of IO's to do before getting a new offset, this is one by
208appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
209would look like \fBrw=randread:8\fR for passing in an offset modifier with a
059b0802
JA
210value of 8. If the postfix is used with a sequential IO pattern, then the value
211specified will be added to the generated offset for each IO. For instance,
212using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
213into sequential IO with holes. See the \fBrw_sequencer\fR option.
d60e92d1
AC
214.RE
215.TP
38dad62d
JA
216.BI rw_sequencer \fR=\fPstr
217If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
218then this option controls how that number modifies the IO offset being
219generated. Accepted values are:
220.RS
221.RS
222.TP
223.B sequential
224Generate sequential offset
225.TP
226.B identical
227Generate the same offset
228.RE
229.P
230\fBsequential\fR is only useful for random IO, where fio would normally
231generate a new random offset for every IO. If you append eg 8 to randread, you
232would get a new random offset for every 8 IO's. The result would be a seek for
233only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
234that. As sequential IO is already sequential, setting \fBsequential\fR for that
235would not result in any differences. \fBidentical\fR behaves in a similar
236fashion, except it sends the same offset 8 number of times before generating a
237new offset.
238.RE
239.P
240.TP
90fef2d1
JA
241.BI kb_base \fR=\fPint
242The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
243manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
244reasons. Allow values are 1024 or 1000, with 1024 being the default.
245.TP
d60e92d1
AC
246.BI randrepeat \fR=\fPbool
247Seed the random number generator in a predictable way so results are repeatable
d1429b5c 248across runs. Default: true.
d60e92d1 249.TP
2615cc4b
JA
250.BI use_os_rand \fR=\fPbool
251Fio can either use the random generator supplied by the OS to generator random
252offsets, or it can use it's own internal generator (based on Tausworthe).
253Default is to use the internal generator, which is often of better quality and
254faster. Default: false.
255.TP
a596f047
EG
256.BI fallocate \fR=\fPstr
257Whether pre-allocation is performed when laying down files. Accepted values
258are:
259.RS
260.RS
261.TP
262.B none
263Do not pre-allocate space.
264.TP
265.B posix
266Pre-allocate via posix_fallocate().
267.TP
268.B keep
269Pre-allocate via fallocate() with FALLOC_FL_KEEP_SIZE set.
270.TP
271.B 0
272Backward-compatible alias for 'none'.
273.TP
274.B 1
275Backward-compatible alias for 'posix'.
276.RE
277.P
278May not be available on all supported platforms. 'keep' is only
279available on Linux. If using ZFS on Solaris this must be set to 'none'
280because ZFS doesn't support it. Default: 'posix'.
281.RE
7bc8c2cf 282.TP
d60e92d1 283.BI fadvise_hint \fR=\fPbool
23a7b043 284Use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
d1429b5c 285are likely to be issued. Default: true.
d60e92d1 286.TP
f7fa2653 287.BI size \fR=\fPint
d60e92d1
AC
288Total size of I/O for this job. \fBfio\fR will run until this many bytes have
289been transfered, unless limited by other options (\fBruntime\fR, for instance).
d7c8be03 290Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
d6667268
JA
291divided between the available files for the job. If not set, fio will use the
292full size of the given files or devices. If the the files do not exist, size
7bb59102
JA
293must be given. It is also possible to give size as a percentage between 1 and
294100. If size=20% is given, fio will use 20% of the full size of the given files
295or devices.
d60e92d1 296.TP
74586c1e 297.BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
3ce9dcaf
JA
298Sets size to something really large and waits for ENOSPC (no space left on
299device) as the terminating condition. Only makes sense with sequential write.
300For a read workload, the mount point will be filled first then IO started on
4f12432e
JA
301the result. This option doesn't make sense if operating on a raw device node,
302since the size of that is already known by the file system. Additionally,
303writing beyond end-of-device will not return ENOSPC there.
3ce9dcaf 304.TP
d60e92d1
AC
305.BI filesize \fR=\fPirange
306Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
d1429b5c
AC
307for files at random within the given range, limited to \fBsize\fR in total (if
308that is given). If \fBfilesize\fR is not specified, each created file is the
309same size.
d60e92d1 310.TP
f7fa2653 311.BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
d60e92d1 312Block size for I/O units. Default: 4k. Values for reads and writes can be
656ebab7 313specified separately in the format \fIread\fR,\fIwrite\fR, either of
d60e92d1
AC
314which may be empty to leave that value at its default.
315.TP
9183788d 316.BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
d1429b5c
AC
317Specify a range of I/O block sizes. The issued I/O unit will always be a
318multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
9183788d 319to both reads and writes if only one range is given, but can be specified
656ebab7 320separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
9183788d
JA
321Also (see \fBblocksize\fR).
322.TP
323.BI bssplit \fR=\fPstr
324This option allows even finer grained control of the block sizes issued,
325not just even splits between them. With this option, you can weight various
326block sizes for exact control of the issued IO for a job that has mixed
327block sizes. The format of the option is bssplit=blocksize/percentage,
5982a925 328optionally adding as many definitions as needed separated by a colon.
9183788d 329Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
c83cdd3e
JA
330blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
331splits to reads and writes. The format is identical to what the
332\fBbs\fR option accepts, the read and write parts are separated with a
333comma.
d60e92d1
AC
334.TP
335.B blocksize_unaligned\fR,\fP bs_unaligned
d1429b5c
AC
336If set, any size in \fBblocksize_range\fR may be used. This typically won't
337work with direct I/O, as that normally requires sector alignment.
d60e92d1 338.TP
2b7a01d0 339.BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
639ce0f3
MS
340At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
341the minimum blocksize given. Minimum alignment is typically 512b
2b7a01d0
JA
342for using direct IO, though it usually depends on the hardware block size.
343This option is mutually exclusive with using a random map for files, so it
344will turn off that option.
43602667 345.TP
d60e92d1
AC
346.B zero_buffers
347Initialise buffers with all zeros. Default: fill buffers with random data.
348.TP
901bb994
JA
349.B refill_buffers
350If this option is given, fio will refill the IO buffers on every submit. The
351default is to only fill it at init time and reuse that data. Only makes sense
352if zero_buffers isn't specified, naturally. If data verification is enabled,
353refill_buffers is also automatically enabled.
354.TP
fd68418e
JA
355.BI scramble_buffers \fR=\fPbool
356If \fBrefill_buffers\fR is too costly and the target is using data
357deduplication, then setting this option will slightly modify the IO buffer
358contents to defeat normal de-dupe attempts. This is not enough to defeat
359more clever block compression attempts, but it will stop naive dedupe
360of blocks. Default: true.
361.TP
c5751c62
JA
362.BI buffer_compress_percentage \fR=\fPint
363If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
364that compress to the specified level. Fio does this by providing a mix of
365random data and zeroes. Note that this is per block size unit, for file/disk
366wide compression level that matches this setting, you'll also want to set
367\fBrefill_buffers\fR.
368.TP
369.BI buffer_compress_chunk \fR=\fPint
370See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
371big the ranges of random data and zeroed data is. Without this set, fio will
372provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
373the remaining zeroed. With this set to some chunk size smaller than the block
374size, fio can alternate random and zeroed data throughout the IO buffer.
375.TP
d60e92d1
AC
376.BI nrfiles \fR=\fPint
377Number of files to use for this job. Default: 1.
378.TP
379.BI openfiles \fR=\fPint
380Number of files to keep open at the same time. Default: \fBnrfiles\fR.
381.TP
382.BI file_service_type \fR=\fPstr
383Defines how files to service are selected. The following types are defined:
384.RS
385.RS
386.TP
387.B random
388Choose a file at random
389.TP
390.B roundrobin
391Round robin over open files (default).
6b7f6851
JA
392.B sequential
393Do each file in the set sequentially.
d60e92d1
AC
394.RE
395.P
396The number of I/Os to issue before switching a new file can be specified by
397appending `:\fIint\fR' to the service type.
398.RE
399.TP
400.BI ioengine \fR=\fPstr
401Defines how the job issues I/O. The following types are defined:
402.RS
403.RS
404.TP
405.B sync
406Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
407position the I/O location.
408.TP
a31041ea 409.B psync
410Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
411.TP
9183788d
JA
412.B vsync
413Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
414coalescing adjacents IOs into a single submission.
415.TP
d60e92d1 416.B libaio
de890a1e 417Linux native asynchronous I/O. This ioengine defines engine specific options.
d60e92d1
AC
418.TP
419.B posixaio
03e20d68
BC
420POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
421.TP
422.B solarisaio
423Solaris native asynchronous I/O.
424.TP
425.B windowsaio
426Windows native asynchronous I/O.
d60e92d1
AC
427.TP
428.B mmap
d1429b5c
AC
429File is memory mapped with \fImmap\fR\|(2) and data copied using
430\fImemcpy\fR\|(3).
d60e92d1
AC
431.TP
432.B splice
d1429b5c
AC
433\fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
434transfer data from user-space to the kernel.
d60e92d1
AC
435.TP
436.B syslet-rw
437Use the syslet system calls to make regular read/write asynchronous.
438.TP
439.B sg
440SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
d1429b5c
AC
441the target is an sg character device, we use \fIread\fR\|(2) and
442\fIwrite\fR\|(2) for asynchronous I/O.
d60e92d1
AC
443.TP
444.B null
445Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
446itself and for debugging and testing purposes.
447.TP
448.B net
de890a1e
SL
449Transfer over the network. The protocol to be used can be defined with the
450\fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
451\fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
452This ioengine defines engine specific options.
d60e92d1
AC
453.TP
454.B netsplice
455Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
de890a1e 456and send/receive. This ioengine defines engine specific options.
d60e92d1 457.TP
53aec0a4 458.B cpuio
d60e92d1
AC
459Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
460\fBcpucycles\fR parameters.
461.TP
462.B guasi
463The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
464approach to asycnronous I/O.
d1429b5c
AC
465.br
466See <http://www.xmailserver.org/guasi\-lib.html>.
d60e92d1 467.TP
21b8aee8 468.B rdma
85286c5c
BVA
469The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
470and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
21b8aee8 471.TP
d60e92d1
AC
472.B external
473Loads an external I/O engine object file. Append the engine filename as
474`:\fIenginepath\fR'.
475.RE
476.RE
477.TP
478.BI iodepth \fR=\fPint
8489dae4
SK
479Number of I/O units to keep in flight against the file. Note that increasing
480iodepth beyond 1 will not affect synchronous ioengines (except for small
ee72ca09
JA
481degress when verify_async is in use). Even async engines my impose OS
482restrictions causing the desired depth not to be achieved. This may happen on
483Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
484not async on that OS. Keep an eye on the IO depth distribution in the
485fio output to verify that the achieved depth is as expected. Default: 1.
d60e92d1
AC
486.TP
487.BI iodepth_batch \fR=\fPint
488Number of I/Os to submit at once. Default: \fBiodepth\fR.
489.TP
3ce9dcaf
JA
490.BI iodepth_batch_complete \fR=\fPint
491This defines how many pieces of IO to retrieve at once. It defaults to 1 which
492 means that we'll ask for a minimum of 1 IO in the retrieval process from the
493kernel. The IO retrieval will go on until we hit the limit set by
494\fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
495completed events before queuing more IO. This helps reduce IO latency, at the
496cost of more retrieval system calls.
497.TP
d60e92d1
AC
498.BI iodepth_low \fR=\fPint
499Low watermark indicating when to start filling the queue again. Default:
500\fBiodepth\fR.
501.TP
502.BI direct \fR=\fPbool
503If true, use non-buffered I/O (usually O_DIRECT). Default: false.
504.TP
505.BI buffered \fR=\fPbool
506If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
507Default: true.
508.TP
f7fa2653 509.BI offset \fR=\fPint
d60e92d1
AC
510Offset in the file to start I/O. Data before the offset will not be touched.
511.TP
591e9e06
JA
512.BI offset_increment \fR=\fPint
513If this is provided, then the real offset becomes the
514offset + offset_increment * thread_number, where the thread number is a counter
515that starts at 0 and is incremented for each job. This option is useful if
516there are several jobs which are intended to operate on a file in parallel in
517disjoint segments, with even spacing between the starting points.
518.TP
d60e92d1 519.BI fsync \fR=\fPint
d1429b5c
AC
520How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
5210, don't sync. Default: 0.
d60e92d1 522.TP
5f9099ea
JA
523.BI fdatasync \fR=\fPint
524Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
525data parts of the file. Default: 0.
526.TP
e76b1da4
JA
527.BI sync_file_range \fR=\fPstr:int
528Use sync_file_range() for every \fRval\fP number of write operations. Fio will
529track range of writes that have happened since the last sync_file_range() call.
530\fRstr\fP can currently be one or more of:
531.RS
532.TP
533.B wait_before
534SYNC_FILE_RANGE_WAIT_BEFORE
535.TP
536.B write
537SYNC_FILE_RANGE_WRITE
538.TP
539.B wait_after
540SYNC_FILE_RANGE_WRITE
541.TP
542.RE
543.P
544So if you do sync_file_range=wait_before,write:8, fio would use
545\fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
546Also see the sync_file_range(2) man page. This option is Linux specific.
547.TP
d60e92d1 548.BI overwrite \fR=\fPbool
d1429b5c 549If writing, setup the file first and do overwrites. Default: false.
d60e92d1
AC
550.TP
551.BI end_fsync \fR=\fPbool
d1429b5c 552Sync file contents when job exits. Default: false.
d60e92d1
AC
553.TP
554.BI fsync_on_close \fR=\fPbool
555If true, sync file contents on close. This differs from \fBend_fsync\fR in that
d1429b5c 556it will happen on every close, not just at the end of the job. Default: false.
d60e92d1
AC
557.TP
558.BI rwmixcycle \fR=\fPint
559How many milliseconds before switching between reads and writes for a mixed
560workload. Default: 500ms.
561.TP
562.BI rwmixread \fR=\fPint
563Percentage of a mixed workload that should be reads. Default: 50.
564.TP
565.BI rwmixwrite \fR=\fPint
d1429b5c 566Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
c35dd7a6
JA
567\fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
568overrides the first. This may interfere with a given rate setting, if fio is
569asked to limit reads or writes to a certain rate. If that is the case, then
570the distribution may be skewed. Default: 50.
d60e92d1
AC
571.TP
572.B norandommap
573Normally \fBfio\fR will cover every block of the file when doing random I/O. If
574this parameter is given, a new offset will be chosen without looking at past
575I/O history. This parameter is mutually exclusive with \fBverify\fR.
576.TP
744492c9 577.BI softrandommap \fR=\fPbool
3ce9dcaf
JA
578See \fBnorandommap\fR. If fio runs with the random block map enabled and it
579fails to allocate the map, if this option is set it will continue without a
580random block map. As coverage will not be as complete as with random maps, this
581option is disabled by default.
582.TP
d60e92d1
AC
583.BI nice \fR=\fPint
584Run job with given nice value. See \fInice\fR\|(2).
585.TP
586.BI prio \fR=\fPint
587Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
588\fIionice\fR\|(1).
589.TP
590.BI prioclass \fR=\fPint
591Set I/O priority class. See \fIionice\fR\|(1).
592.TP
593.BI thinktime \fR=\fPint
594Stall job for given number of microseconds between issuing I/Os.
595.TP
596.BI thinktime_spin \fR=\fPint
597Pretend to spend CPU time for given number of microseconds, sleeping the rest
598of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
599.TP
600.BI thinktime_blocks \fR=\fPint
601Number of blocks to issue before waiting \fBthinktime\fR microseconds.
602Default: 1.
603.TP
604.BI rate \fR=\fPint
c35dd7a6
JA
605Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
606rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
607or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
608limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
609can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
610limit writes (to 500KB/sec), the latter will only limit reads.
d60e92d1
AC
611.TP
612.BI ratemin \fR=\fPint
613Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
c35dd7a6
JA
614Failing to meet this requirement will cause the job to exit. The same format
615as \fBrate\fR is used for read vs write separation.
d60e92d1
AC
616.TP
617.BI rate_iops \fR=\fPint
c35dd7a6
JA
618Cap the bandwidth to this number of IOPS. Basically the same as rate, just
619specified independently of bandwidth. The same format as \fBrate\fR is used for
620read vs write seperation. If \fBblocksize\fR is a range, the smallest block
621size is used as the metric.
d60e92d1
AC
622.TP
623.BI rate_iops_min \fR=\fPint
c35dd7a6
JA
624If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
625is used for read vs write seperation.
d60e92d1
AC
626.TP
627.BI ratecycle \fR=\fPint
628Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
629milliseconds. Default: 1000ms.
630.TP
631.BI cpumask \fR=\fPint
632Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
633may run on. See \fBsched_setaffinity\fR\|(2).
634.TP
635.BI cpus_allowed \fR=\fPstr
636Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
637.TP
638.BI startdelay \fR=\fPint
639Delay start of job for the specified number of seconds.
640.TP
641.BI runtime \fR=\fPint
642Terminate processing after the specified number of seconds.
643.TP
644.B time_based
645If given, run for the specified \fBruntime\fR duration even if the files are
646completely read or written. The same workload will be repeated as many times
647as \fBruntime\fR allows.
648.TP
901bb994
JA
649.BI ramp_time \fR=\fPint
650If set, fio will run the specified workload for this amount of time before
651logging any performance numbers. Useful for letting performance settle before
652logging results, thus minimizing the runtime required for stable results. Note
c35dd7a6
JA
653that the \fBramp_time\fR is considered lead in time for a job, thus it will
654increase the total runtime if a special timeout or runtime is specified.
901bb994 655.TP
d60e92d1
AC
656.BI invalidate \fR=\fPbool
657Invalidate buffer-cache for the file prior to starting I/O. Default: true.
658.TP
659.BI sync \fR=\fPbool
660Use synchronous I/O for buffered writes. For the majority of I/O engines,
d1429b5c 661this means using O_SYNC. Default: false.
d60e92d1
AC
662.TP
663.BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
664Allocation method for I/O unit buffer. Allowed values are:
665.RS
666.RS
667.TP
668.B malloc
669Allocate memory with \fImalloc\fR\|(3).
670.TP
671.B shm
672Use shared memory buffers allocated through \fIshmget\fR\|(2).
673.TP
674.B shmhuge
675Same as \fBshm\fR, but use huge pages as backing.
676.TP
677.B mmap
678Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
679is given after the option in the format `:\fIfile\fR'.
680.TP
681.B mmaphuge
682Same as \fBmmap\fR, but use huge files as backing.
683.RE
684.P
685The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
686job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
687the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
2e266ba6
JA
688have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
689huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
690and the documentation for that. Normally you just need to echo an appropriate
691number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
692use.
d60e92d1
AC
693.RE
694.TP
d392365e 695.BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
d529ee19
JA
696This indiciates the memory alignment of the IO memory buffers. Note that the
697given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
698the alignment of the following buffers are given by the \fBbs\fR used. In
699other words, if using a \fBbs\fR that is a multiple of the page sized in the
700system, all buffers will be aligned to this value. If using a \fBbs\fR that
701is not page aligned, the alignment of subsequent IO memory buffers is the
702sum of the \fBiomem_align\fR and \fBbs\fR used.
703.TP
f7fa2653 704.BI hugepage\-size \fR=\fPint
d60e92d1 705Defines the size of a huge page. Must be at least equal to the system setting.
b22989b9 706Should be a multiple of 1MB. Default: 4MB.
d60e92d1
AC
707.TP
708.B exitall
709Terminate all jobs when one finishes. Default: wait for each job to finish.
710.TP
711.BI bwavgtime \fR=\fPint
712Average bandwidth calculations over the given time in milliseconds. Default:
713500ms.
714.TP
c8eeb9df
JA
715.BI iopsavgtime \fR=\fPint
716Average IOPS calculations over the given time in milliseconds. Default:
717500ms.
718.TP
d60e92d1 719.BI create_serialize \fR=\fPbool
d1429b5c 720If true, serialize file creation for the jobs. Default: true.
d60e92d1
AC
721.TP
722.BI create_fsync \fR=\fPbool
723\fIfsync\fR\|(2) data file after creation. Default: true.
724.TP
6b7f6851
JA
725.BI create_on_open \fR=\fPbool
726If true, the files are not created until they are opened for IO by the job.
727.TP
25460cf6
JA
728.BI create_only \fR=\fPbool
729If true, fio will only run the setup phase of the job. If files need to be
730laid out or updated on disk, only that will be done. The actual job contents
731are not executed.
732.TP
e9f48479
JA
733.BI pre_read \fR=\fPbool
734If this is given, files will be pre-read into memory before starting the given
735IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
9c0d2241
JA
736pointless to pre-read and then drop the cache. This will only work for IO
737engines that are seekable, since they allow you to read the same data
738multiple times. Thus it will not work on eg network or splice IO.
e9f48479 739.TP
d60e92d1
AC
740.BI unlink \fR=\fPbool
741Unlink job files when done. Default: false.
742.TP
743.BI loops \fR=\fPint
744Specifies the number of iterations (runs of the same workload) of this job.
745Default: 1.
746.TP
747.BI do_verify \fR=\fPbool
748Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
749Default: true.
750.TP
751.BI verify \fR=\fPstr
752Method of verifying file contents after each iteration of the job. Allowed
753values are:
754.RS
755.RS
756.TP
b892dc08 757.B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
0539d758
JA
758Store appropriate checksum in the header of each block. crc32c-intel is
759hardware accelerated SSE4.2 driven, falls back to regular crc32c if
760not supported by the system.
d60e92d1
AC
761.TP
762.B meta
763Write extra information about each I/O (timestamp, block number, etc.). The
996093bb 764block number is verified. See \fBverify_pattern\fR as well.
d60e92d1
AC
765.TP
766.B null
767Pretend to verify. Used for testing internals.
768.RE
b892dc08
JA
769
770This option can be used for repeated burn-in tests of a system to make sure
771that the written data is also correctly read back. If the data direction given
772is a read or random read, fio will assume that it should verify a previously
773written file. If the data direction includes any form of write, the verify will
774be of the newly written data.
d60e92d1
AC
775.RE
776.TP
777.BI verify_sort \fR=\fPbool
778If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
779read them back in a sorted manner. Default: true.
780.TP
f7fa2653 781.BI verify_offset \fR=\fPint
d60e92d1 782Swap the verification header with data somewhere else in the block before
d1429b5c 783writing. It is swapped back before verifying.
d60e92d1 784.TP
f7fa2653 785.BI verify_interval \fR=\fPint
d60e92d1
AC
786Write the verification header for this number of bytes, which should divide
787\fBblocksize\fR. Default: \fBblocksize\fR.
788.TP
996093bb
JA
789.BI verify_pattern \fR=\fPstr
790If set, fio will fill the io buffers with this pattern. Fio defaults to filling
791with totally random bytes, but sometimes it's interesting to fill with a known
792pattern for io verification purposes. Depending on the width of the pattern,
793fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
794decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
795has to be a hex number that starts with either "0x" or "0X". Use with
796\fBverify\fP=meta.
797.TP
d60e92d1
AC
798.BI verify_fatal \fR=\fPbool
799If true, exit the job on the first observed verification failure. Default:
800false.
801.TP
b463e936
JA
802.BI verify_dump \fR=\fPbool
803If set, dump the contents of both the original data block and the data block we
804read off disk to files. This allows later analysis to inspect just what kind of
ef71e317 805data corruption occurred. Off by default.
b463e936 806.TP
e8462bd8
JA
807.BI verify_async \fR=\fPint
808Fio will normally verify IO inline from the submitting thread. This option
809takes an integer describing how many async offload threads to create for IO
810verification instead, causing fio to offload the duty of verifying IO contents
c85c324c
JA
811to one or more separate threads. If using this offload option, even sync IO
812engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
813allows them to have IO in flight while verifies are running.
e8462bd8
JA
814.TP
815.BI verify_async_cpus \fR=\fPstr
816Tell fio to set the given CPU affinity on the async IO verification threads.
817See \fBcpus_allowed\fP for the format used.
818.TP
6f87418f
JA
819.BI verify_backlog \fR=\fPint
820Fio will normally verify the written contents of a job that utilizes verify
821once that job has completed. In other words, everything is written then
822everything is read back and verified. You may want to verify continually
823instead for a variety of reasons. Fio stores the meta data associated with an
824IO block in memory, so for large verify workloads, quite a bit of memory would
092f707f
DN
825be used up holding this meta data. If this option is enabled, fio will write
826only N blocks before verifying these blocks.
6f87418f
JA
827.TP
828.BI verify_backlog_batch \fR=\fPint
829Control how many blocks fio will verify if verify_backlog is set. If not set,
830will default to the value of \fBverify_backlog\fR (meaning the entire queue is
092f707f
DN
831read back and verified). If \fBverify_backlog_batch\fR is less than
832\fBverify_backlog\fR then not all blocks will be verified, if
833\fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
834will be verified more than once.
6f87418f 835.TP
d392365e 836.B stonewall "\fR,\fP wait_for_previous"
5982a925 837Wait for preceding jobs in the job file to exit before starting this one.
d60e92d1
AC
838\fBstonewall\fR implies \fBnew_group\fR.
839.TP
840.B new_group
841Start a new reporting group. If not given, all jobs in a file will be part
842of the same reporting group, unless separated by a stonewall.
843.TP
844.BI numjobs \fR=\fPint
845Number of clones (processes/threads performing the same workload) of this job.
846Default: 1.
847.TP
848.B group_reporting
849If set, display per-group reports instead of per-job when \fBnumjobs\fR is
850specified.
851.TP
852.B thread
853Use threads created with \fBpthread_create\fR\|(3) instead of processes created
854with \fBfork\fR\|(2).
855.TP
f7fa2653 856.BI zonesize \fR=\fPint
d60e92d1
AC
857Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
858.TP
f7fa2653 859.BI zoneskip \fR=\fPint
d1429b5c 860Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
d60e92d1
AC
861read.
862.TP
863.BI write_iolog \fR=\fPstr
5b42a488
SH
864Write the issued I/O patterns to the specified file. Specify a separate file
865for each job, otherwise the iologs will be interspersed and the file may be
866corrupt.
d60e92d1
AC
867.TP
868.BI read_iolog \fR=\fPstr
869Replay the I/O patterns contained in the specified file generated by
870\fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
871.TP
64bbb865
DN
872.BI replay_no_stall \fR=\fPint
873While replaying I/O patterns using \fBread_iolog\fR the default behavior
874attempts to respect timing information between I/Os. Enabling
875\fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
876still respecting ordering.
877.TP
d1c46c04
DN
878.BI replay_redirect \fR=\fPstr
879While replaying I/O patterns using \fBread_iolog\fR the default behavior
880is to replay the IOPS onto the major/minor device that each IOP was recorded
881from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
882single specified device regardless of the device it was recorded from.
883.TP
836bad52 884.BI write_bw_log \fR=\fPstr
901bb994
JA
885If given, write a bandwidth log of the jobs in this job file. Can be used to
886store data of the bandwidth of the jobs in their lifetime. The included
887fio_generate_plots script uses gnuplot to turn these text files into nice
888graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
889option, the postfix is _bw.log.
d60e92d1 890.TP
836bad52 891.BI write_lat_log \fR=\fPstr
901bb994
JA
892Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
893filename is given with this option, the default filename of "jobname_type.log"
894is used. Even if the filename is given, fio will still append the type of log.
895.TP
c8eeb9df
JA
896.BI write_iops_log \fR=\fPstr
897Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
898option, the default filename of "jobname_type.log" is used. Even if the
899filename is given, fio will still append the type of log.
900.TP
b8bc8cba
JA
901.BI log_avg_msec \fR=\fPint
902By default, fio will log an entry in the iops, latency, or bw log for every
903IO that completes. When writing to the disk log, that can quickly grow to a
904very large size. Setting this option makes fio average the each log entry
905over the specified period of time, reducing the resolution of the log.
906Defaults to 0.
907.TP
836bad52 908.BI disable_lat \fR=\fPbool
02af0988 909Disable measurements of total latency numbers. Useful only for cutting
901bb994
JA
910back the number of calls to gettimeofday, as that does impact performance at
911really high IOPS rates. Note that to really get rid of a large amount of these
912calls, this option must be used with disable_slat and disable_bw as well.
913.TP
836bad52 914.BI disable_clat \fR=\fPbool
c95f9daf 915Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
02af0988 916.TP
836bad52 917.BI disable_slat \fR=\fPbool
02af0988 918Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
901bb994 919.TP
836bad52 920.BI disable_bw_measurement \fR=\fPbool
02af0988 921Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
d60e92d1 922.TP
f7fa2653 923.BI lockmem \fR=\fPint
d60e92d1
AC
924Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
925simulate a smaller amount of memory.
926.TP
927.BI exec_prerun \fR=\fPstr
928Before running the job, execute the specified command with \fBsystem\fR\|(3).
929.TP
930.BI exec_postrun \fR=\fPstr
931Same as \fBexec_prerun\fR, but the command is executed after the job completes.
932.TP
933.BI ioscheduler \fR=\fPstr
934Attempt to switch the device hosting the file to the specified I/O scheduler.
935.TP
936.BI cpuload \fR=\fPint
937If the job is a CPU cycle-eater, attempt to use the specified percentage of
938CPU cycles.
939.TP
940.BI cpuchunks \fR=\fPint
941If the job is a CPU cycle-eater, split the load into cycles of the
942given time in milliseconds.
943.TP
944.BI disk_util \fR=\fPbool
d1429b5c 945Generate disk utilization statistics if the platform supports it. Default: true.
901bb994
JA
946.TP
947.BI gtod_reduce \fR=\fPbool
948Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
949disable_bw) plus reduce precision of the timeout somewhat to really shrink the
950gettimeofday() call count. With this option enabled, we only do about 0.4% of
951the gtod() calls we would have done if all time keeping was enabled.
952.TP
953.BI gtod_cpu \fR=\fPint
954Sometimes it's cheaper to dedicate a single thread of execution to just getting
955the current time. Fio (and databases, for instance) are very intensive on
956gettimeofday() calls. With this option, you can set one CPU aside for doing
957nothing but logging current time to a shared memory location. Then the other
958threads/processes that run IO workloads need only copy that segment, instead of
959entering the kernel with a gettimeofday() call. The CPU set aside for doing
960these time calls will be excluded from other uses. Fio will manually clear it
961from the CPU mask of other jobs.
f2bba182 962.TP
a696fa2a
JA
963.BI cgroup \fR=\fPstr
964Add job to this control group. If it doesn't exist, it will be created.
6adb38a1
JA
965The system must have a mounted cgroup blkio mount point for this to work. If
966your system doesn't have it mounted, you can do so with:
967
5982a925 968# mount \-t cgroup \-o blkio none /cgroup
a696fa2a
JA
969.TP
970.BI cgroup_weight \fR=\fPint
971Set the weight of the cgroup to this value. See the documentation that comes
972with the kernel, allowed values are in the range of 100..1000.
e0b0d892 973.TP
7de87099
VG
974.BI cgroup_nodelete \fR=\fPbool
975Normally fio will delete the cgroups it has created after the job completion.
976To override this behavior and to leave cgroups around after the job completion,
977set cgroup_nodelete=1. This can be useful if one wants to inspect various
978cgroup files after job completion. Default: false
979.TP
e0b0d892
JA
980.BI uid \fR=\fPint
981Instead of running as the invoking user, set the user ID to this value before
982the thread/process does any work.
983.TP
984.BI gid \fR=\fPint
985Set group ID, see \fBuid\fR.
83349190 986.TP
9e684a49
DE
987.BI flow_id \fR=\fPint
988The ID of the flow. If not specified, it defaults to being a global flow. See
989\fBflow\fR.
990.TP
991.BI flow \fR=\fPint
992Weight in token-based flow control. If this value is used, then there is a
993\fBflow counter\fR which is used to regulate the proportion of activity between
994two or more jobs. fio attempts to keep this flow counter near zero. The
995\fBflow\fR parameter stands for how much should be added or subtracted to the
996flow counter on each iteration of the main I/O loop. That is, if one job has
997\fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
9981:8 ratio in how much one runs vs the other.
999.TP
1000.BI flow_watermark \fR=\fPint
1001The maximum value that the absolute value of the flow counter is allowed to
1002reach before the job must wait for a lower value of the counter.
1003.TP
1004.BI flow_sleep \fR=\fPint
1005The period of time, in microseconds, to wait after the flow watermark has been
1006exceeded before retrying operations
1007.TP
83349190
YH
1008.BI clat_percentiles \fR=\fPbool
1009Enable the reporting of percentiles of completion latencies.
1010.TP
1011.BI percentile_list \fR=\fPfloat_list
1012Overwrite the default list of percentiles for completion
1013latencies. Each number is a floating number in the range (0,100], and
1014the maximum length of the list is 20. Use ':' to separate the
3eb07285 1015numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
83349190
YH
1016report the values of completion latency below which 99.5% and 99.9% of
1017the observed latencies fell, respectively.
de890a1e
SL
1018.SS "Ioengine Parameters List"
1019Some parameters are only valid when a specific ioengine is in use. These are
1020used identically to normal parameters, with the caveat that when used on the
1021command line, the must come after the ioengine that defines them is selected.
1022.TP
1023.BI (libaio)userspace_reap
1024Normally, with the libaio engine in use, fio will use
1025the io_getevents system call to reap newly returned events.
1026With this flag turned on, the AIO ring will be read directly
1027from user-space to reap events. The reaping mode is only
1028enabled when polling for a minimum of 0 events (eg when
1029iodepth_batch_complete=0).
1030.TP
1031.BI (net,netsplice)hostname \fR=\fPstr
1032The host name or IP address to use for TCP or UDP based IO.
1033If the job is a TCP listener or UDP reader, the hostname is not
1034used and must be omitted.
1035.TP
1036.BI (net,netsplice)port \fR=\fPint
1037The TCP or UDP port to bind to or connect to.
1038.TP
1039.BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1040The network protocol to use. Accepted values are:
1041.RS
1042.RS
1043.TP
1044.B tcp
1045Transmission control protocol
1046.TP
1047.B udp
1048Unreliable datagram protocol
1049.TP
1050.B unix
1051UNIX domain socket
1052.RE
1053.P
1054When the protocol is TCP or UDP, the port must also be given,
1055as well as the hostname if the job is a TCP listener or UDP
1056reader. For unix sockets, the normal filename option should be
1057used and the port is invalid.
1058.RE
1059.TP
1060.BI (net,netsplice)listen
1061For TCP network connections, tell fio to listen for incoming
1062connections rather than initiating an outgoing connection. The
1063hostname must be omitted if this option is used.
d60e92d1 1064.SH OUTPUT
d1429b5c
AC
1065While running, \fBfio\fR will display the status of the created jobs. For
1066example:
d60e92d1 1067.RS
d1429b5c 1068.P
d60e92d1
AC
1069Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1070.RE
1071.P
d1429b5c
AC
1072The characters in the first set of brackets denote the current status of each
1073threads. The possible values are:
1074.P
1075.PD 0
d60e92d1
AC
1076.RS
1077.TP
1078.B P
1079Setup but not started.
1080.TP
1081.B C
1082Thread created.
1083.TP
1084.B I
1085Initialized, waiting.
1086.TP
1087.B R
1088Running, doing sequential reads.
1089.TP
1090.B r
1091Running, doing random reads.
1092.TP
1093.B W
1094Running, doing sequential writes.
1095.TP
1096.B w
1097Running, doing random writes.
1098.TP
1099.B M
1100Running, doing mixed sequential reads/writes.
1101.TP
1102.B m
1103Running, doing mixed random reads/writes.
1104.TP
1105.B F
1106Running, currently waiting for \fBfsync\fR\|(2).
1107.TP
1108.B V
1109Running, verifying written data.
1110.TP
1111.B E
1112Exited, not reaped by main thread.
1113.TP
1114.B \-
1115Exited, thread reaped.
1116.RE
d1429b5c 1117.PD
d60e92d1
AC
1118.P
1119The second set of brackets shows the estimated completion percentage of
1120the current group. The third set shows the read and write I/O rate,
1121respectively. Finally, the estimated run time of the job is displayed.
1122.P
1123When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1124for each thread, each group of threads, and each disk, in that order.
1125.P
1126Per-thread statistics first show the threads client number, group-id, and
1127error code. The remaining figures are as follows:
1128.RS
d60e92d1
AC
1129.TP
1130.B io
1131Number of megabytes of I/O performed.
1132.TP
1133.B bw
1134Average data rate (bandwidth).
1135.TP
1136.B runt
1137Threads run time.
1138.TP
1139.B slat
1140Submission latency minimum, maximum, average and standard deviation. This is
1141the time it took to submit the I/O.
1142.TP
1143.B clat
1144Completion latency minimum, maximum, average and standard deviation. This
1145is the time between submission and completion.
1146.TP
1147.B bw
1148Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1149and standard deviation.
1150.TP
1151.B cpu
1152CPU usage statistics. Includes user and system time, number of context switches
1153this thread went through and number of major and minor page faults.
1154.TP
1155.B IO depths
1156Distribution of I/O depths. Each depth includes everything less than (or equal)
1157to it, but greater than the previous depth.
1158.TP
1159.B IO issued
1160Number of read/write requests issued, and number of short read/write requests.
1161.TP
1162.B IO latencies
1163Distribution of I/O completion latencies. The numbers follow the same pattern
1164as \fBIO depths\fR.
1165.RE
d60e92d1
AC
1166.P
1167The group statistics show:
d1429b5c 1168.PD 0
d60e92d1
AC
1169.RS
1170.TP
1171.B io
1172Number of megabytes I/O performed.
1173.TP
1174.B aggrb
1175Aggregate bandwidth of threads in the group.
1176.TP
1177.B minb
1178Minimum average bandwidth a thread saw.
1179.TP
1180.B maxb
1181Maximum average bandwidth a thread saw.
1182.TP
1183.B mint
d1429b5c 1184Shortest runtime of threads in the group.
d60e92d1
AC
1185.TP
1186.B maxt
1187Longest runtime of threads in the group.
1188.RE
d1429b5c 1189.PD
d60e92d1
AC
1190.P
1191Finally, disk statistics are printed with reads first:
d1429b5c 1192.PD 0
d60e92d1
AC
1193.RS
1194.TP
1195.B ios
1196Number of I/Os performed by all groups.
1197.TP
1198.B merge
1199Number of merges in the I/O scheduler.
1200.TP
1201.B ticks
1202Number of ticks we kept the disk busy.
1203.TP
1204.B io_queue
1205Total time spent in the disk queue.
1206.TP
1207.B util
1208Disk utilization.
1209.RE
d1429b5c 1210.PD
8423bd11
JA
1211.P
1212It is also possible to get fio to dump the current output while it is
1213running, without terminating the job. To do that, send fio the \fBUSR1\fR
1214signal.
d60e92d1
AC
1215.SH TERSE OUTPUT
1216If the \fB\-\-minimal\fR option is given, the results will be printed in a
562c2d2f
DN
1217semicolon-delimited format suitable for scripted use - a job description
1218(if provided) follows on a new line. Note that the first
525c2bfa
JA
1219number in the line is the version number. If the output has to be changed
1220for some reason, this number will be incremented by 1 to signify that
1221change. The fields are:
d60e92d1
AC
1222.P
1223.RS
5e726d0a 1224.B terse version, fio version, jobname, groupid, error
d60e92d1
AC
1225.P
1226Read status:
1227.RS
312b4af2 1228.B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
d60e92d1
AC
1229.P
1230Submission latency:
1231.RS
1232.B min, max, mean, standard deviation
1233.RE
1234Completion latency:
1235.RS
1236.B min, max, mean, standard deviation
1237.RE
1db92cb6
JA
1238Completion latency percentiles (20 fields):
1239.RS
1240.B Xth percentile=usec
1241.RE
525c2bfa
JA
1242Total latency:
1243.RS
1244.B min, max, mean, standard deviation
1245.RE
d60e92d1
AC
1246Bandwidth:
1247.RS
1248.B min, max, aggregate percentage of total, mean, standard deviation
1249.RE
1250.RE
1251.P
1252Write status:
1253.RS
312b4af2 1254.B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
d60e92d1
AC
1255.P
1256Submission latency:
1257.RS
1258.B min, max, mean, standard deviation
1259.RE
1260Completion latency:
1261.RS
1262.B min, max, mean, standard deviation
1263.RE
1db92cb6
JA
1264Completion latency percentiles (20 fields):
1265.RS
1266.B Xth percentile=usec
1267.RE
525c2bfa
JA
1268Total latency:
1269.RS
1270.B min, max, mean, standard deviation
1271.RE
d60e92d1
AC
1272Bandwidth:
1273.RS
1274.B min, max, aggregate percentage of total, mean, standard deviation
1275.RE
1276.RE
1277.P
d1429b5c 1278CPU usage:
d60e92d1 1279.RS
bd2626f0 1280.B user, system, context switches, major page faults, minor page faults
d60e92d1
AC
1281.RE
1282.P
1283IO depth distribution:
1284.RS
1285.B <=1, 2, 4, 8, 16, 32, >=64
1286.RE
1287.P
562c2d2f 1288IO latency distribution:
d60e92d1 1289.RS
562c2d2f
DN
1290Microseconds:
1291.RS
1292.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1293.RE
1294Milliseconds:
1295.RS
1296.B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1297.RE
1298.RE
1299.P
f2f788dd
JA
1300Disk utilization (1 for each disk used):
1301.RS
1302.B name, read ios, write ios, read merges, write merges, read ticks, write ticks, read in-queue time, write in-queue time, disk utilization percentage
1303.RE
1304.P
5982a925 1305Error Info (dependent on continue_on_error, default off):
562c2d2f
DN
1306.RS
1307.B total # errors, first error code
d60e92d1
AC
1308.RE
1309.P
562c2d2f 1310.B text description (if provided in config - appears on newline)
d60e92d1 1311.RE
49da1240
JA
1312.SH CLIENT / SERVER
1313Normally you would run fio as a stand-alone application on the machine
1314where the IO workload should be generated. However, it is also possible to
1315run the frontend and backend of fio separately. This makes it possible to
1316have a fio server running on the machine(s) where the IO workload should
1317be running, while controlling it from another machine.
1318
1319To start the server, you would do:
1320
1321\fBfio \-\-server=args\fR
1322
1323on that machine, where args defines what fio listens to. The arguments
811826be
JA
1324are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1325for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain socket.
1326'hostname' is either a hostname or IP address, and 'port' is the port to
1327listen to (only valid for TCP/IP, not a local socket). Some examples:
49da1240
JA
1328
13291) fio --server
1330
1331 Start a fio server, listening on all interfaces on the default port (8765).
1332
811826be 13332) fio --server=ip:hostname,4444
49da1240
JA
1334
1335 Start a fio server, listening on IP belonging to hostname and on port 4444.
1336
811826be
JA
13373) fio --server=ip6:::1,4444
1338
1339 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1340
13414) fio --server=,4444
49da1240
JA
1342
1343 Start a fio server, listening on all interfaces on port 4444.
1344
811826be 13455) fio --server=1.2.3.4
49da1240
JA
1346
1347 Start a fio server, listening on IP 1.2.3.4 on the default port.
1348
811826be 13496) fio --server=sock:/tmp/fio.sock
49da1240
JA
1350
1351 Start a fio server, listening on the local socket /tmp/fio.sock.
1352
1353When a server is running, you can connect to it from a client. The client
1354is run with:
1355
1356fio --local-args --client=server --remote-args <job file(s)>
1357
1358where --local-args are arguments that are local to the client where it is
1359running, 'server' is the connect string, and --remote-args and <job file(s)>
1360are sent to the server. The 'server' string follows the same format as it
1361does on the server side, to allow IP/hostname/socket and port strings.
1362You can connect to multiple clients as well, to do that you could run:
1363
1364fio --client=server2 --client=server2 <job file(s)>
d60e92d1 1365.SH AUTHORS
49da1240 1366
d60e92d1 1367.B fio
aa58d252
JA
1368was written by Jens Axboe <jens.axboe@oracle.com>,
1369now Jens Axboe <jaxboe@fusionio.com>.
d1429b5c
AC
1370.br
1371This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
d60e92d1
AC
1372on documentation by Jens Axboe.
1373.SH "REPORTING BUGS"
482900c9 1374Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
d1429b5c 1375See \fBREADME\fR.
d60e92d1 1376.SH "SEE ALSO"
d1429b5c
AC
1377For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1378.br
1379Sample jobfiles are available in the \fBexamples\fR directory.
d60e92d1 1380