1 .TH fio 1 "October 2013" "User Manual"
3 fio \- flexible I/O tester
6 [\fIoptions\fR] [\fIjobfile\fR]...
9 is a tool that will spawn a number of threads or processes doing a
10 particular type of I/O action as specified by the user.
11 The typical use of fio is to write a job file matching the I/O load
12 one wants to simulate.
15 .BI \-\-debug \fR=\fPtype
16 Enable verbose tracing of various fio actions. May be `all' for all types
17 or individual types separated by a comma (eg \-\-debug=io,file). `help' will
18 list all available tracing options.
20 .BI \-\-output \fR=\fPfilename
21 Write output to \fIfilename\fR.
23 .BI \-\-runtime \fR=\fPruntime
24 Limit run time to \fIruntime\fR seconds.
27 Generate per-job latency logs.
30 Generate per-job bandwidth logs.
33 Print statistics in a terse, semicolon-delimited format.
36 Print statistics in selected mode AND terse, semicolon-delimited format.
39 Display version information and exit.
41 .BI \-\-terse\-version \fR=\fPversion
42 Set terse version output format (Current version 3, or older version 2).
45 Display usage information and exit.
48 Perform test and validation of internal CPU clock
50 .BI \-\-crctest[\fR=\fPtest]
51 Test the speed of the builtin checksumming functions. If no argument is given,
52 all of them are tested. Or a comma separated list can be passed, in which
53 case the given ones are tested.
55 .BI \-\-cmdhelp \fR=\fPcommand
56 Print help information for \fIcommand\fR. May be `all' for all commands.
58 .BI \-\-enghelp \fR=\fPioengine[,command]
59 List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR.
61 .BI \-\-showcmd \fR=\fPjobfile
62 Convert \fIjobfile\fR to a set of command-line options.
64 .BI \-\-eta \fR=\fPwhen
65 Specifies when real-time ETA estimate should be printed. \fIwhen\fR may
66 be one of `always', `never' or `auto'.
68 .BI \-\-eta\-newline \fR=\fPtime
69 Force an ETA newline for every `time` period passed.
71 .BI \-\-status\-interval \fR=\fPtime
72 Report full output status every `time` period passed.
75 Turn on safety read-only checks, preventing any attempted write.
77 .BI \-\-section \fR=\fPsec
78 Only run section \fIsec\fR from job file. Multiple of these options can be given, adding more sections to run.
80 .BI \-\-alloc\-size \fR=\fPkb
81 Set the internal smalloc pool size to \fIkb\fP kilobytes.
83 .BI \-\-warnings\-fatal
84 All fio parser warnings are fatal, causing fio to exit with an error.
86 .BI \-\-max\-jobs \fR=\fPnr
87 Set the maximum allowed number of jobs (threads/processes) to support.
89 .BI \-\-server \fR=\fPargs
90 Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section.
92 .BI \-\-daemonize \fR=\fPpidfile
93 Background a fio server, writing the pid to the given pid file.
95 .BI \-\-client \fR=\fPhost
96 Instead of running the jobs locally, send and run them on the given host.
98 .BI \-\-idle\-prof \fR=\fPoption
99 Report cpu idleness on a system or percpu basis (\fIoption\fP=system,percpu) or run unit work calibration only (\fIoption\fP=calibrate).
100 .SH "JOB FILE FORMAT"
101 Job files are in `ini' format. They consist of one or more
102 job definitions, which begin with a job name in square brackets and
103 extend to the next job name. The job name can be any ASCII string
104 except `global', which has a special meaning. Following the job name is
105 a sequence of zero or more parameters, one per line, that define the
106 behavior of the job. Any line starting with a `;' or `#' character is
107 considered a comment and ignored.
109 If \fIjobfile\fR is specified as `-', the job file will be read from
112 The global section contains default parameters for jobs specified in the
113 job file. A job is only affected by global sections residing above it,
114 and there may be any number of global sections. Specific job definitions
115 may override any parameter set in global sections.
118 Some parameters may take arguments of a specific type. The types used are:
121 String: a sequence of alphanumeric characters.
124 SI integer: a whole number, possibly containing a suffix denoting the base unit
125 of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
126 kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
127 respectively. If prefixed with '0x', the value is assumed to be base 16
128 (hexadecimal). A suffix may include a trailing 'b', for instance 'kb' is
129 identical to 'k'. You can specify a base 10 value by using 'KiB', 'MiB','GiB',
130 etc. This is useful for disk drives where values are often given in base 10
131 values. Specifying '30GiB' will get you 30*1000^3 bytes.
132 When specifying times the default suffix meaning changes, still denoting the
133 base unit of the value, but accepted suffixes are 'D' (days), 'H' (hours), 'M'
134 (minutes), 'S' Seconds, 'ms' (or msec) milli seconds, 'us' (or 'usec') micro
135 seconds. Time values without a unit specify seconds.
136 The suffixes are not case sensitive.
139 Boolean: a true or false value. `0' denotes false, `1' denotes true.
142 Integer range: a range of integers specified in the format
143 \fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
144 \fIupper\fR may contain a suffix as described above. If an option allows two
145 sets of ranges, they are separated with a `,' or `/' character. For example:
149 List of floating numbers: A list of floating numbers, separated by
154 May be used to override the job name. On the command line, this parameter
155 has the special purpose of signalling the start of a new job.
157 .BI description \fR=\fPstr
158 Human-readable description of the job. It is printed when the job is run, but
159 otherwise has no special purpose.
161 .BI directory \fR=\fPstr
162 Prefix filenames with this directory. Used to place files in a location other
164 You can specify a number of directories by separating the names with a ':'
165 character. These directories will be assigned equally distributed to job clones
166 creates with \fInumjobs\fR as long as they are using generated filenames.
167 If specific \fIfilename(s)\fR are set fio will use the first listed directory,
168 and thereby matching the \fIfilename\fR semantic which generates a file each
169 clone if not specified, but let all clones use the same if set. See
170 \fIfilename\fR for considerations regarding escaping certain characters on
173 .BI filename \fR=\fPstr
175 normally makes up a file name based on the job name, thread number, and file
176 number. If you want to share files between threads in a job or several jobs,
177 specify a \fIfilename\fR for each of them to override the default.
178 If the I/O engine is file-based, you can specify
179 a number of files by separating the names with a `:' character. `\-' is a
180 reserved name, meaning stdin or stdout, depending on the read/write direction
181 set. On Windows, disk devices are accessed as \\.\PhysicalDrive0 for the first
182 device, \\.\PhysicalDrive1 for the second etc. Note: Windows and FreeBSD
183 prevent write access to areas of the disk containing in-use data
184 (e.g. filesystems). If the wanted filename does need to include a colon, then
185 escape that with a '\' character. For instance, if the filename is
186 "/dev/dsk/foo@3,0:c", then you would use filename="/dev/dsk/foo@3,0\:c".
188 .BI filename_format \fR=\fPstr
189 If sharing multiple files between jobs, it is usually necessary to have
190 fio generate the exact names that you want. By default, fio will name a file
191 based on the default file format specification of
192 \fBjobname.jobnumber.filenumber\fP. With this option, that can be
193 customized. Fio will recognize and replace the following keywords in this
199 The name of the worker thread or process.
202 The incremental number of the worker thread or process.
205 The incremental number of the file for that worker thread or process.
208 To have dependent jobs share a set of files, this option can be set to
209 have fio generate filenames that are shared between the two. For instance,
210 if \fBtestfiles.$filenum\fR is specified, file number 4 for any job will
211 be named \fBtestfiles.4\fR. The default of \fB$jobname.$jobnum.$filenum\fR
212 will be used if no other format specifier is given.
216 .BI lockfile \fR=\fPstr
217 Fio defaults to not locking any files before it does IO to them. If a file or
218 file descriptor is shared, fio can serialize IO to that file to make the end
219 result consistent. This is usual for emulating real workloads that share files.
225 No locking. This is the default.
228 Only one thread or process may do IO at the time, excluding all others.
231 Read-write locking on the file. Many readers may access the file at the same
232 time, but writes get exclusive access.
236 .BI opendir \fR=\fPstr
237 Recursively open any files below directory \fIstr\fR.
239 .BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
240 Type of I/O pattern. Accepted values are:
251 Sequential trim (Linux block devices only).
260 Random trim (Linux block devices only).
263 Mixed sequential reads and writes.
266 Mixed random reads and writes.
269 For mixed I/O, the default split is 50/50. For certain types of io the result
270 may still be skewed a bit, since the speed may be different. It is possible to
271 specify a number of IO's to do before getting a new offset, this is done by
272 appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
273 would look like \fBrw=randread:8\fR for passing in an offset modifier with a
274 value of 8. If the postfix is used with a sequential IO pattern, then the value
275 specified will be added to the generated offset for each IO. For instance,
276 using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
277 into sequential IO with holes. See the \fBrw_sequencer\fR option.
280 .BI rw_sequencer \fR=\fPstr
281 If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
282 then this option controls how that number modifies the IO offset being
283 generated. Accepted values are:
288 Generate sequential offset
291 Generate the same offset
294 \fBsequential\fR is only useful for random IO, where fio would normally
295 generate a new random offset for every IO. If you append eg 8 to randread, you
296 would get a new random offset for every 8 IO's. The result would be a seek for
297 only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
298 that. As sequential IO is already sequential, setting \fBsequential\fR for that
299 would not result in any differences. \fBidentical\fR behaves in a similar
300 fashion, except it sends the same offset 8 number of times before generating a
305 .BI kb_base \fR=\fPint
306 The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
307 manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
308 reasons. Allowed values are 1024 or 1000, with 1024 being the default.
310 .BI unified_rw_reporting \fR=\fPbool
311 Fio normally reports statistics on a per data direction basis, meaning that
312 read, write, and trim are accounted and reported separately. If this option is
313 set, the fio will sum the results and report them as "mixed" instead.
315 .BI randrepeat \fR=\fPbool
316 Seed the random number generator used for random I/O patterns in a predictable
317 way so the pattern is repeatable across runs. Default: true.
319 .BI allrandrepeat \fR=\fPbool
320 Seed all random number generators in a predictable way so results are
321 repeatable across runs. Default: false.
323 .BI randseed \fR=\fPint
324 Seed the random number generators based on this seed value, to be able to
325 control what sequence of output is being generated. If not set, the random
326 sequence depends on the \fBrandrepeat\fR setting.
328 .BI use_os_rand \fR=\fPbool
329 Fio can either use the random generator supplied by the OS to generator random
330 offsets, or it can use it's own internal generator (based on Tausworthe).
331 Default is to use the internal generator, which is often of better quality and
332 faster. Default: false.
334 .BI fallocate \fR=\fPstr
335 Whether pre-allocation is performed when laying down files. Accepted values
341 Do not pre-allocate space.
344 Pre-allocate via \fBposix_fallocate\fR\|(3).
347 Pre-allocate via \fBfallocate\fR\|(2) with FALLOC_FL_KEEP_SIZE set.
350 Backward-compatible alias for 'none'.
353 Backward-compatible alias for 'posix'.
356 May not be available on all supported platforms. 'keep' is only
357 available on Linux. If using ZFS on Solaris this must be set to 'none'
358 because ZFS doesn't support it. Default: 'posix'.
361 .BI fadvise_hint \fR=\fPbool
362 Use of \fBposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
363 are likely to be issued. Default: true.
366 Total size of I/O for this job. \fBfio\fR will run until this many bytes have
367 been transferred, unless limited by other options (\fBruntime\fR, for instance).
368 Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
369 divided between the available files for the job. If not set, fio will use the
370 full size of the given files or devices. If the files do not exist, size
371 must be given. It is also possible to give size as a percentage between 1 and
372 100. If size=20% is given, fio will use 20% of the full size of the given files
375 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
376 Sets size to something really large and waits for ENOSPC (no space left on
377 device) as the terminating condition. Only makes sense with sequential write.
378 For a read workload, the mount point will be filled first then IO started on
379 the result. This option doesn't make sense if operating on a raw device node,
380 since the size of that is already known by the file system. Additionally,
381 writing beyond end-of-device will not return ENOSPC there.
383 .BI filesize \fR=\fPirange
384 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
385 for files at random within the given range, limited to \fBsize\fR in total (if
386 that is given). If \fBfilesize\fR is not specified, each created file is the
389 .BI file_append \fR=\fPbool
390 Perform IO after the end of the file. Normally fio will operate within the
391 size of a file. If this option is set, then fio will append to the file
392 instead. This has identical behavior to setting \fRoffset\fP to the size
393 of a file. This option is ignored on non-regular files.
395 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
396 Block size for I/O units. Default: 4k. Values for reads, writes, and trims
397 can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR
398 either of which may be empty to leave that value at its default. If a trailing
399 comma isn't given, the remainder will inherit the last value set.
401 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
402 Specify a range of I/O block sizes. The issued I/O unit will always be a
403 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
404 to both reads and writes if only one range is given, but can be specified
405 separately with a comma separating the values. Example: bsrange=1k-4k,2k-8k.
406 Also (see \fBblocksize\fR).
408 .BI bssplit \fR=\fPstr
409 This option allows even finer grained control of the block sizes issued,
410 not just even splits between them. With this option, you can weight various
411 block sizes for exact control of the issued IO for a job that has mixed
412 block sizes. The format of the option is bssplit=blocksize/percentage,
413 optionally adding as many definitions as needed separated by a colon.
414 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
415 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
416 splits to reads and writes. The format is identical to what the
417 \fBbs\fR option accepts, the read and write parts are separated with a
420 .B blocksize_unaligned\fR,\fP bs_unaligned
421 If set, any size in \fBblocksize_range\fR may be used. This typically won't
422 work with direct I/O, as that normally requires sector alignment.
424 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
425 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
426 the minimum blocksize given. Minimum alignment is typically 512b
427 for using direct IO, though it usually depends on the hardware block size.
428 This option is mutually exclusive with using a random map for files, so it
429 will turn off that option.
431 .BI bs_is_seq_rand \fR=\fPbool
432 If this option is set, fio will use the normal read,write blocksize settings as
433 sequential,random instead. Any random read or write will use the WRITE
434 blocksize settings, and any sequential read or write will use the READ
438 Initialise buffers with all zeros. Default: fill buffers with random data.
439 The resulting IO buffers will not be completely zeroed, unless
440 \fPscramble_buffers\fR is also turned off.
443 If this option is given, fio will refill the IO buffers on every submit. The
444 default is to only fill it at init time and reuse that data. Only makes sense
445 if zero_buffers isn't specified, naturally. If data verification is enabled,
446 refill_buffers is also automatically enabled.
448 .BI scramble_buffers \fR=\fPbool
449 If \fBrefill_buffers\fR is too costly and the target is using data
450 deduplication, then setting this option will slightly modify the IO buffer
451 contents to defeat normal de-dupe attempts. This is not enough to defeat
452 more clever block compression attempts, but it will stop naive dedupe
453 of blocks. Default: true.
455 .BI buffer_compress_percentage \fR=\fPint
456 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
457 that compress to the specified level. Fio does this by providing a mix of
458 random data and zeroes. Note that this is per block size unit, for file/disk
459 wide compression level that matches this setting, you'll also want to set
460 \fBrefill_buffers\fR.
462 .BI buffer_compress_chunk \fR=\fPint
463 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
464 big the ranges of random data and zeroed data is. Without this set, fio will
465 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
466 the remaining zeroed. With this set to some chunk size smaller than the block
467 size, fio can alternate random and zeroed data throughout the IO buffer.
469 .BI buffer_pattern \fR=\fPstr
470 If set, fio will fill the io buffers with this pattern. If not set, the contents
471 of io buffers is defined by the other options related to buffer contents. The
472 setting can be any pattern of bytes, and can be prefixed with 0x for hex
475 .BI nrfiles \fR=\fPint
476 Number of files to use for this job. Default: 1.
478 .BI openfiles \fR=\fPint
479 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
481 .BI file_service_type \fR=\fPstr
482 Defines how files to service are selected. The following types are defined:
487 Choose a file at random.
490 Round robin over open files (default).
493 Do each file in the set sequentially.
496 The number of I/Os to issue before switching a new file can be specified by
497 appending `:\fIint\fR' to the service type.
500 .BI ioengine \fR=\fPstr
501 Defines how the job issues I/O. The following types are defined:
506 Basic \fBread\fR\|(2) or \fBwrite\fR\|(2) I/O. \fBfseek\fR\|(2) is used to
507 position the I/O location.
510 Basic \fBpread\fR\|(2) or \fBpwrite\fR\|(2) I/O.
513 Basic \fBreadv\fR\|(2) or \fBwritev\fR\|(2) I/O. Will emulate queuing by
514 coalescing adjacent IOs into a single submission.
517 Basic \fBpreadv\fR\|(2) or \fBpwritev\fR\|(2) I/O.
520 Linux native asynchronous I/O. This ioengine defines engine specific options.
523 POSIX asynchronous I/O using \fBaio_read\fR\|(3) and \fBaio_write\fR\|(3).
526 Solaris native asynchronous I/O.
529 Windows native asynchronous I/O.
532 File is memory mapped with \fBmmap\fR\|(2) and data copied using
536 \fBsplice\fR\|(2) is used to transfer the data and \fBvmsplice\fR\|(2) to
537 transfer data from user-space to the kernel.
540 Use the syslet system calls to make regular read/write asynchronous.
543 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
544 the target is an sg character device, we use \fBread\fR\|(2) and
545 \fBwrite\fR\|(2) for asynchronous I/O.
548 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
549 itself and for debugging and testing purposes.
552 Transfer over the network. The protocol to be used can be defined with the
553 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
554 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
555 This ioengine defines engine specific options.
558 Like \fBnet\fR, but uses \fBsplice\fR\|(2) and \fBvmsplice\fR\|(2) to map data
559 and send/receive. This ioengine defines engine specific options.
562 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
563 \fBcpucycles\fR parameters.
566 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
567 approach to asynchronous I/O.
569 See <http://www.xmailserver.org/guasi\-lib.html>.
572 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
573 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
576 Loads an external I/O engine object file. Append the engine filename as
580 IO engine that does regular linux native fallocate call to simulate data
581 transfer as fio ioengine
583 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
585 DIR_WRITE does fallocate(,mode = 0)
587 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
590 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
591 request to DDIR_WRITE event
594 IO engine supporting direct access to Ceph Rados Block Devices (RBD) via librbd
595 without the need to use the kernel rbd driver. This ioengine defines engine specific
599 Using Glusterfs libgfapi to direct access to Glusterfs volumes without
600 having to go through FUSE. This ioengine defines engine specific
606 .BI iodepth \fR=\fPint
607 Number of I/O units to keep in flight against the file. Note that increasing
608 iodepth beyond 1 will not affect synchronous ioengines (except for small
609 degress when verify_async is in use). Even async engines my impose OS
610 restrictions causing the desired depth not to be achieved. This may happen on
611 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
612 not async on that OS. Keep an eye on the IO depth distribution in the
613 fio output to verify that the achieved depth is as expected. Default: 1.
615 .BI iodepth_batch \fR=\fPint
616 Number of I/Os to submit at once. Default: \fBiodepth\fR.
618 .BI iodepth_batch_complete \fR=\fPint
619 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
620 means that we'll ask for a minimum of 1 IO in the retrieval process from the
621 kernel. The IO retrieval will go on until we hit the limit set by
622 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
623 completed events before queuing more IO. This helps reduce IO latency, at the
624 cost of more retrieval system calls.
626 .BI iodepth_low \fR=\fPint
627 Low watermark indicating when to start filling the queue again. Default:
630 .BI direct \fR=\fPbool
631 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
633 .BI atomic \fR=\fPbool
634 If value is true, attempt to use atomic direct IO. Atomic writes are guaranteed
635 to be stable once acknowledged by the operating system. Only Linux supports
638 .BI buffered \fR=\fPbool
639 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
642 .BI offset \fR=\fPint
643 Offset in the file to start I/O. Data before the offset will not be touched.
645 .BI offset_increment \fR=\fPint
646 If this is provided, then the real offset becomes the
647 offset + offset_increment * thread_number, where the thread number is a counter
648 that starts at 0 and is incremented for each job. This option is useful if
649 there are several jobs which are intended to operate on a file in parallel in
650 disjoint segments, with even spacing between the starting points.
652 .BI number_ios \fR=\fPint
653 Fio will normally perform IOs until it has exhausted the size of the region
654 set by \fBsize\fR, or if it exhaust the allocated time (or hits an error
655 condition). With this setting, the range/size can be set independently of
656 the number of IOs to perform. When fio reaches this number, it will exit
657 normally and report status.
660 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
661 0, don't sync. Default: 0.
663 .BI fdatasync \fR=\fPint
664 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
665 data parts of the file. Default: 0.
667 .BI write_barrier \fR=\fPint
668 Make every Nth write a barrier write.
670 .BI sync_file_range \fR=\fPstr:int
671 Use \fBsync_file_range\fR\|(2) for every \fRval\fP number of write operations. Fio will
672 track range of writes that have happened since the last \fBsync_file_range\fR\|(2) call.
673 \fRstr\fP can currently be one or more of:
677 SYNC_FILE_RANGE_WAIT_BEFORE
680 SYNC_FILE_RANGE_WRITE
683 SYNC_FILE_RANGE_WRITE
687 So if you do sync_file_range=wait_before,write:8, fio would use
688 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
689 Also see the \fBsync_file_range\fR\|(2) man page. This option is Linux specific.
691 .BI overwrite \fR=\fPbool
692 If writing, setup the file first and do overwrites. Default: false.
694 .BI end_fsync \fR=\fPbool
695 Sync file contents when a write stage has completed. Default: false.
697 .BI fsync_on_close \fR=\fPbool
698 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
699 it will happen on every close, not just at the end of the job. Default: false.
701 .BI rwmixread \fR=\fPint
702 Percentage of a mixed workload that should be reads. Default: 50.
704 .BI rwmixwrite \fR=\fPint
705 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
706 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
707 overrides the first. This may interfere with a given rate setting, if fio is
708 asked to limit reads or writes to a certain rate. If that is the case, then
709 the distribution may be skewed. Default: 50.
711 .BI random_distribution \fR=\fPstr:float
712 By default, fio will use a completely uniform random distribution when asked
713 to perform random IO. Sometimes it is useful to skew the distribution in
714 specific ways, ensuring that some parts of the data is more hot than others.
715 Fio includes the following distribution models:
719 Uniform random distribution
729 When using a zipf or pareto distribution, an input value is also needed to
730 define the access pattern. For zipf, this is the zipf theta. For pareto,
731 it's the pareto power. Fio includes a test program, genzipf, that can be
732 used visualize what the given input values will yield in terms of hit rates.
733 If you wanted to use zipf with a theta of 1.2, you would use
734 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
735 fio will disable use of the random map.
737 .BI percentage_random \fR=\fPint
738 For a random workload, set how big a percentage should be random. This defaults
739 to 100%, in which case the workload is fully random. It can be set from
740 anywhere from 0 to 100. Setting it to 0 would make the workload fully
741 sequential. It is possible to set different values for reads, writes, and
742 trim. To do so, simply use a comma separated list. See \fBblocksize\fR.
745 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
746 this parameter is given, a new offset will be chosen without looking at past
747 I/O history. This parameter is mutually exclusive with \fBverify\fR.
749 .BI softrandommap \fR=\fPbool
750 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
751 fails to allocate the map, if this option is set it will continue without a
752 random block map. As coverage will not be as complete as with random maps, this
753 option is disabled by default.
755 .BI random_generator \fR=\fPstr
756 Fio supports the following engines for generating IO offsets for random IO:
760 Strong 2^88 cycle random number generator
763 Linear feedback shift register generator
767 Tausworthe is a strong random number generator, but it requires tracking on the
768 side if we want to ensure that blocks are only read or written once. LFSR
769 guarantees that we never generate the same offset twice, and it's also less
770 computationally expensive. It's not a true random generator, however, though
771 for IO purposes it's typically good enough. LFSR only works with single block
772 sizes, not with workloads that use multiple block sizes. If used with such a
773 workload, fio may read or write some blocks multiple times.
776 Run job with given nice value. See \fBnice\fR\|(2).
779 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
782 .BI prioclass \fR=\fPint
783 Set I/O priority class. See \fBionice\fR\|(1).
785 .BI thinktime \fR=\fPint
786 Stall job for given number of microseconds between issuing I/Os.
788 .BI thinktime_spin \fR=\fPint
789 Pretend to spend CPU time for given number of microseconds, sleeping the rest
790 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
792 .BI thinktime_blocks \fR=\fPint
793 Only valid if thinktime is set - control how many blocks to issue, before
794 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
795 make fio wait \fBthinktime\fR microseconds after every block. This
796 effectively makes any queue depth setting redundant, since no more than 1 IO
797 will be queued before we have to complete it and do our thinktime. In other
798 words, this setting effectively caps the queue depth if the latter is larger.
802 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
803 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
804 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
805 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
806 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
807 limit writes (to 500KB/sec), the latter will only limit reads.
809 .BI ratemin \fR=\fPint
810 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
811 Failing to meet this requirement will cause the job to exit. The same format
812 as \fBrate\fR is used for read vs write separation.
814 .BI rate_iops \fR=\fPint
815 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
816 specified independently of bandwidth. The same format as \fBrate\fR is used for
817 read vs write separation. If \fBblocksize\fR is a range, the smallest block
818 size is used as the metric.
820 .BI rate_iops_min \fR=\fPint
821 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
822 is used for read vs write separation.
824 .BI ratecycle \fR=\fPint
825 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
826 milliseconds. Default: 1000ms.
828 .BI latency_target \fR=\fPint
829 If set, fio will attempt to find the max performance point that the given
830 workload will run at while maintaining a latency below this target. The
831 values is given in microseconds. See \fBlatency_window\fR and
832 \fBlatency_percentile\fR.
834 .BI latency_window \fR=\fPint
835 Used with \fBlatency_target\fR to specify the sample window that the job
836 is run at varying queue depths to test the performance. The value is given
839 .BI latency_percentile \fR=\fPfloat
840 The percentage of IOs that must fall within the criteria specified by
841 \fBlatency_target\fR and \fBlatency_window\fR. If not set, this defaults
842 to 100.0, meaning that all IOs must be equal or below to the value set
843 by \fBlatency_target\fR.
845 .BI max_latency \fR=\fPint
846 If set, fio will exit the job if it exceeds this maximum latency. It will exit
849 .BI cpumask \fR=\fPint
850 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
851 may run on. See \fBsched_setaffinity\fR\|(2).
853 .BI cpus_allowed \fR=\fPstr
854 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
856 .BI cpus_allowed_policy \fR=\fPstr
857 Set the policy of how fio distributes the CPUs specified by \fBcpus_allowed\fR
858 or \fBcpumask\fR. Two policies are supported:
863 All jobs will share the CPU set specified.
866 Each job will get a unique CPU from the CPU set.
869 \fBshared\fR is the default behaviour, if the option isn't specified. If
870 \fBsplit\fR is specified, then fio will assign one cpu per job. If not enough
871 CPUs are given for the jobs listed, then fio will roundrobin the CPUs in
876 .BI numa_cpu_nodes \fR=\fPstr
877 Set this job running on specified NUMA nodes' CPUs. The arguments allow
878 comma delimited list of cpu numbers, A-B ranges, or 'all'.
880 .BI numa_mem_policy \fR=\fPstr
881 Set this job's memory policy and corresponding NUMA nodes. Format of
885 .B <mode>[:<nodelist>]
888 is one of the following memory policy:
890 .B default, prefer, bind, interleave, local
893 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
894 needed to be specified. For \fBprefer\fR, only one node is
895 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
896 comma delimited list of numbers, A-B ranges, or 'all'.
898 .BI startdelay \fR=\fPirange
899 Delay start of job for the specified number of seconds. Supports all time
900 suffixes to allow specification of hours, minutes, seconds and
901 milliseconds - seconds are the default if a unit is ommited.
902 Can be given as a range which causes each thread to choose randomly out of the
905 .BI runtime \fR=\fPint
906 Terminate processing after the specified number of seconds.
909 If given, run for the specified \fBruntime\fR duration even if the files are
910 completely read or written. The same workload will be repeated as many times
911 as \fBruntime\fR allows.
913 .BI ramp_time \fR=\fPint
914 If set, fio will run the specified workload for this amount of time before
915 logging any performance numbers. Useful for letting performance settle before
916 logging results, thus minimizing the runtime required for stable results. Note
917 that the \fBramp_time\fR is considered lead in time for a job, thus it will
918 increase the total runtime if a special timeout or runtime is specified.
920 .BI invalidate \fR=\fPbool
921 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
924 Use synchronous I/O for buffered writes. For the majority of I/O engines,
925 this means using O_SYNC. Default: false.
927 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
928 Allocation method for I/O unit buffer. Allowed values are:
933 Allocate memory with \fBmalloc\fR\|(3).
936 Use shared memory buffers allocated through \fBshmget\fR\|(2).
939 Same as \fBshm\fR, but use huge pages as backing.
942 Use \fBmmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
943 is given after the option in the format `:\fIfile\fR'.
946 Same as \fBmmap\fR, but use huge files as backing.
949 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
950 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
951 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
952 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
953 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
954 and the documentation for that. Normally you just need to echo an appropriate
955 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
959 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
960 This indicates the memory alignment of the IO memory buffers. Note that the
961 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
962 the alignment of the following buffers are given by the \fBbs\fR used. In
963 other words, if using a \fBbs\fR that is a multiple of the page sized in the
964 system, all buffers will be aligned to this value. If using a \fBbs\fR that
965 is not page aligned, the alignment of subsequent IO memory buffers is the
966 sum of the \fBiomem_align\fR and \fBbs\fR used.
968 .BI hugepage\-size \fR=\fPint
969 Defines the size of a huge page. Must be at least equal to the system setting.
970 Should be a multiple of 1MB. Default: 4MB.
973 Terminate all jobs when one finishes. Default: wait for each job to finish.
975 .BI bwavgtime \fR=\fPint
976 Average bandwidth calculations over the given time in milliseconds. Default:
979 .BI iopsavgtime \fR=\fPint
980 Average IOPS calculations over the given time in milliseconds. Default:
983 .BI create_serialize \fR=\fPbool
984 If true, serialize file creation for the jobs. Default: true.
986 .BI create_fsync \fR=\fPbool
987 \fBfsync\fR\|(2) data file after creation. Default: true.
989 .BI create_on_open \fR=\fPbool
990 If true, the files are not created until they are opened for IO by the job.
992 .BI create_only \fR=\fPbool
993 If true, fio will only run the setup phase of the job. If files need to be
994 laid out or updated on disk, only that will be done. The actual job contents
997 .BI pre_read \fR=\fPbool
998 If this is given, files will be pre-read into memory before starting the given
999 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
1000 pointless to pre-read and then drop the cache. This will only work for IO
1001 engines that are seekable, since they allow you to read the same data
1002 multiple times. Thus it will not work on eg network or splice IO.
1004 .BI unlink \fR=\fPbool
1005 Unlink job files when done. Default: false.
1007 .BI loops \fR=\fPint
1008 Specifies the number of iterations (runs of the same workload) of this job.
1011 .BI verify_only \fR=\fPbool
1012 Do not perform the specified workload, only verify data still matches previous
1013 invocation of this workload. This option allows one to check data multiple
1014 times at a later date without overwriting it. This option makes sense only for
1015 workloads that write data, and does not support workloads with the
1016 \fBtime_based\fR option set.
1018 .BI do_verify \fR=\fPbool
1019 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
1022 .BI verify \fR=\fPstr
1023 Method of verifying file contents after each iteration of the job. Allowed
1028 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 xxhash
1029 Store appropriate checksum in the header of each block. crc32c-intel is
1030 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
1031 not supported by the system.
1034 Write extra information about each I/O (timestamp, block number, etc.). The
1035 block number is verified. See \fBverify_pattern\fR as well.
1038 Pretend to verify. Used for testing internals.
1041 This option can be used for repeated burn-in tests of a system to make sure
1042 that the written data is also correctly read back. If the data direction given
1043 is a read or random read, fio will assume that it should verify a previously
1044 written file. If the data direction includes any form of write, the verify will
1045 be of the newly written data.
1048 .BI verifysort \fR=\fPbool
1049 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
1050 read them back in a sorted manner. Default: true.
1052 .BI verifysort_nr \fR=\fPint
1053 Pre-load and sort verify blocks for a read workload.
1055 .BI verify_offset \fR=\fPint
1056 Swap the verification header with data somewhere else in the block before
1057 writing. It is swapped back before verifying.
1059 .BI verify_interval \fR=\fPint
1060 Write the verification header for this number of bytes, which should divide
1061 \fBblocksize\fR. Default: \fBblocksize\fR.
1063 .BI verify_pattern \fR=\fPstr
1064 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
1065 with totally random bytes, but sometimes it's interesting to fill with a known
1066 pattern for io verification purposes. Depending on the width of the pattern,
1067 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
1068 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
1069 has to be a hex number that starts with either "0x" or "0X". Use with
1072 .BI verify_fatal \fR=\fPbool
1073 If true, exit the job on the first observed verification failure. Default:
1076 .BI verify_dump \fR=\fPbool
1077 If set, dump the contents of both the original data block and the data block we
1078 read off disk to files. This allows later analysis to inspect just what kind of
1079 data corruption occurred. Off by default.
1081 .BI verify_async \fR=\fPint
1082 Fio will normally verify IO inline from the submitting thread. This option
1083 takes an integer describing how many async offload threads to create for IO
1084 verification instead, causing fio to offload the duty of verifying IO contents
1085 to one or more separate threads. If using this offload option, even sync IO
1086 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
1087 allows them to have IO in flight while verifies are running.
1089 .BI verify_async_cpus \fR=\fPstr
1090 Tell fio to set the given CPU affinity on the async IO verification threads.
1091 See \fBcpus_allowed\fP for the format used.
1093 .BI verify_backlog \fR=\fPint
1094 Fio will normally verify the written contents of a job that utilizes verify
1095 once that job has completed. In other words, everything is written then
1096 everything is read back and verified. You may want to verify continually
1097 instead for a variety of reasons. Fio stores the meta data associated with an
1098 IO block in memory, so for large verify workloads, quite a bit of memory would
1099 be used up holding this meta data. If this option is enabled, fio will write
1100 only N blocks before verifying these blocks.
1102 .BI verify_backlog_batch \fR=\fPint
1103 Control how many blocks fio will verify if verify_backlog is set. If not set,
1104 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
1105 read back and verified). If \fBverify_backlog_batch\fR is less than
1106 \fBverify_backlog\fR then not all blocks will be verified, if
1107 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
1108 will be verified more than once.
1110 .BI trim_percentage \fR=\fPint
1111 Number of verify blocks to discard/trim.
1113 .BI trim_verify_zero \fR=\fPbool
1114 Verify that trim/discarded blocks are returned as zeroes.
1116 .BI trim_backlog \fR=\fPint
1117 Trim after this number of blocks are written.
1119 .BI trim_backlog_batch \fR=\fPint
1120 Trim this number of IO blocks.
1122 .BI experimental_verify \fR=\fPbool
1123 Enable experimental verification.
1125 .B stonewall "\fR,\fP wait_for_previous"
1126 Wait for preceding jobs in the job file to exit before starting this one.
1127 \fBstonewall\fR implies \fBnew_group\fR.
1130 Start a new reporting group. If not given, all jobs in a file will be part
1131 of the same reporting group, unless separated by a stonewall.
1133 .BI numjobs \fR=\fPint
1134 Number of clones (processes/threads performing the same workload) of this job.
1138 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
1142 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
1143 with \fBfork\fR\|(2).
1145 .BI zonesize \fR=\fPint
1146 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
1148 .BI zonerange \fR=\fPint
1149 Give size of an IO zone. See \fBzoneskip\fR.
1151 .BI zoneskip \fR=\fPint
1152 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
1155 .BI write_iolog \fR=\fPstr
1156 Write the issued I/O patterns to the specified file. Specify a separate file
1157 for each job, otherwise the iologs will be interspersed and the file may be
1160 .BI read_iolog \fR=\fPstr
1161 Replay the I/O patterns contained in the specified file generated by
1162 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1164 .BI replay_no_stall \fR=\fPint
1165 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1166 attempts to respect timing information between I/Os. Enabling
1167 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1168 still respecting ordering.
1170 .BI replay_redirect \fR=\fPstr
1171 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1172 is to replay the IOPS onto the major/minor device that each IOP was recorded
1173 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1174 single specified device regardless of the device it was recorded from.
1176 .BI write_bw_log \fR=\fPstr
1177 If given, write a bandwidth log of the jobs in this job file. Can be used to
1178 store data of the bandwidth of the jobs in their lifetime. The included
1179 fio_generate_plots script uses gnuplot to turn these text files into nice
1180 graphs. See \fBwrite_lat_log\fR for behaviour of given filename. For this
1181 option, the postfix is _bw.log.
1183 .BI write_lat_log \fR=\fPstr
1184 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1185 filename is given with this option, the default filename of "jobname_type.log"
1186 is used. Even if the filename is given, fio will still append the type of log.
1188 .BI write_iops_log \fR=\fPstr
1189 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1190 option, the default filename of "jobname_type.log" is used. Even if the
1191 filename is given, fio will still append the type of log.
1193 .BI log_avg_msec \fR=\fPint
1194 By default, fio will log an entry in the iops, latency, or bw log for every
1195 IO that completes. When writing to the disk log, that can quickly grow to a
1196 very large size. Setting this option makes fio average the each log entry
1197 over the specified period of time, reducing the resolution of the log.
1200 .BI disable_lat \fR=\fPbool
1201 Disable measurements of total latency numbers. Useful only for cutting
1202 back the number of calls to \fBgettimeofday\fR\|(2), as that does impact performance at
1203 really high IOPS rates. Note that to really get rid of a large amount of these
1204 calls, this option must be used with disable_slat and disable_bw as well.
1206 .BI disable_clat \fR=\fPbool
1207 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1209 .BI disable_slat \fR=\fPbool
1210 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1212 .BI disable_bw_measurement \fR=\fPbool
1213 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1215 .BI lockmem \fR=\fPint
1216 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1217 simulate a smaller amount of memory. The amount specified is per worker.
1219 .BI exec_prerun \fR=\fPstr
1220 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1222 Output is redirected in a file called \fBjobname.prerun.txt\fR
1225 .BI exec_postrun \fR=\fPstr
1226 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1228 Output is redirected in a file called \fBjobname.postrun.txt\fR
1231 .BI ioscheduler \fR=\fPstr
1232 Attempt to switch the device hosting the file to the specified I/O scheduler.
1234 .BI cpuload \fR=\fPint
1235 If the job is a CPU cycle-eater, attempt to use the specified percentage of
1238 .BI cpuchunks \fR=\fPint
1239 If the job is a CPU cycle-eater, split the load into cycles of the
1240 given time in milliseconds.
1242 .BI disk_util \fR=\fPbool
1243 Generate disk utilization statistics if the platform supports it. Default: true.
1245 .BI clocksource \fR=\fPstr
1246 Use the given clocksource as the base of timing. The supported options are:
1250 \fBgettimeofday\fR\|(2)
1253 \fBclock_gettime\fR\|(2)
1256 Internal CPU clock source
1260 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1261 (and fio is heavy on time calls). Fio will automatically use this clocksource
1262 if it's supported and considered reliable on the system it is running on,
1263 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1264 means supporting TSC Invariant.
1266 .BI gtod_reduce \fR=\fPbool
1267 Enable all of the \fBgettimeofday\fR\|(2) reducing options (disable_clat, disable_slat,
1268 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1269 \fBgettimeofday\fR\|(2) call count. With this option enabled, we only do about 0.4% of
1270 the gtod() calls we would have done if all time keeping was enabled.
1272 .BI gtod_cpu \fR=\fPint
1273 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1274 the current time. Fio (and databases, for instance) are very intensive on
1275 \fBgettimeofday\fR\|(2) calls. With this option, you can set one CPU aside for doing
1276 nothing but logging current time to a shared memory location. Then the other
1277 threads/processes that run IO workloads need only copy that segment, instead of
1278 entering the kernel with a \fBgettimeofday\fR\|(2) call. The CPU set aside for doing
1279 these time calls will be excluded from other uses. Fio will manually clear it
1280 from the CPU mask of other jobs.
1282 .BI ignore_error \fR=\fPstr
1283 Sometimes you want to ignore some errors during test in that case you can specify
1284 error list for each error type.
1286 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1288 errors for given error type is separated with ':'.
1289 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1291 Example: ignore_error=EAGAIN,ENOSPC:122 .
1293 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1295 .BI error_dump \fR=\fPbool
1296 If set dump every error even if it is non fatal, true by default. If disabled
1297 only fatal error will be dumped
1299 .BI profile \fR=\fPstr
1300 Select a specific builtin performance test.
1302 .BI cgroup \fR=\fPstr
1303 Add job to this control group. If it doesn't exist, it will be created.
1304 The system must have a mounted cgroup blkio mount point for this to work. If
1305 your system doesn't have it mounted, you can do so with:
1307 # mount \-t cgroup \-o blkio none /cgroup
1309 .BI cgroup_weight \fR=\fPint
1310 Set the weight of the cgroup to this value. See the documentation that comes
1311 with the kernel, allowed values are in the range of 100..1000.
1313 .BI cgroup_nodelete \fR=\fPbool
1314 Normally fio will delete the cgroups it has created after the job completion.
1315 To override this behavior and to leave cgroups around after the job completion,
1316 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1317 cgroup files after job completion. Default: false
1320 Instead of running as the invoking user, set the user ID to this value before
1321 the thread/process does any work.
1324 Set group ID, see \fBuid\fR.
1326 .BI unit_base \fR=\fPint
1327 Base unit for reporting. Allowed values are:
1331 Use auto-detection (default).
1341 .BI flow_id \fR=\fPint
1342 The ID of the flow. If not specified, it defaults to being a global flow. See
1346 Weight in token-based flow control. If this value is used, then there is a
1347 \fBflow counter\fR which is used to regulate the proportion of activity between
1348 two or more jobs. fio attempts to keep this flow counter near zero. The
1349 \fBflow\fR parameter stands for how much should be added or subtracted to the
1350 flow counter on each iteration of the main I/O loop. That is, if one job has
1351 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1352 1:8 ratio in how much one runs vs the other.
1354 .BI flow_watermark \fR=\fPint
1355 The maximum value that the absolute value of the flow counter is allowed to
1356 reach before the job must wait for a lower value of the counter.
1358 .BI flow_sleep \fR=\fPint
1359 The period of time, in microseconds, to wait after the flow watermark has been
1360 exceeded before retrying operations
1362 .BI clat_percentiles \fR=\fPbool
1363 Enable the reporting of percentiles of completion latencies.
1365 .BI percentile_list \fR=\fPfloat_list
1366 Overwrite the default list of percentiles for completion
1367 latencies. Each number is a floating number in the range (0,100], and
1368 the maximum length of the list is 20. Use ':' to separate the
1369 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1370 report the values of completion latency below which 99.5% and 99.9% of
1371 the observed latencies fell, respectively.
1372 .SS "Ioengine Parameters List"
1373 Some parameters are only valid when a specific ioengine is in use. These are
1374 used identically to normal parameters, with the caveat that when used on the
1375 command line, the must come after the ioengine that defines them is selected.
1377 .BI (cpu)cpuload \fR=\fPint
1378 Attempt to use the specified percentage of CPU cycles.
1380 .BI (cpu)cpuchunks \fR=\fPint
1381 Split the load into cycles of the given time. In microseconds.
1383 .BI (libaio)userspace_reap
1384 Normally, with the libaio engine in use, fio will use
1385 the io_getevents system call to reap newly returned events.
1386 With this flag turned on, the AIO ring will be read directly
1387 from user-space to reap events. The reaping mode is only
1388 enabled when polling for a minimum of 0 events (eg when
1389 iodepth_batch_complete=0).
1391 .BI (net,netsplice)hostname \fR=\fPstr
1392 The host name or IP address to use for TCP or UDP based IO.
1393 If the job is a TCP listener or UDP reader, the hostname is not
1394 used and must be omitted unless it is a valid UDP multicast address.
1396 .BI (net,netsplice)port \fR=\fPint
1397 The TCP or UDP port to bind to or connect to.
1399 .BI (net,netsplice)interface \fR=\fPstr
1400 The IP address of the network interface used to send or receive UDP multicast
1403 .BI (net,netsplice)ttl \fR=\fPint
1404 Time-to-live value for outgoing UDP multicast packets. Default: 1
1406 .BI (net,netsplice)nodelay \fR=\fPbool
1407 Set TCP_NODELAY on TCP connections.
1409 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1410 The network protocol to use. Accepted values are:
1415 Transmission control protocol
1418 Transmission control protocol V6
1421 User datagram protocol
1424 User datagram protocol V6
1430 When the protocol is TCP or UDP, the port must also be given,
1431 as well as the hostname if the job is a TCP listener or UDP
1432 reader. For unix sockets, the normal filename option should be
1433 used and the port is invalid.
1436 .BI (net,netsplice)listen
1437 For TCP network connections, tell fio to listen for incoming
1438 connections rather than initiating an outgoing connection. The
1439 hostname must be omitted if this option is used.
1441 .BI (net, pingpong) \fR=\fPbool
1442 Normally a network writer will just continue writing data, and a network reader
1443 will just consume packages. If pingpong=1 is set, a writer will send its normal
1444 payload to the reader, then wait for the reader to send the same payload back.
1445 This allows fio to measure network latencies. The submission and completion
1446 latencies then measure local time spent sending or receiving, and the
1447 completion latency measures how long it took for the other end to receive and
1448 send back. For UDP multicast traffic pingpong=1 should only be set for a single
1449 reader when multiple readers are listening to the same address.
1451 .BI (e4defrag,donorname) \fR=\fPstr
1452 File will be used as a block donor (swap extents between files)
1454 .BI (e4defrag,inplace) \fR=\fPint
1455 Configure donor file block allocation strategy
1458 Preallocate donor's file on init
1461 allocate space immediately inside defragment event, and free right after event
1464 .BI (rbd)rbdname \fR=\fPstr
1465 Specifies the name of the RBD.
1467 .BI (rbd)pool \fR=\fPstr
1468 Specifies the name of the Ceph pool containing the RBD.
1470 .BI (rbd)clientname \fR=\fPstr
1471 Specifies the username (without the 'client.' prefix) used to access the Ceph cluster.
1473 While running, \fBfio\fR will display the status of the created jobs. For
1477 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1480 The characters in the first set of brackets denote the current status of each
1481 threads. The possible values are:
1487 Setup but not started.
1493 Initialized, waiting.
1496 Running, doing sequential reads.
1499 Running, doing random reads.
1502 Running, doing sequential writes.
1505 Running, doing random writes.
1508 Running, doing mixed sequential reads/writes.
1511 Running, doing mixed random reads/writes.
1514 Running, currently waiting for \fBfsync\fR\|(2).
1517 Running, verifying written data.
1520 Exited, not reaped by main thread.
1523 Exited, thread reaped.
1527 The second set of brackets shows the estimated completion percentage of
1528 the current group. The third set shows the read and write I/O rate,
1529 respectively. Finally, the estimated run time of the job is displayed.
1531 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1532 for each thread, each group of threads, and each disk, in that order.
1534 Per-thread statistics first show the threads client number, group-id, and
1535 error code. The remaining figures are as follows:
1539 Number of megabytes of I/O performed.
1542 Average data rate (bandwidth).
1548 Submission latency minimum, maximum, average and standard deviation. This is
1549 the time it took to submit the I/O.
1552 Completion latency minimum, maximum, average and standard deviation. This
1553 is the time between submission and completion.
1556 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1557 and standard deviation.
1560 CPU usage statistics. Includes user and system time, number of context switches
1561 this thread went through and number of major and minor page faults.
1564 Distribution of I/O depths. Each depth includes everything less than (or equal)
1565 to it, but greater than the previous depth.
1568 Number of read/write requests issued, and number of short read/write requests.
1571 Distribution of I/O completion latencies. The numbers follow the same pattern
1575 The group statistics show:
1580 Number of megabytes I/O performed.
1583 Aggregate bandwidth of threads in the group.
1586 Minimum average bandwidth a thread saw.
1589 Maximum average bandwidth a thread saw.
1592 Shortest runtime of threads in the group.
1595 Longest runtime of threads in the group.
1599 Finally, disk statistics are printed with reads first:
1604 Number of I/Os performed by all groups.
1607 Number of merges in the I/O scheduler.
1610 Number of ticks we kept the disk busy.
1613 Total time spent in the disk queue.
1620 It is also possible to get fio to dump the current output while it is
1621 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1624 If the \fB\-\-minimal\fR / \fB\-\-append-terse\fR options are given, the
1625 results will be printed/appended in a semicolon-delimited format suitable for
1627 A job description (if provided) follows on a new line. Note that the first
1628 number in the line is the version number. If the output has to be changed
1629 for some reason, this number will be incremented by 1 to signify that
1630 change. The fields are:
1633 .B terse version, fio version, jobname, groupid, error
1637 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1641 .B min, max, mean, standard deviation
1645 .B min, max, mean, standard deviation
1647 Completion latency percentiles (20 fields):
1649 .B Xth percentile=usec
1653 .B min, max, mean, standard deviation
1657 .B min, max, aggregate percentage of total, mean, standard deviation
1663 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1667 .B min, max, mean, standard deviation
1671 .B min, max, mean, standard deviation
1673 Completion latency percentiles (20 fields):
1675 .B Xth percentile=usec
1679 .B min, max, mean, standard deviation
1683 .B min, max, aggregate percentage of total, mean, standard deviation
1689 .B user, system, context switches, major page faults, minor page faults
1692 IO depth distribution:
1694 .B <=1, 2, 4, 8, 16, 32, >=64
1697 IO latency distribution:
1701 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1705 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1709 Disk utilization (1 for each disk used):
1711 .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
1714 Error Info (dependent on continue_on_error, default off):
1716 .B total # errors, first error code
1719 .B text description (if provided in config - appears on newline)
1722 Normally you would run fio as a stand-alone application on the machine
1723 where the IO workload should be generated. However, it is also possible to
1724 run the frontend and backend of fio separately. This makes it possible to
1725 have a fio server running on the machine(s) where the IO workload should
1726 be running, while controlling it from another machine.
1728 To start the server, you would do:
1730 \fBfio \-\-server=args\fR
1732 on that machine, where args defines what fio listens to. The arguments
1733 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1734 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1735 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1736 listen to (only valid for TCP/IP, not a local socket). Some examples:
1740 Start a fio server, listening on all interfaces on the default port (8765).
1742 2) fio \-\-server=ip:hostname,4444
1744 Start a fio server, listening on IP belonging to hostname and on port 4444.
1746 3) fio \-\-server=ip6:::1,4444
1748 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1750 4) fio \-\-server=,4444
1752 Start a fio server, listening on all interfaces on port 4444.
1754 5) fio \-\-server=1.2.3.4
1756 Start a fio server, listening on IP 1.2.3.4 on the default port.
1758 6) fio \-\-server=sock:/tmp/fio.sock
1760 Start a fio server, listening on the local socket /tmp/fio.sock.
1762 When a server is running, you can connect to it from a client. The client
1765 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1767 where \-\-local-args are arguments that are local to the client where it is
1768 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1769 are sent to the server. The 'server' string follows the same format as it
1770 does on the server side, to allow IP/hostname/socket and port strings.
1771 You can connect to multiple clients as well, to do that you could run:
1773 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1777 was written by Jens Axboe <jens.axboe@oracle.com>,
1778 now Jens Axboe <jaxboe@fusionio.com>.
1780 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1781 on documentation by Jens Axboe.
1782 .SH "REPORTING BUGS"
1783 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1786 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1788 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob