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. This option can be used multiple times to add 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 a 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 fio sums the results and reports 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 generate random
330 offsets, or it can use its 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 \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
375 .BI io_limit \fR=\fPint
376 Normally fio operates within the region set by \fBsize\fR, which means that
377 the \fBsize\fR option sets both the region and size of IO to be performed.
378 Sometimes that is not what you want. With this option, it is possible to
379 define just the amount of IO that fio should do. For instance, if \fBsize\fR
380 is set to 20G and \fBio_limit\fR is set to 5G, fio will perform IO within
381 the first 20G but exit when 5G have been done.
383 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
384 Sets size to something really large and waits for ENOSPC (no space left on
385 device) as the terminating condition. Only makes sense with sequential write.
386 For a read workload, the mount point will be filled first then IO started on
387 the result. This option doesn't make sense if operating on a raw device node,
388 since the size of that is already known by the file system. Additionally,
389 writing beyond end-of-device will not return ENOSPC there.
391 .BI filesize \fR=\fPirange
392 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
393 for files at random within the given range, limited to \fBsize\fR in total (if
394 that is given). If \fBfilesize\fR is not specified, each created file is the
397 .BI file_append \fR=\fPbool
398 Perform IO after the end of the file. Normally fio will operate within the
399 size of a file. If this option is set, then fio will append to the file
400 instead. This has identical behavior to setting \fRoffset\fP to the size
401 of a file. This option is ignored on non-regular files.
403 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
404 Block size for I/O units. Default: 4k. Values for reads, writes, and trims
405 can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR
406 either of which may be empty to leave that value at its default. If a trailing
407 comma isn't given, the remainder will inherit the last value set.
409 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
410 Specify a range of I/O block sizes. The issued I/O unit will always be a
411 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
412 to both reads and writes if only one range is given, but can be specified
413 separately with a comma separating the values. Example: bsrange=1k-4k,2k-8k.
414 Also (see \fBblocksize\fR).
416 .BI bssplit \fR=\fPstr
417 This option allows even finer grained control of the block sizes issued,
418 not just even splits between them. With this option, you can weight various
419 block sizes for exact control of the issued IO for a job that has mixed
420 block sizes. The format of the option is bssplit=blocksize/percentage,
421 optionally adding as many definitions as needed separated by a colon.
422 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
423 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
424 splits to reads and writes. The format is identical to what the
425 \fBbs\fR option accepts, the read and write parts are separated with a
428 .B blocksize_unaligned\fR,\fP bs_unaligned
429 If set, any size in \fBblocksize_range\fR may be used. This typically won't
430 work with direct I/O, as that normally requires sector alignment.
432 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
433 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
434 the minimum blocksize given. Minimum alignment is typically 512b
435 for using direct IO, though it usually depends on the hardware block size.
436 This option is mutually exclusive with using a random map for files, so it
437 will turn off that option.
439 .BI bs_is_seq_rand \fR=\fPbool
440 If this option is set, fio will use the normal read,write blocksize settings as
441 sequential,random instead. Any random read or write will use the WRITE
442 blocksize settings, and any sequential read or write will use the READ
446 Initialize buffers with all zeros. Default: fill buffers with random data.
447 The resulting IO buffers will not be completely zeroed, unless
448 \fPscramble_buffers\fR is also turned off.
451 If this option is given, fio will refill the IO buffers on every submit. The
452 default is to only fill it at init time and reuse that data. Only makes sense
453 if zero_buffers isn't specified, naturally. If data verification is enabled,
454 refill_buffers is also automatically enabled.
456 .BI scramble_buffers \fR=\fPbool
457 If \fBrefill_buffers\fR is too costly and the target is using data
458 deduplication, then setting this option will slightly modify the IO buffer
459 contents to defeat normal de-dupe attempts. This is not enough to defeat
460 more clever block compression attempts, but it will stop naive dedupe
461 of blocks. Default: true.
463 .BI buffer_compress_percentage \fR=\fPint
464 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
465 that compress to the specified level. Fio does this by providing a mix of
466 random data and zeroes. Note that this is per block size unit, for file/disk
467 wide compression level that matches this setting, you'll also want to set
468 \fBrefill_buffers\fR.
470 .BI buffer_compress_chunk \fR=\fPint
471 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
472 big the ranges of random data and zeroed data is. Without this set, fio will
473 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
474 the remaining zeroed. With this set to some chunk size smaller than the block
475 size, fio can alternate random and zeroed data throughout the IO buffer.
477 .BI buffer_pattern \fR=\fPstr
478 If set, fio will fill the IO buffers with this pattern. If not set, the contents
479 of IO buffers is defined by the other options related to buffer contents. The
480 setting can be any pattern of bytes, and can be prefixed with 0x for hex
481 values. It may also be a string, where the string must then be wrapped with
484 .BI dedupe_percentage \fR=\fPint
485 If set, fio will generate this percentage of identical buffers when writing.
486 These buffers will be naturally dedupable. The contents of the buffers depend
487 on what other buffer compression settings have been set. It's possible to have
488 the individual buffers either fully compressible, or not at all. This option
489 only controls the distribution of unique buffers.
491 .BI nrfiles \fR=\fPint
492 Number of files to use for this job. Default: 1.
494 .BI openfiles \fR=\fPint
495 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
497 .BI file_service_type \fR=\fPstr
498 Defines how files to service are selected. The following types are defined:
503 Choose a file at random.
506 Round robin over opened files (default).
509 Do each file in the set sequentially.
512 The number of I/Os to issue before switching to a new file can be specified by
513 appending `:\fIint\fR' to the service type.
516 .BI ioengine \fR=\fPstr
517 Defines how the job issues I/O. The following types are defined:
522 Basic \fBread\fR\|(2) or \fBwrite\fR\|(2) I/O. \fBfseek\fR\|(2) is used to
523 position the I/O location.
526 Basic \fBpread\fR\|(2) or \fBpwrite\fR\|(2) I/O.
529 Basic \fBreadv\fR\|(2) or \fBwritev\fR\|(2) I/O. Will emulate queuing by
530 coalescing adjacent IOs into a single submission.
533 Basic \fBpreadv\fR\|(2) or \fBpwritev\fR\|(2) I/O.
536 Linux native asynchronous I/O. This ioengine defines engine specific options.
539 POSIX asynchronous I/O using \fBaio_read\fR\|(3) and \fBaio_write\fR\|(3).
542 Solaris native asynchronous I/O.
545 Windows native asynchronous I/O.
548 File is memory mapped with \fBmmap\fR\|(2) and data copied using
552 \fBsplice\fR\|(2) is used to transfer the data and \fBvmsplice\fR\|(2) to
553 transfer data from user-space to the kernel.
556 Use the syslet system calls to make regular read/write asynchronous.
559 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
560 the target is an sg character device, we use \fBread\fR\|(2) and
561 \fBwrite\fR\|(2) for asynchronous I/O.
564 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
565 itself and for debugging and testing purposes.
568 Transfer over the network. The protocol to be used can be defined with the
569 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
570 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
571 This ioengine defines engine specific options.
574 Like \fBnet\fR, but uses \fBsplice\fR\|(2) and \fBvmsplice\fR\|(2) to map data
575 and send/receive. This ioengine defines engine specific options.
578 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
579 \fBcpucycles\fR parameters.
582 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
583 approach to asynchronous I/O.
585 See <http://www.xmailserver.org/guasi\-lib.html>.
588 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
589 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
592 Loads an external I/O engine object file. Append the engine filename as
596 IO engine that does regular linux native fallocate call to simulate data
597 transfer as fio ioengine
599 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
601 DIR_WRITE does fallocate(,mode = 0)
603 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
606 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
607 request to DDIR_WRITE event
610 IO engine supporting direct access to Ceph Rados Block Devices (RBD) via librbd
611 without the need to use the kernel rbd driver. This ioengine defines engine specific
615 Using Glusterfs libgfapi sync interface to direct access to Glusterfs volumes without
616 having to go through FUSE. This ioengine defines engine specific
620 Using Glusterfs libgfapi async interface to direct access to Glusterfs volumes without
621 having to go through FUSE. This ioengine defines engine specific
625 Read and write through Hadoop (HDFS). The \fBfilename\fR option is used to
626 specify host,port of the hdfs name-node to connect. This engine interprets
627 offsets a little differently. In HDFS, files once created cannot be modified.
628 So random writes are not possible. To imitate this, libhdfs engine expects
629 bunch of small files to be created over HDFS, and engine will randomly pick a
630 file out of those files based on the offset generated by fio backend. (see the
631 example job file to create such files, use rw=write option). Please note, you
632 might want to set necessary environment variables to work with hdfs/libhdfs
638 .BI iodepth \fR=\fPint
639 Number of I/O units to keep in flight against the file. Note that increasing
640 iodepth beyond 1 will not affect synchronous ioengines (except for small
641 degress when verify_async is in use). Even async engines may impose OS
642 restrictions causing the desired depth not to be achieved. This may happen on
643 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
644 not async on that OS. Keep an eye on the IO depth distribution in the
645 fio output to verify that the achieved depth is as expected. Default: 1.
647 .BI iodepth_batch \fR=\fPint
648 Number of I/Os to submit at once. Default: \fBiodepth\fR.
650 .BI iodepth_batch_complete \fR=\fPint
651 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
652 means that we'll ask for a minimum of 1 IO in the retrieval process from the
653 kernel. The IO retrieval will go on until we hit the limit set by
654 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
655 completed events before queuing more IO. This helps reduce IO latency, at the
656 cost of more retrieval system calls.
658 .BI iodepth_low \fR=\fPint
659 Low watermark indicating when to start filling the queue again. Default:
662 .BI direct \fR=\fPbool
663 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
665 .BI atomic \fR=\fPbool
666 If value is true, attempt to use atomic direct IO. Atomic writes are guaranteed
667 to be stable once acknowledged by the operating system. Only Linux supports
670 .BI buffered \fR=\fPbool
671 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
674 .BI offset \fR=\fPint
675 Offset in the file to start I/O. Data before the offset will not be touched.
677 .BI offset_increment \fR=\fPint
678 If this is provided, then the real offset becomes the
679 offset + offset_increment * thread_number, where the thread number is a
680 counter that starts at 0 and is incremented for each sub-job (i.e. when
681 numjobs option is specified). This option is useful if there are several jobs
682 which are intended to operate on a file in parallel disjoint segments, with
683 even spacing between the starting points.
685 .BI number_ios \fR=\fPint
686 Fio will normally perform IOs until it has exhausted the size of the region
687 set by \fBsize\fR, or if it exhaust the allocated time (or hits an error
688 condition). With this setting, the range/size can be set independently of
689 the number of IOs to perform. When fio reaches this number, it will exit
690 normally and report status.
693 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
694 0, don't sync. Default: 0.
696 .BI fdatasync \fR=\fPint
697 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
698 data parts of the file. Default: 0.
700 .BI write_barrier \fR=\fPint
701 Make every Nth write a barrier write.
703 .BI sync_file_range \fR=\fPstr:int
704 Use \fBsync_file_range\fR\|(2) for every \fRval\fP number of write operations. Fio will
705 track range of writes that have happened since the last \fBsync_file_range\fR\|(2) call.
706 \fRstr\fP can currently be one or more of:
710 SYNC_FILE_RANGE_WAIT_BEFORE
713 SYNC_FILE_RANGE_WRITE
716 SYNC_FILE_RANGE_WRITE
720 So if you do sync_file_range=wait_before,write:8, fio would use
721 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
722 Also see the \fBsync_file_range\fR\|(2) man page. This option is Linux specific.
724 .BI overwrite \fR=\fPbool
725 If writing, setup the file first and do overwrites. Default: false.
727 .BI end_fsync \fR=\fPbool
728 Sync file contents when a write stage has completed. Default: false.
730 .BI fsync_on_close \fR=\fPbool
731 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
732 it will happen on every close, not just at the end of the job. Default: false.
734 .BI rwmixread \fR=\fPint
735 Percentage of a mixed workload that should be reads. Default: 50.
737 .BI rwmixwrite \fR=\fPint
738 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
739 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
740 overrides the first. This may interfere with a given rate setting, if fio is
741 asked to limit reads or writes to a certain rate. If that is the case, then
742 the distribution may be skewed. Default: 50.
744 .BI random_distribution \fR=\fPstr:float
745 By default, fio will use a completely uniform random distribution when asked
746 to perform random IO. Sometimes it is useful to skew the distribution in
747 specific ways, ensuring that some parts of the data is more hot than others.
748 Fio includes the following distribution models:
752 Uniform random distribution
762 When using a zipf or pareto distribution, an input value is also needed to
763 define the access pattern. For zipf, this is the zipf theta. For pareto,
764 it's the pareto power. Fio includes a test program, genzipf, that can be
765 used visualize what the given input values will yield in terms of hit rates.
766 If you wanted to use zipf with a theta of 1.2, you would use
767 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
768 fio will disable use of the random map.
770 .BI percentage_random \fR=\fPint
771 For a random workload, set how big a percentage should be random. This defaults
772 to 100%, in which case the workload is fully random. It can be set from
773 anywhere from 0 to 100. Setting it to 0 would make the workload fully
774 sequential. It is possible to set different values for reads, writes, and
775 trim. To do so, simply use a comma separated list. See \fBblocksize\fR.
778 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
779 this parameter is given, a new offset will be chosen without looking at past
780 I/O history. This parameter is mutually exclusive with \fBverify\fR.
782 .BI softrandommap \fR=\fPbool
783 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
784 fails to allocate the map, if this option is set it will continue without a
785 random block map. As coverage will not be as complete as with random maps, this
786 option is disabled by default.
788 .BI random_generator \fR=\fPstr
789 Fio supports the following engines for generating IO offsets for random IO:
793 Strong 2^88 cycle random number generator
796 Linear feedback shift register generator
800 Tausworthe is a strong random number generator, but it requires tracking on the
801 side if we want to ensure that blocks are only read or written once. LFSR
802 guarantees that we never generate the same offset twice, and it's also less
803 computationally expensive. It's not a true random generator, however, though
804 for IO purposes it's typically good enough. LFSR only works with single block
805 sizes, not with workloads that use multiple block sizes. If used with such a
806 workload, fio may read or write some blocks multiple times.
809 Run job with given nice value. See \fBnice\fR\|(2).
812 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
815 .BI prioclass \fR=\fPint
816 Set I/O priority class. See \fBionice\fR\|(1).
818 .BI thinktime \fR=\fPint
819 Stall job for given number of microseconds between issuing I/Os.
821 .BI thinktime_spin \fR=\fPint
822 Pretend to spend CPU time for given number of microseconds, sleeping the rest
823 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
825 .BI thinktime_blocks \fR=\fPint
826 Only valid if thinktime is set - control how many blocks to issue, before
827 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
828 make fio wait \fBthinktime\fR microseconds after every block. This
829 effectively makes any queue depth setting redundant, since no more than 1 IO
830 will be queued before we have to complete it and do our thinktime. In other
831 words, this setting effectively caps the queue depth if the latter is larger.
835 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
836 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
837 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
838 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
839 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
840 limit writes (to 500KB/sec), the latter will only limit reads.
842 .BI ratemin \fR=\fPint
843 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
844 Failing to meet this requirement will cause the job to exit. The same format
845 as \fBrate\fR is used for read vs write separation.
847 .BI rate_iops \fR=\fPint
848 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
849 specified independently of bandwidth. The same format as \fBrate\fR is used for
850 read vs write separation. If \fBblocksize\fR is a range, the smallest block
851 size is used as the metric.
853 .BI rate_iops_min \fR=\fPint
854 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
855 is used for read vs write separation.
857 .BI ratecycle \fR=\fPint
858 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
859 milliseconds. Default: 1000ms.
861 .BI latency_target \fR=\fPint
862 If set, fio will attempt to find the max performance point that the given
863 workload will run at while maintaining a latency below this target. The
864 values is given in microseconds. See \fBlatency_window\fR and
865 \fBlatency_percentile\fR.
867 .BI latency_window \fR=\fPint
868 Used with \fBlatency_target\fR to specify the sample window that the job
869 is run at varying queue depths to test the performance. The value is given
872 .BI latency_percentile \fR=\fPfloat
873 The percentage of IOs that must fall within the criteria specified by
874 \fBlatency_target\fR and \fBlatency_window\fR. If not set, this defaults
875 to 100.0, meaning that all IOs must be equal or below to the value set
876 by \fBlatency_target\fR.
878 .BI max_latency \fR=\fPint
879 If set, fio will exit the job if it exceeds this maximum latency. It will exit
882 .BI cpumask \fR=\fPint
883 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
884 may run on. See \fBsched_setaffinity\fR\|(2).
886 .BI cpus_allowed \fR=\fPstr
887 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
889 .BI cpus_allowed_policy \fR=\fPstr
890 Set the policy of how fio distributes the CPUs specified by \fBcpus_allowed\fR
891 or \fBcpumask\fR. Two policies are supported:
896 All jobs will share the CPU set specified.
899 Each job will get a unique CPU from the CPU set.
902 \fBshared\fR is the default behaviour, if the option isn't specified. If
903 \fBsplit\fR is specified, then fio will assign one cpu per job. If not enough
904 CPUs are given for the jobs listed, then fio will roundrobin the CPUs in
909 .BI numa_cpu_nodes \fR=\fPstr
910 Set this job running on specified NUMA nodes' CPUs. The arguments allow
911 comma delimited list of cpu numbers, A-B ranges, or 'all'.
913 .BI numa_mem_policy \fR=\fPstr
914 Set this job's memory policy and corresponding NUMA nodes. Format of
918 .B <mode>[:<nodelist>]
921 is one of the following memory policy:
923 .B default, prefer, bind, interleave, local
926 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
927 needed to be specified. For \fBprefer\fR, only one node is
928 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
929 comma delimited list of numbers, A-B ranges, or 'all'.
931 .BI startdelay \fR=\fPirange
932 Delay start of job for the specified number of seconds. Supports all time
933 suffixes to allow specification of hours, minutes, seconds and
934 milliseconds - seconds are the default if a unit is ommited.
935 Can be given as a range which causes each thread to choose randomly out of the
938 .BI runtime \fR=\fPint
939 Terminate processing after the specified number of seconds.
942 If given, run for the specified \fBruntime\fR duration even if the files are
943 completely read or written. The same workload will be repeated as many times
944 as \fBruntime\fR allows.
946 .BI ramp_time \fR=\fPint
947 If set, fio will run the specified workload for this amount of time before
948 logging any performance numbers. Useful for letting performance settle before
949 logging results, thus minimizing the runtime required for stable results. Note
950 that the \fBramp_time\fR is considered lead in time for a job, thus it will
951 increase the total runtime if a special timeout or runtime is specified.
953 .BI invalidate \fR=\fPbool
954 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
957 Use synchronous I/O for buffered writes. For the majority of I/O engines,
958 this means using O_SYNC. Default: false.
960 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
961 Allocation method for I/O unit buffer. Allowed values are:
966 Allocate memory with \fBmalloc\fR\|(3).
969 Use shared memory buffers allocated through \fBshmget\fR\|(2).
972 Same as \fBshm\fR, but use huge pages as backing.
975 Use \fBmmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
976 is given after the option in the format `:\fIfile\fR'.
979 Same as \fBmmap\fR, but use huge files as backing.
982 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
983 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
984 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
985 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
986 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
987 and the documentation for that. Normally you just need to echo an appropriate
988 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
992 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
993 This indicates the memory alignment of the IO memory buffers. Note that the
994 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
995 the alignment of the following buffers are given by the \fBbs\fR used. In
996 other words, if using a \fBbs\fR that is a multiple of the page sized in the
997 system, all buffers will be aligned to this value. If using a \fBbs\fR that
998 is not page aligned, the alignment of subsequent IO memory buffers is the
999 sum of the \fBiomem_align\fR and \fBbs\fR used.
1001 .BI hugepage\-size \fR=\fPint
1002 Defines the size of a huge page. Must be at least equal to the system setting.
1003 Should be a multiple of 1MB. Default: 4MB.
1006 Terminate all jobs when one finishes. Default: wait for each job to finish.
1008 .BI bwavgtime \fR=\fPint
1009 Average bandwidth calculations over the given time in milliseconds. Default:
1012 .BI iopsavgtime \fR=\fPint
1013 Average IOPS calculations over the given time in milliseconds. Default:
1016 .BI create_serialize \fR=\fPbool
1017 If true, serialize file creation for the jobs. Default: true.
1019 .BI create_fsync \fR=\fPbool
1020 \fBfsync\fR\|(2) data file after creation. Default: true.
1022 .BI create_on_open \fR=\fPbool
1023 If true, the files are not created until they are opened for IO by the job.
1025 .BI create_only \fR=\fPbool
1026 If true, fio will only run the setup phase of the job. If files need to be
1027 laid out or updated on disk, only that will be done. The actual job contents
1030 .BI pre_read \fR=\fPbool
1031 If this is given, files will be pre-read into memory before starting the given
1032 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
1033 pointless to pre-read and then drop the cache. This will only work for IO
1034 engines that are seekable, since they allow you to read the same data
1035 multiple times. Thus it will not work on eg network or splice IO.
1037 .BI unlink \fR=\fPbool
1038 Unlink job files when done. Default: false.
1040 .BI loops \fR=\fPint
1041 Specifies the number of iterations (runs of the same workload) of this job.
1044 .BI verify_only \fR=\fPbool
1045 Do not perform the specified workload, only verify data still matches previous
1046 invocation of this workload. This option allows one to check data multiple
1047 times at a later date without overwriting it. This option makes sense only for
1048 workloads that write data, and does not support workloads with the
1049 \fBtime_based\fR option set.
1051 .BI do_verify \fR=\fPbool
1052 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
1055 .BI verify \fR=\fPstr
1056 Method of verifying file contents after each iteration of the job. Allowed
1061 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 xxhash
1062 Store appropriate checksum in the header of each block. crc32c-intel is
1063 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
1064 not supported by the system.
1067 Write extra information about each I/O (timestamp, block number, etc.). The
1068 block number is verified. See \fBverify_pattern\fR as well.
1071 Pretend to verify. Used for testing internals.
1074 This option can be used for repeated burn-in tests of a system to make sure
1075 that the written data is also correctly read back. If the data direction given
1076 is a read or random read, fio will assume that it should verify a previously
1077 written file. If the data direction includes any form of write, the verify will
1078 be of the newly written data.
1081 .BI verifysort \fR=\fPbool
1082 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
1083 read them back in a sorted manner. Default: true.
1085 .BI verifysort_nr \fR=\fPint
1086 Pre-load and sort verify blocks for a read workload.
1088 .BI verify_offset \fR=\fPint
1089 Swap the verification header with data somewhere else in the block before
1090 writing. It is swapped back before verifying.
1092 .BI verify_interval \fR=\fPint
1093 Write the verification header for this number of bytes, which should divide
1094 \fBblocksize\fR. Default: \fBblocksize\fR.
1096 .BI verify_pattern \fR=\fPstr
1097 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
1098 with totally random bytes, but sometimes it's interesting to fill with a known
1099 pattern for io verification purposes. Depending on the width of the pattern,
1100 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
1101 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
1102 has to be a hex number that starts with either "0x" or "0X". Use with
1105 .BI verify_fatal \fR=\fPbool
1106 If true, exit the job on the first observed verification failure. Default:
1109 .BI verify_dump \fR=\fPbool
1110 If set, dump the contents of both the original data block and the data block we
1111 read off disk to files. This allows later analysis to inspect just what kind of
1112 data corruption occurred. Off by default.
1114 .BI verify_async \fR=\fPint
1115 Fio will normally verify IO inline from the submitting thread. This option
1116 takes an integer describing how many async offload threads to create for IO
1117 verification instead, causing fio to offload the duty of verifying IO contents
1118 to one or more separate threads. If using this offload option, even sync IO
1119 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
1120 allows them to have IO in flight while verifies are running.
1122 .BI verify_async_cpus \fR=\fPstr
1123 Tell fio to set the given CPU affinity on the async IO verification threads.
1124 See \fBcpus_allowed\fP for the format used.
1126 .BI verify_backlog \fR=\fPint
1127 Fio will normally verify the written contents of a job that utilizes verify
1128 once that job has completed. In other words, everything is written then
1129 everything is read back and verified. You may want to verify continually
1130 instead for a variety of reasons. Fio stores the meta data associated with an
1131 IO block in memory, so for large verify workloads, quite a bit of memory would
1132 be used up holding this meta data. If this option is enabled, fio will write
1133 only N blocks before verifying these blocks.
1135 .BI verify_backlog_batch \fR=\fPint
1136 Control how many blocks fio will verify if verify_backlog is set. If not set,
1137 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
1138 read back and verified). If \fBverify_backlog_batch\fR is less than
1139 \fBverify_backlog\fR then not all blocks will be verified, if
1140 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
1141 will be verified more than once.
1143 .BI trim_percentage \fR=\fPint
1144 Number of verify blocks to discard/trim.
1146 .BI trim_verify_zero \fR=\fPbool
1147 Verify that trim/discarded blocks are returned as zeroes.
1149 .BI trim_backlog \fR=\fPint
1150 Trim after this number of blocks are written.
1152 .BI trim_backlog_batch \fR=\fPint
1153 Trim this number of IO blocks.
1155 .BI experimental_verify \fR=\fPbool
1156 Enable experimental verification.
1158 .B stonewall "\fR,\fP wait_for_previous"
1159 Wait for preceding jobs in the job file to exit before starting this one.
1160 \fBstonewall\fR implies \fBnew_group\fR.
1163 Start a new reporting group. If not given, all jobs in a file will be part
1164 of the same reporting group, unless separated by a stonewall.
1166 .BI numjobs \fR=\fPint
1167 Number of clones (processes/threads performing the same workload) of this job.
1171 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
1175 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
1176 with \fBfork\fR\|(2).
1178 .BI zonesize \fR=\fPint
1179 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
1181 .BI zonerange \fR=\fPint
1182 Give size of an IO zone. See \fBzoneskip\fR.
1184 .BI zoneskip \fR=\fPint
1185 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
1188 .BI write_iolog \fR=\fPstr
1189 Write the issued I/O patterns to the specified file. Specify a separate file
1190 for each job, otherwise the iologs will be interspersed and the file may be
1193 .BI read_iolog \fR=\fPstr
1194 Replay the I/O patterns contained in the specified file generated by
1195 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1197 .BI replay_no_stall \fR=\fPint
1198 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1199 attempts to respect timing information between I/Os. Enabling
1200 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1201 still respecting ordering.
1203 .BI replay_redirect \fR=\fPstr
1204 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1205 is to replay the IOPS onto the major/minor device that each IOP was recorded
1206 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1207 single specified device regardless of the device it was recorded from.
1209 .BI write_bw_log \fR=\fPstr
1210 If given, write a bandwidth log of the jobs in this job file. Can be used to
1211 store data of the bandwidth of the jobs in their lifetime. The included
1212 fio_generate_plots script uses gnuplot to turn these text files into nice
1213 graphs. See \fBwrite_lat_log\fR for behaviour of given filename. For this
1214 option, the postfix is _bw.x.log, where x is the index of the job (1..N,
1215 where N is the number of jobs)
1217 .BI write_lat_log \fR=\fPstr
1218 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1219 filename is given with this option, the default filename of
1220 "jobname_type.x.log" is used, where x is the index of the job (1..N, where
1221 N is the number of jobs). Even if the filename is given, fio will still
1222 append the type of log.
1224 .BI write_iops_log \fR=\fPstr
1225 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1226 option, the default filename of "jobname_type.x.log" is used, where x is the
1227 index of the job (1..N, where N is the number of jobs). Even if the filename
1228 is given, fio will still append the type of log.
1230 .BI log_avg_msec \fR=\fPint
1231 By default, fio will log an entry in the iops, latency, or bw log for every
1232 IO that completes. When writing to the disk log, that can quickly grow to a
1233 very large size. Setting this option makes fio average the each log entry
1234 over the specified period of time, reducing the resolution of the log.
1237 .BI log_offset \fR=\fPbool
1238 If this is set, the iolog options will include the byte offset for the IO
1239 entry as well as the other data values.
1241 .BI log_compression \fR=\fPint
1242 If this is set, fio will compress the IO logs as it goes, to keep the memory
1243 footprint lower. When a log reaches the specified size, that chunk is removed
1244 and compressed in the background. Given that IO logs are fairly highly
1245 compressible, this yields a nice memory savings for longer runs. The downside
1246 is that the compression will consume some background CPU cycles, so it may
1247 impact the run. This, however, is also true if the logging ends up consuming
1248 most of the system memory. So pick your poison. The IO logs are saved
1249 normally at the end of a run, by decompressing the chunks and storing them
1250 in the specified log file. This feature depends on the availability of zlib.
1252 .BI log_store_compressed \fR=\fPbool
1253 If set, and \fBlog\fR_compression is also set, fio will store the log files in
1254 a compressed format. They can be decompressed with fio, using the
1255 \fB\-\-inflate-log\fR command line parameter. The files will be stored with a
1258 .BI disable_lat \fR=\fPbool
1259 Disable measurements of total latency numbers. Useful only for cutting
1260 back the number of calls to \fBgettimeofday\fR\|(2), as that does impact performance at
1261 really high IOPS rates. Note that to really get rid of a large amount of these
1262 calls, this option must be used with disable_slat and disable_bw as well.
1264 .BI disable_clat \fR=\fPbool
1265 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1267 .BI disable_slat \fR=\fPbool
1268 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1270 .BI disable_bw_measurement \fR=\fPbool
1271 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1273 .BI lockmem \fR=\fPint
1274 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1275 simulate a smaller amount of memory. The amount specified is per worker.
1277 .BI exec_prerun \fR=\fPstr
1278 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1280 Output is redirected in a file called \fBjobname.prerun.txt\fR
1283 .BI exec_postrun \fR=\fPstr
1284 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1286 Output is redirected in a file called \fBjobname.postrun.txt\fR
1289 .BI ioscheduler \fR=\fPstr
1290 Attempt to switch the device hosting the file to the specified I/O scheduler.
1292 .BI disk_util \fR=\fPbool
1293 Generate disk utilization statistics if the platform supports it. Default: true.
1295 .BI clocksource \fR=\fPstr
1296 Use the given clocksource as the base of timing. The supported options are:
1300 \fBgettimeofday\fR\|(2)
1303 \fBclock_gettime\fR\|(2)
1306 Internal CPU clock source
1310 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1311 (and fio is heavy on time calls). Fio will automatically use this clocksource
1312 if it's supported and considered reliable on the system it is running on,
1313 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1314 means supporting TSC Invariant.
1316 .BI gtod_reduce \fR=\fPbool
1317 Enable all of the \fBgettimeofday\fR\|(2) reducing options (disable_clat, disable_slat,
1318 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1319 \fBgettimeofday\fR\|(2) call count. With this option enabled, we only do about 0.4% of
1320 the gtod() calls we would have done if all time keeping was enabled.
1322 .BI gtod_cpu \fR=\fPint
1323 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1324 the current time. Fio (and databases, for instance) are very intensive on
1325 \fBgettimeofday\fR\|(2) calls. With this option, you can set one CPU aside for doing
1326 nothing but logging current time to a shared memory location. Then the other
1327 threads/processes that run IO workloads need only copy that segment, instead of
1328 entering the kernel with a \fBgettimeofday\fR\|(2) call. The CPU set aside for doing
1329 these time calls will be excluded from other uses. Fio will manually clear it
1330 from the CPU mask of other jobs.
1332 .BI ignore_error \fR=\fPstr
1333 Sometimes you want to ignore some errors during test in that case you can specify
1334 error list for each error type.
1336 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1338 errors for given error type is separated with ':'.
1339 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1341 Example: ignore_error=EAGAIN,ENOSPC:122 .
1343 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1345 .BI error_dump \fR=\fPbool
1346 If set dump every error even if it is non fatal, true by default. If disabled
1347 only fatal error will be dumped
1349 .BI profile \fR=\fPstr
1350 Select a specific builtin performance test.
1352 .BI cgroup \fR=\fPstr
1353 Add job to this control group. If it doesn't exist, it will be created.
1354 The system must have a mounted cgroup blkio mount point for this to work. If
1355 your system doesn't have it mounted, you can do so with:
1357 # mount \-t cgroup \-o blkio none /cgroup
1359 .BI cgroup_weight \fR=\fPint
1360 Set the weight of the cgroup to this value. See the documentation that comes
1361 with the kernel, allowed values are in the range of 100..1000.
1363 .BI cgroup_nodelete \fR=\fPbool
1364 Normally fio will delete the cgroups it has created after the job completion.
1365 To override this behavior and to leave cgroups around after the job completion,
1366 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1367 cgroup files after job completion. Default: false
1370 Instead of running as the invoking user, set the user ID to this value before
1371 the thread/process does any work.
1374 Set group ID, see \fBuid\fR.
1376 .BI unit_base \fR=\fPint
1377 Base unit for reporting. Allowed values are:
1381 Use auto-detection (default).
1391 .BI flow_id \fR=\fPint
1392 The ID of the flow. If not specified, it defaults to being a global flow. See
1396 Weight in token-based flow control. If this value is used, then there is a
1397 \fBflow counter\fR which is used to regulate the proportion of activity between
1398 two or more jobs. fio attempts to keep this flow counter near zero. The
1399 \fBflow\fR parameter stands for how much should be added or subtracted to the
1400 flow counter on each iteration of the main I/O loop. That is, if one job has
1401 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1402 1:8 ratio in how much one runs vs the other.
1404 .BI flow_watermark \fR=\fPint
1405 The maximum value that the absolute value of the flow counter is allowed to
1406 reach before the job must wait for a lower value of the counter.
1408 .BI flow_sleep \fR=\fPint
1409 The period of time, in microseconds, to wait after the flow watermark has been
1410 exceeded before retrying operations
1412 .BI clat_percentiles \fR=\fPbool
1413 Enable the reporting of percentiles of completion latencies.
1415 .BI percentile_list \fR=\fPfloat_list
1416 Overwrite the default list of percentiles for completion
1417 latencies. Each number is a floating number in the range (0,100], and
1418 the maximum length of the list is 20. Use ':' to separate the
1419 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1420 report the values of completion latency below which 99.5% and 99.9% of
1421 the observed latencies fell, respectively.
1422 .SS "Ioengine Parameters List"
1423 Some parameters are only valid when a specific ioengine is in use. These are
1424 used identically to normal parameters, with the caveat that when used on the
1425 command line, they must come after the ioengine.
1427 .BI (cpu)cpuload \fR=\fPint
1428 Attempt to use the specified percentage of CPU cycles.
1430 .BI (cpu)cpuchunks \fR=\fPint
1431 Split the load into cycles of the given time. In microseconds.
1433 .BI (cpu)exit_on_io_done \fR=\fPbool
1434 Detect when IO threads are done, then exit.
1436 .BI (libaio)userspace_reap
1437 Normally, with the libaio engine in use, fio will use
1438 the io_getevents system call to reap newly returned events.
1439 With this flag turned on, the AIO ring will be read directly
1440 from user-space to reap events. The reaping mode is only
1441 enabled when polling for a minimum of 0 events (eg when
1442 iodepth_batch_complete=0).
1444 .BI (net,netsplice)hostname \fR=\fPstr
1445 The host name or IP address to use for TCP or UDP based IO.
1446 If the job is a TCP listener or UDP reader, the hostname is not
1447 used and must be omitted unless it is a valid UDP multicast address.
1449 .BI (net,netsplice)port \fR=\fPint
1450 The TCP or UDP port to bind to or connect to.
1452 .BI (net,netsplice)interface \fR=\fPstr
1453 The IP address of the network interface used to send or receive UDP multicast
1456 .BI (net,netsplice)ttl \fR=\fPint
1457 Time-to-live value for outgoing UDP multicast packets. Default: 1
1459 .BI (net,netsplice)nodelay \fR=\fPbool
1460 Set TCP_NODELAY on TCP connections.
1462 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1463 The network protocol to use. Accepted values are:
1468 Transmission control protocol
1471 Transmission control protocol V6
1474 User datagram protocol
1477 User datagram protocol V6
1483 When the protocol is TCP or UDP, the port must also be given,
1484 as well as the hostname if the job is a TCP listener or UDP
1485 reader. For unix sockets, the normal filename option should be
1486 used and the port is invalid.
1489 .BI (net,netsplice)listen
1490 For TCP network connections, tell fio to listen for incoming
1491 connections rather than initiating an outgoing connection. The
1492 hostname must be omitted if this option is used.
1494 .BI (net, pingpong) \fR=\fPbool
1495 Normally a network writer will just continue writing data, and a network reader
1496 will just consume packets. If pingpong=1 is set, a writer will send its normal
1497 payload to the reader, then wait for the reader to send the same payload back.
1498 This allows fio to measure network latencies. The submission and completion
1499 latencies then measure local time spent sending or receiving, and the
1500 completion latency measures how long it took for the other end to receive and
1501 send back. For UDP multicast traffic pingpong=1 should only be set for a single
1502 reader when multiple readers are listening to the same address.
1504 .BI (e4defrag,donorname) \fR=\fPstr
1505 File will be used as a block donor (swap extents between files)
1507 .BI (e4defrag,inplace) \fR=\fPint
1508 Configure donor file block allocation strategy
1511 Preallocate donor's file on init
1514 allocate space immediately inside defragment event, and free right after event
1517 .BI (rbd)rbdname \fR=\fPstr
1518 Specifies the name of the RBD.
1520 .BI (rbd)pool \fR=\fPstr
1521 Specifies the name of the Ceph pool containing the RBD.
1523 .BI (rbd)clientname \fR=\fPstr
1524 Specifies the username (without the 'client.' prefix) used to access the Ceph cluster.
1526 While running, \fBfio\fR will display the status of the created jobs. For
1530 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1533 The characters in the first set of brackets denote the current status of each
1534 threads. The possible values are:
1540 Setup but not started.
1546 Initialized, waiting.
1549 Running, doing sequential reads.
1552 Running, doing random reads.
1555 Running, doing sequential writes.
1558 Running, doing random writes.
1561 Running, doing mixed sequential reads/writes.
1564 Running, doing mixed random reads/writes.
1567 Running, currently waiting for \fBfsync\fR\|(2).
1570 Running, verifying written data.
1573 Exited, not reaped by main thread.
1576 Exited, thread reaped.
1580 The second set of brackets shows the estimated completion percentage of
1581 the current group. The third set shows the read and write I/O rate,
1582 respectively. Finally, the estimated run time of the job is displayed.
1584 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1585 for each thread, each group of threads, and each disk, in that order.
1587 Per-thread statistics first show the threads client number, group-id, and
1588 error code. The remaining figures are as follows:
1592 Number of megabytes of I/O performed.
1595 Average data rate (bandwidth).
1601 Submission latency minimum, maximum, average and standard deviation. This is
1602 the time it took to submit the I/O.
1605 Completion latency minimum, maximum, average and standard deviation. This
1606 is the time between submission and completion.
1609 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1610 and standard deviation.
1613 CPU usage statistics. Includes user and system time, number of context switches
1614 this thread went through and number of major and minor page faults.
1617 Distribution of I/O depths. Each depth includes everything less than (or equal)
1618 to it, but greater than the previous depth.
1621 Number of read/write requests issued, and number of short read/write requests.
1624 Distribution of I/O completion latencies. The numbers follow the same pattern
1628 The group statistics show:
1633 Number of megabytes I/O performed.
1636 Aggregate bandwidth of threads in the group.
1639 Minimum average bandwidth a thread saw.
1642 Maximum average bandwidth a thread saw.
1645 Shortest runtime of threads in the group.
1648 Longest runtime of threads in the group.
1652 Finally, disk statistics are printed with reads first:
1657 Number of I/Os performed by all groups.
1660 Number of merges in the I/O scheduler.
1663 Number of ticks we kept the disk busy.
1666 Total time spent in the disk queue.
1673 It is also possible to get fio to dump the current output while it is
1674 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1677 If the \fB\-\-minimal\fR / \fB\-\-append-terse\fR options are given, the
1678 results will be printed/appended in a semicolon-delimited format suitable for
1680 A job description (if provided) follows on a new line. Note that the first
1681 number in the line is the version number. If the output has to be changed
1682 for some reason, this number will be incremented by 1 to signify that
1683 change. The fields are:
1686 .B terse version, fio version, jobname, groupid, error
1690 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1694 .B min, max, mean, standard deviation
1698 .B min, max, mean, standard deviation
1700 Completion latency percentiles (20 fields):
1702 .B Xth percentile=usec
1706 .B min, max, mean, standard deviation
1710 .B min, max, aggregate percentage of total, mean, standard deviation
1716 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1720 .B min, max, mean, standard deviation
1724 .B min, max, mean, standard deviation
1726 Completion latency percentiles (20 fields):
1728 .B Xth percentile=usec
1732 .B min, max, mean, standard deviation
1736 .B min, max, aggregate percentage of total, mean, standard deviation
1742 .B user, system, context switches, major page faults, minor page faults
1745 IO depth distribution:
1747 .B <=1, 2, 4, 8, 16, 32, >=64
1750 IO latency distribution:
1754 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1758 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1762 Disk utilization (1 for each disk used):
1764 .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
1767 Error Info (dependent on continue_on_error, default off):
1769 .B total # errors, first error code
1772 .B text description (if provided in config - appears on newline)
1775 Normally you would run fio as a stand-alone application on the machine
1776 where the IO workload should be generated. However, it is also possible to
1777 run the frontend and backend of fio separately. This makes it possible to
1778 have a fio server running on the machine(s) where the IO workload should
1779 be running, while controlling it from another machine.
1781 To start the server, you would do:
1783 \fBfio \-\-server=args\fR
1785 on that machine, where args defines what fio listens to. The arguments
1786 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1787 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1788 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1789 listen to (only valid for TCP/IP, not a local socket). Some examples:
1793 Start a fio server, listening on all interfaces on the default port (8765).
1795 2) fio \-\-server=ip:hostname,4444
1797 Start a fio server, listening on IP belonging to hostname and on port 4444.
1799 3) fio \-\-server=ip6:::1,4444
1801 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1803 4) fio \-\-server=,4444
1805 Start a fio server, listening on all interfaces on port 4444.
1807 5) fio \-\-server=1.2.3.4
1809 Start a fio server, listening on IP 1.2.3.4 on the default port.
1811 6) fio \-\-server=sock:/tmp/fio.sock
1813 Start a fio server, listening on the local socket /tmp/fio.sock.
1815 When a server is running, you can connect to it from a client. The client
1818 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1820 where \-\-local-args are arguments that are local to the client where it is
1821 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1822 are sent to the server. The 'server' string follows the same format as it
1823 does on the server side, to allow IP/hostname/socket and port strings.
1824 You can connect to multiple clients as well, to do that you could run:
1826 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1830 was written by Jens Axboe <jens.axboe@oracle.com>,
1831 now Jens Axboe <jaxboe@fusionio.com>.
1833 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1834 on documentation by Jens Axboe.
1835 .SH "REPORTING BUGS"
1836 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1839 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1841 Sample jobfiles are available in the \fBexamples\fR directory.