1 .TH fio 1 "December 2014" "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 \-\-output-format \fR=\fPformat
24 Set the reporting format to \fInormal\fR, \fIterse\fR, or \fIjson\fR.
26 .BI \-\-runtime \fR=\fPruntime
27 Limit run time to \fIruntime\fR seconds.
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 or set of hosts. See client/server section.
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.
119 Anywhere a numeric value is required, an arithmetic expression may be used,
120 provided it is surrounded by parentheses. Supported operators are:
128 .B multiplication (*)
134 .B exponentiation (^)
138 For time values in expressions, units are microseconds by default. This is
139 different than for time values not in expressions (not enclosed in
140 parentheses). The types used are:
143 String: a sequence of alphanumeric characters.
146 SI integer: a whole number, possibly containing a suffix denoting the base unit
147 of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
148 kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
149 respectively. If prefixed with '0x', the value is assumed to be base 16
150 (hexadecimal). A suffix may include a trailing 'b', for instance 'kb' is
151 identical to 'k'. You can specify a base 10 value by using 'KiB', 'MiB','GiB',
152 etc. This is useful for disk drives where values are often given in base 10
153 values. Specifying '30GiB' will get you 30*1000^3 bytes.
154 When specifying times the default suffix meaning changes, still denoting the
155 base unit of the value, but accepted suffixes are 'D' (days), 'H' (hours), 'M'
156 (minutes), 'S' Seconds, 'ms' (or msec) milli seconds, 'us' (or 'usec') micro
157 seconds. Time values without a unit specify seconds.
158 The suffixes are not case sensitive.
161 Boolean: a true or false value. `0' denotes false, `1' denotes true.
164 Integer range: a range of integers specified in the format
165 \fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
166 \fIupper\fR may contain a suffix as described above. If an option allows two
167 sets of ranges, they are separated with a `,' or `/' character. For example:
171 List of floating numbers: A list of floating numbers, separated by
176 May be used to override the job name. On the command line, this parameter
177 has the special purpose of signalling the start of a new job.
179 .BI description \fR=\fPstr
180 Human-readable description of the job. It is printed when the job is run, but
181 otherwise has no special purpose.
183 .BI directory \fR=\fPstr
184 Prefix filenames with this directory. Used to place files in a location other
186 You can specify a number of directories by separating the names with a ':'
187 character. These directories will be assigned equally distributed to job clones
188 creates with \fInumjobs\fR as long as they are using generated filenames.
189 If specific \fIfilename(s)\fR are set fio will use the first listed directory,
190 and thereby matching the \fIfilename\fR semantic which generates a file each
191 clone if not specified, but let all clones use the same if set. See
192 \fIfilename\fR for considerations regarding escaping certain characters on
195 .BI filename \fR=\fPstr
197 normally makes up a file name based on the job name, thread number, and file
198 number. If you want to share files between threads in a job or several jobs,
199 specify a \fIfilename\fR for each of them to override the default.
200 If the I/O engine is file-based, you can specify
201 a number of files by separating the names with a `:' character. `\-' is a
202 reserved name, meaning stdin or stdout, depending on the read/write direction
203 set. On Windows, disk devices are accessed as \\.\PhysicalDrive0 for the first
204 device, \\.\PhysicalDrive1 for the second etc. Note: Windows and FreeBSD
205 prevent write access to areas of the disk containing in-use data
206 (e.g. filesystems). If the wanted filename does need to include a colon, then
207 escape that with a '\\' character. For instance, if the filename is
208 "/dev/dsk/foo@3,0:c", then you would use filename="/dev/dsk/foo@3,0\\:c".
210 .BI filename_format \fR=\fPstr
211 If sharing multiple files between jobs, it is usually necessary to have
212 fio generate the exact names that you want. By default, fio will name a file
213 based on the default file format specification of
214 \fBjobname.jobnumber.filenumber\fP. With this option, that can be
215 customized. Fio will recognize and replace the following keywords in this
221 The name of the worker thread or process.
224 The incremental number of the worker thread or process.
227 The incremental number of the file for that worker thread or process.
230 To have dependent jobs share a set of files, this option can be set to
231 have fio generate filenames that are shared between the two. For instance,
232 if \fBtestfiles.$filenum\fR is specified, file number 4 for any job will
233 be named \fBtestfiles.4\fR. The default of \fB$jobname.$jobnum.$filenum\fR
234 will be used if no other format specifier is given.
238 .BI lockfile \fR=\fPstr
239 Fio defaults to not locking any files before it does IO to them. If a file or
240 file descriptor is shared, fio can serialize IO to that file to make the end
241 result consistent. This is usual for emulating real workloads that share files.
247 No locking. This is the default.
250 Only one thread or process may do IO at a time, excluding all others.
253 Read-write locking on the file. Many readers may access the file at the same
254 time, but writes get exclusive access.
258 .BI opendir \fR=\fPstr
259 Recursively open any files below directory \fIstr\fR.
261 .BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
262 Type of I/O pattern. Accepted values are:
273 Sequential trim (Linux block devices only).
282 Random trim (Linux block devices only).
285 Mixed sequential reads and writes.
288 Mixed random reads and writes.
291 Trim and write mixed workload. Blocks will be trimmed first, then the same
292 blocks will be written to.
295 For mixed I/O, the default split is 50/50. For certain types of io the result
296 may still be skewed a bit, since the speed may be different. It is possible to
297 specify a number of IO's to do before getting a new offset, this is done by
298 appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
299 would look like \fBrw=randread:8\fR for passing in an offset modifier with a
300 value of 8. If the postfix is used with a sequential IO pattern, then the value
301 specified will be added to the generated offset for each IO. For instance,
302 using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
303 into sequential IO with holes. See the \fBrw_sequencer\fR option.
306 .BI rw_sequencer \fR=\fPstr
307 If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
308 then this option controls how that number modifies the IO offset being
309 generated. Accepted values are:
314 Generate sequential offset
317 Generate the same offset
320 \fBsequential\fR is only useful for random IO, where fio would normally
321 generate a new random offset for every IO. If you append eg 8 to randread, you
322 would get a new random offset for every 8 IO's. The result would be a seek for
323 only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
324 that. As sequential IO is already sequential, setting \fBsequential\fR for that
325 would not result in any differences. \fBidentical\fR behaves in a similar
326 fashion, except it sends the same offset 8 number of times before generating a
331 .BI kb_base \fR=\fPint
332 The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
333 manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
334 reasons. Allowed values are 1024 or 1000, with 1024 being the default.
336 .BI unified_rw_reporting \fR=\fPbool
337 Fio normally reports statistics on a per data direction basis, meaning that
338 read, write, and trim are accounted and reported separately. If this option is
339 set fio sums the results and reports them as "mixed" instead.
341 .BI randrepeat \fR=\fPbool
342 Seed the random number generator used for random I/O patterns in a predictable
343 way so the pattern is repeatable across runs. Default: true.
345 .BI allrandrepeat \fR=\fPbool
346 Seed all random number generators in a predictable way so results are
347 repeatable across runs. Default: false.
349 .BI randseed \fR=\fPint
350 Seed the random number generators based on this seed value, to be able to
351 control what sequence of output is being generated. If not set, the random
352 sequence depends on the \fBrandrepeat\fR setting.
354 .BI fallocate \fR=\fPstr
355 Whether pre-allocation is performed when laying down files. Accepted values
361 Do not pre-allocate space.
364 Pre-allocate via \fBposix_fallocate\fR\|(3).
367 Pre-allocate via \fBfallocate\fR\|(2) with FALLOC_FL_KEEP_SIZE set.
370 Backward-compatible alias for 'none'.
373 Backward-compatible alias for 'posix'.
376 May not be available on all supported platforms. 'keep' is only
377 available on Linux. If using ZFS on Solaris this must be set to 'none'
378 because ZFS doesn't support it. Default: 'posix'.
381 .BI fadvise_hint \fR=\fPbool
382 Use \fBposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
383 are likely to be issued. Default: true.
385 .BI fadvise_stream \fR=\fPint
386 Use \fBposix_fadvise\fR\|(2) to advise the kernel what stream ID the
387 writes issued belong to. Only supported on Linux. Note, this option
388 may change going forward.
391 Total size of I/O for this job. \fBfio\fR will run until this many bytes have
392 been transferred, unless limited by other options (\fBruntime\fR, for instance,
393 or increased/descreased by \fBio_size\fR). Unless \fBnrfiles\fR and
394 \fBfilesize\fR options are given, this amount will be divided between the
395 available files for the job. If not set, fio will use the full size of the
396 given files or devices. If the files do not exist, size must be given. It is
397 also possible to give size as a percentage between 1 and 100. If size=20% is
398 given, fio will use 20% of the full size of the given files or devices.
400 .BI io_size \fR=\fPint "\fR,\fB io_limit \fR=\fPint
401 Normally fio operates within the region set by \fBsize\fR, which means that
402 the \fBsize\fR option sets both the region and size of IO to be performed.
403 Sometimes that is not what you want. With this option, it is possible to
404 define just the amount of IO that fio should do. For instance, if \fBsize\fR
405 is set to 20G and \fBio_limit\fR is set to 5G, fio will perform IO within
406 the first 20G but exit when 5G have been done. The opposite is also
407 possible - if \fBsize\fR is set to 20G, and \fBio_size\fR is set to 40G, then
408 fio will do 40G of IO within the 0..20G region.
410 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
411 Sets size to something really large and waits for ENOSPC (no space left on
412 device) as the terminating condition. Only makes sense with sequential write.
413 For a read workload, the mount point will be filled first then IO started on
414 the result. This option doesn't make sense if operating on a raw device node,
415 since the size of that is already known by the file system. Additionally,
416 writing beyond end-of-device will not return ENOSPC there.
418 .BI filesize \fR=\fPirange
419 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
420 for files at random within the given range, limited to \fBsize\fR in total (if
421 that is given). If \fBfilesize\fR is not specified, each created file is the
424 .BI file_append \fR=\fPbool
425 Perform IO after the end of the file. Normally fio will operate within the
426 size of a file. If this option is set, then fio will append to the file
427 instead. This has identical behavior to setting \fRoffset\fP to the size
428 of a file. This option is ignored on non-regular files.
430 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
431 Block size for I/O units. Default: 4k. Values for reads, writes, and trims
432 can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR
433 either of which may be empty to leave that value at its default. If a trailing
434 comma isn't given, the remainder will inherit the last value set.
436 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
437 Specify a range of I/O block sizes. The issued I/O unit will always be a
438 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
439 to both reads and writes if only one range is given, but can be specified
440 separately with a comma separating the values. Example: bsrange=1k-4k,2k-8k.
441 Also (see \fBblocksize\fR).
443 .BI bssplit \fR=\fPstr
444 This option allows even finer grained control of the block sizes issued,
445 not just even splits between them. With this option, you can weight various
446 block sizes for exact control of the issued IO for a job that has mixed
447 block sizes. The format of the option is bssplit=blocksize/percentage,
448 optionally adding as many definitions as needed separated by a colon.
449 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
450 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
451 splits to reads and writes. The format is identical to what the
452 \fBbs\fR option accepts, the read and write parts are separated with a
455 .B blocksize_unaligned\fR,\fP bs_unaligned
456 If set, any size in \fBblocksize_range\fR may be used. This typically won't
457 work with direct I/O, as that normally requires sector alignment.
459 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
460 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
461 the minimum blocksize given. Minimum alignment is typically 512b
462 for using direct IO, though it usually depends on the hardware block size.
463 This option is mutually exclusive with using a random map for files, so it
464 will turn off that option.
466 .BI bs_is_seq_rand \fR=\fPbool
467 If this option is set, fio will use the normal read,write blocksize settings as
468 sequential,random instead. Any random read or write will use the WRITE
469 blocksize settings, and any sequential read or write will use the READ
473 Initialize buffers with all zeros. Default: fill buffers with random data.
476 If this option is given, fio will refill the IO buffers on every submit. The
477 default is to only fill it at init time and reuse that data. Only makes sense
478 if zero_buffers isn't specified, naturally. If data verification is enabled,
479 refill_buffers is also automatically enabled.
481 .BI scramble_buffers \fR=\fPbool
482 If \fBrefill_buffers\fR is too costly and the target is using data
483 deduplication, then setting this option will slightly modify the IO buffer
484 contents to defeat normal de-dupe attempts. This is not enough to defeat
485 more clever block compression attempts, but it will stop naive dedupe
486 of blocks. Default: true.
488 .BI buffer_compress_percentage \fR=\fPint
489 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
490 that compress to the specified level. Fio does this by providing a mix of
491 random data and a fixed pattern. The fixed pattern is either zeroes, or the
492 pattern specified by \fBbuffer_pattern\fR. If the pattern option is used, it
493 might skew the compression ratio slightly. Note that this is per block size
494 unit, for file/disk wide compression level that matches this setting. Note
495 that this is per block size unit, for file/disk wide compression level that
496 matches this setting, you'll also want to set refill_buffers.
498 .BI buffer_compress_chunk \fR=\fPint
499 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
500 big the ranges of random data and zeroed data is. Without this set, fio will
501 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
502 the remaining zeroed. With this set to some chunk size smaller than the block
503 size, fio can alternate random and zeroed data throughout the IO buffer.
505 .BI buffer_pattern \fR=\fPstr
506 If set, fio will fill the IO buffers with this pattern. If not set, the contents
507 of IO buffers is defined by the other options related to buffer contents. The
508 setting can be any pattern of bytes, and can be prefixed with 0x for hex
509 values. It may also be a string, where the string must then be wrapped with
513 \fBbuffer_pattern\fR="abcd"
517 \fBbuffer_pattern\fR=-12
521 \fBbuffer_pattern\fR=0xdeadface
524 Also you can combine everything together in any order:
527 \fBbuffer_pattern\fR=0xdeadface"abcd"-12
531 .BI dedupe_percentage \fR=\fPint
532 If set, fio will generate this percentage of identical buffers when writing.
533 These buffers will be naturally dedupable. The contents of the buffers depend
534 on what other buffer compression settings have been set. It's possible to have
535 the individual buffers either fully compressible, or not at all. This option
536 only controls the distribution of unique buffers.
538 .BI nrfiles \fR=\fPint
539 Number of files to use for this job. Default: 1.
541 .BI openfiles \fR=\fPint
542 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
544 .BI file_service_type \fR=\fPstr
545 Defines how files to service are selected. The following types are defined:
550 Choose a file at random.
553 Round robin over opened files (default).
556 Do each file in the set sequentially.
559 The number of I/Os to issue before switching to a new file can be specified by
560 appending `:\fIint\fR' to the service type.
563 .BI ioengine \fR=\fPstr
564 Defines how the job issues I/O. The following types are defined:
569 Basic \fBread\fR\|(2) or \fBwrite\fR\|(2) I/O. \fBfseek\fR\|(2) is used to
570 position the I/O location.
573 Basic \fBpread\fR\|(2) or \fBpwrite\fR\|(2) I/O.
576 Basic \fBreadv\fR\|(2) or \fBwritev\fR\|(2) I/O. Will emulate queuing by
577 coalescing adjacent IOs into a single submission.
580 Basic \fBpreadv\fR\|(2) or \fBpwritev\fR\|(2) I/O.
583 Linux native asynchronous I/O. This ioengine defines engine specific options.
586 POSIX asynchronous I/O using \fBaio_read\fR\|(3) and \fBaio_write\fR\|(3).
589 Solaris native asynchronous I/O.
592 Windows native asynchronous I/O.
595 File is memory mapped with \fBmmap\fR\|(2) and data copied using
599 \fBsplice\fR\|(2) is used to transfer the data and \fBvmsplice\fR\|(2) to
600 transfer data from user-space to the kernel.
603 Use the syslet system calls to make regular read/write asynchronous.
606 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
607 the target is an sg character device, we use \fBread\fR\|(2) and
608 \fBwrite\fR\|(2) for asynchronous I/O.
611 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
612 itself and for debugging and testing purposes.
615 Transfer over the network. The protocol to be used can be defined with the
616 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
617 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
618 This ioengine defines engine specific options.
621 Like \fBnet\fR, but uses \fBsplice\fR\|(2) and \fBvmsplice\fR\|(2) to map data
622 and send/receive. This ioengine defines engine specific options.
625 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
626 \fBcpucycles\fR parameters.
629 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
630 approach to asynchronous I/O.
632 See <http://www.xmailserver.org/guasi\-lib.html>.
635 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
636 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
639 Loads an external I/O engine object file. Append the engine filename as
643 IO engine that does regular linux native fallocate call to simulate data
644 transfer as fio ioengine
646 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
648 DIR_WRITE does fallocate(,mode = 0)
650 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
653 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
654 request to DDIR_WRITE event
657 IO engine supporting direct access to Ceph Rados Block Devices (RBD) via librbd
658 without the need to use the kernel rbd driver. This ioengine defines engine specific
662 Using Glusterfs libgfapi sync interface to direct access to Glusterfs volumes without
663 having to go through FUSE. This ioengine defines engine specific
667 Using Glusterfs libgfapi async interface to direct access to Glusterfs volumes without
668 having to go through FUSE. This ioengine defines engine specific
672 Read and write through Hadoop (HDFS). The \fBfilename\fR option is used to
673 specify host,port of the hdfs name-node to connect. This engine interprets
674 offsets a little differently. In HDFS, files once created cannot be modified.
675 So random writes are not possible. To imitate this, libhdfs engine expects
676 bunch of small files to be created over HDFS, and engine will randomly pick a
677 file out of those files based on the offset generated by fio backend. (see the
678 example job file to create such files, use rw=write option). Please note, you
679 might want to set necessary environment variables to work with hdfs/libhdfs
683 Read, write and erase an MTD character device (e.g., /dev/mtd0). Discards are
684 treated as erases. Depending on the underlying device type, the I/O may have
685 to go in a certain pattern, e.g., on NAND, writing sequentially to erase blocks
686 and discarding before overwriting. The writetrim mode works well for this
692 .BI iodepth \fR=\fPint
693 Number of I/O units to keep in flight against the file. Note that increasing
694 iodepth beyond 1 will not affect synchronous ioengines (except for small
695 degress when verify_async is in use). Even async engines may impose OS
696 restrictions causing the desired depth not to be achieved. This may happen on
697 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
698 not async on that OS. Keep an eye on the IO depth distribution in the
699 fio output to verify that the achieved depth is as expected. Default: 1.
701 .BI iodepth_batch \fR=\fPint "\fR,\fP iodepth_batch_submit" \fR=\fPint
702 This defines how many pieces of IO to submit at once. It defaults to 1
703 which means that we submit each IO as soon as it is available, but can
704 be raised to submit bigger batches of IO at the time. If it is set to 0
705 the \fBiodepth\fR value will be used.
707 .BI iodepth_batch_complete_min \fR=\fPint "\fR,\fP iodepth_batch_complete" \fR=\fPint
708 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
709 means that we'll ask for a minimum of 1 IO in the retrieval process from the
710 kernel. The IO retrieval will go on until we hit the limit set by
711 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
712 completed events before queuing more IO. This helps reduce IO latency, at the
713 cost of more retrieval system calls.
715 .BI iodepth_batch_complete_max \fR=\fPint
716 This defines maximum pieces of IO to
717 retrieve at once. This variable should be used along with
718 \fBiodepth_batch_complete_min\fR=int variable, specifying the range
719 of min and max amount of IO which should be retrieved. By default
720 it is equal to \fBiodepth_batch_complete_min\fR value.
725 \fBiodepth_batch_complete_min\fR=1
727 \fBiodepth_batch_complete_max\fR=<iodepth>
730 which means that we will retrieve at leat 1 IO and up to the
731 whole submitted queue depth. If none of IO has been completed
736 \fBiodepth_batch_complete_min\fR=0
738 \fBiodepth_batch_complete_max\fR=<iodepth>
741 which means that we can retrieve up to the whole submitted
742 queue depth, but if none of IO has been completed yet, we will
743 NOT wait and immediately exit the system call. In this example
744 we simply do polling.
747 .BI iodepth_low \fR=\fPint
748 Low watermark indicating when to start filling the queue again. Default:
751 .BI io_submit_mode \fR=\fPstr
752 This option controls how fio submits the IO to the IO engine. The default is
753 \fBinline\fR, which means that the fio job threads submit and reap IO directly.
754 If set to \fBoffload\fR, the job threads will offload IO submission to a
755 dedicated pool of IO threads. This requires some coordination and thus has a
756 bit of extra overhead, especially for lower queue depth IO where it can
757 increase latencies. The benefit is that fio can manage submission rates
758 independently of the device completion rates. This avoids skewed latency
759 reporting if IO gets back up on the device side (the coordinated omission
762 .BI direct \fR=\fPbool
763 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
765 .BI atomic \fR=\fPbool
766 If value is true, attempt to use atomic direct IO. Atomic writes are guaranteed
767 to be stable once acknowledged by the operating system. Only Linux supports
770 .BI buffered \fR=\fPbool
771 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
774 .BI offset \fR=\fPint
775 Offset in the file to start I/O. Data before the offset will not be touched.
777 .BI offset_increment \fR=\fPint
778 If this is provided, then the real offset becomes the
779 offset + offset_increment * thread_number, where the thread number is a
780 counter that starts at 0 and is incremented for each sub-job (i.e. when
781 numjobs option is specified). This option is useful if there are several jobs
782 which are intended to operate on a file in parallel disjoint segments, with
783 even spacing between the starting points.
785 .BI number_ios \fR=\fPint
786 Fio will normally perform IOs until it has exhausted the size of the region
787 set by \fBsize\fR, or if it exhaust the allocated time (or hits an error
788 condition). With this setting, the range/size can be set independently of
789 the number of IOs to perform. When fio reaches this number, it will exit
790 normally and report status. Note that this does not extend the amount
791 of IO that will be done, it will only stop fio if this condition is met
792 before other end-of-job criteria.
795 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
796 0, don't sync. Default: 0.
798 .BI fdatasync \fR=\fPint
799 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
800 data parts of the file. Default: 0.
802 .BI write_barrier \fR=\fPint
803 Make every Nth write a barrier write.
805 .BI sync_file_range \fR=\fPstr:int
806 Use \fBsync_file_range\fR\|(2) for every \fRval\fP number of write operations. Fio will
807 track range of writes that have happened since the last \fBsync_file_range\fR\|(2) call.
808 \fRstr\fP can currently be one or more of:
812 SYNC_FILE_RANGE_WAIT_BEFORE
815 SYNC_FILE_RANGE_WRITE
818 SYNC_FILE_RANGE_WRITE
822 So if you do sync_file_range=wait_before,write:8, fio would use
823 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
824 Also see the \fBsync_file_range\fR\|(2) man page. This option is Linux specific.
826 .BI overwrite \fR=\fPbool
827 If writing, setup the file first and do overwrites. Default: false.
829 .BI end_fsync \fR=\fPbool
830 Sync file contents when a write stage has completed. Default: false.
832 .BI fsync_on_close \fR=\fPbool
833 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
834 it will happen on every close, not just at the end of the job. Default: false.
836 .BI rwmixread \fR=\fPint
837 Percentage of a mixed workload that should be reads. Default: 50.
839 .BI rwmixwrite \fR=\fPint
840 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
841 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
842 overrides the first. This may interfere with a given rate setting, if fio is
843 asked to limit reads or writes to a certain rate. If that is the case, then
844 the distribution may be skewed. Default: 50.
846 .BI random_distribution \fR=\fPstr:float
847 By default, fio will use a completely uniform random distribution when asked
848 to perform random IO. Sometimes it is useful to skew the distribution in
849 specific ways, ensuring that some parts of the data is more hot than others.
850 Fio includes the following distribution models:
854 Uniform random distribution
864 When using a zipf or pareto distribution, an input value is also needed to
865 define the access pattern. For zipf, this is the zipf theta. For pareto,
866 it's the pareto power. Fio includes a test program, genzipf, that can be
867 used visualize what the given input values will yield in terms of hit rates.
868 If you wanted to use zipf with a theta of 1.2, you would use
869 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
870 fio will disable use of the random map.
872 .BI percentage_random \fR=\fPint
873 For a random workload, set how big a percentage should be random. This defaults
874 to 100%, in which case the workload is fully random. It can be set from
875 anywhere from 0 to 100. Setting it to 0 would make the workload fully
876 sequential. It is possible to set different values for reads, writes, and
877 trim. To do so, simply use a comma separated list. See \fBblocksize\fR.
880 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
881 this parameter is given, a new offset will be chosen without looking at past
882 I/O history. This parameter is mutually exclusive with \fBverify\fR.
884 .BI softrandommap \fR=\fPbool
885 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
886 fails to allocate the map, if this option is set it will continue without a
887 random block map. As coverage will not be as complete as with random maps, this
888 option is disabled by default.
890 .BI random_generator \fR=\fPstr
891 Fio supports the following engines for generating IO offsets for random IO:
895 Strong 2^88 cycle random number generator
898 Linear feedback shift register generator
901 Strong 64-bit 2^258 cycle random number generator
905 Tausworthe is a strong random number generator, but it requires tracking on the
906 side if we want to ensure that blocks are only read or written once. LFSR
907 guarantees that we never generate the same offset twice, and it's also less
908 computationally expensive. It's not a true random generator, however, though
909 for IO purposes it's typically good enough. LFSR only works with single block
910 sizes, not with workloads that use multiple block sizes. If used with such a
911 workload, fio may read or write some blocks multiple times.
914 Run job with given nice value. See \fBnice\fR\|(2).
917 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
920 .BI prioclass \fR=\fPint
921 Set I/O priority class. See \fBionice\fR\|(1).
923 .BI thinktime \fR=\fPint
924 Stall job for given number of microseconds between issuing I/Os.
926 .BI thinktime_spin \fR=\fPint
927 Pretend to spend CPU time for given number of microseconds, sleeping the rest
928 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
930 .BI thinktime_blocks \fR=\fPint
931 Only valid if thinktime is set - control how many blocks to issue, before
932 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
933 make fio wait \fBthinktime\fR microseconds after every block. This
934 effectively makes any queue depth setting redundant, since no more than 1 IO
935 will be queued before we have to complete it and do our thinktime. In other
936 words, this setting effectively caps the queue depth if the latter is larger.
940 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
941 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
942 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
943 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
944 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
945 limit writes (to 500KB/sec), the latter will only limit reads.
947 .BI ratemin \fR=\fPint
948 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
949 Failing to meet this requirement will cause the job to exit. The same format
950 as \fBrate\fR is used for read vs write separation.
952 .BI rate_iops \fR=\fPint
953 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
954 specified independently of bandwidth. The same format as \fBrate\fR is used for
955 read vs write separation. If \fBblocksize\fR is a range, the smallest block
956 size is used as the metric.
958 .BI rate_iops_min \fR=\fPint
959 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
960 is used for read vs write separation.
962 .BI ratecycle \fR=\fPint
963 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
964 milliseconds. Default: 1000ms.
966 .BI latency_target \fR=\fPint
967 If set, fio will attempt to find the max performance point that the given
968 workload will run at while maintaining a latency below this target. The
969 values is given in microseconds. See \fBlatency_window\fR and
970 \fBlatency_percentile\fR.
972 .BI latency_window \fR=\fPint
973 Used with \fBlatency_target\fR to specify the sample window that the job
974 is run at varying queue depths to test the performance. The value is given
977 .BI latency_percentile \fR=\fPfloat
978 The percentage of IOs that must fall within the criteria specified by
979 \fBlatency_target\fR and \fBlatency_window\fR. If not set, this defaults
980 to 100.0, meaning that all IOs must be equal or below to the value set
981 by \fBlatency_target\fR.
983 .BI max_latency \fR=\fPint
984 If set, fio will exit the job if it exceeds this maximum latency. It will exit
987 .BI cpumask \fR=\fPint
988 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
989 may run on. See \fBsched_setaffinity\fR\|(2).
991 .BI cpus_allowed \fR=\fPstr
992 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
994 .BI cpus_allowed_policy \fR=\fPstr
995 Set the policy of how fio distributes the CPUs specified by \fBcpus_allowed\fR
996 or \fBcpumask\fR. Two policies are supported:
1001 All jobs will share the CPU set specified.
1004 Each job will get a unique CPU from the CPU set.
1007 \fBshared\fR is the default behaviour, if the option isn't specified. If
1008 \fBsplit\fR is specified, then fio will assign one cpu per job. If not enough
1009 CPUs are given for the jobs listed, then fio will roundrobin the CPUs in
1014 .BI numa_cpu_nodes \fR=\fPstr
1015 Set this job running on specified NUMA nodes' CPUs. The arguments allow
1016 comma delimited list of cpu numbers, A-B ranges, or 'all'.
1018 .BI numa_mem_policy \fR=\fPstr
1019 Set this job's memory policy and corresponding NUMA nodes. Format of
1023 .B <mode>[:<nodelist>]
1026 is one of the following memory policy:
1028 .B default, prefer, bind, interleave, local
1031 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
1032 needed to be specified. For \fBprefer\fR, only one node is
1033 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
1034 comma delimited list of numbers, A-B ranges, or 'all'.
1036 .BI startdelay \fR=\fPirange
1037 Delay start of job for the specified number of seconds. Supports all time
1038 suffixes to allow specification of hours, minutes, seconds and
1039 milliseconds - seconds are the default if a unit is omitted.
1040 Can be given as a range which causes each thread to choose randomly out of the
1043 .BI runtime \fR=\fPint
1044 Terminate processing after the specified number of seconds.
1047 If given, run for the specified \fBruntime\fR duration even if the files are
1048 completely read or written. The same workload will be repeated as many times
1049 as \fBruntime\fR allows.
1051 .BI ramp_time \fR=\fPint
1052 If set, fio will run the specified workload for this amount of time before
1053 logging any performance numbers. Useful for letting performance settle before
1054 logging results, thus minimizing the runtime required for stable results. Note
1055 that the \fBramp_time\fR is considered lead in time for a job, thus it will
1056 increase the total runtime if a special timeout or runtime is specified.
1058 .BI invalidate \fR=\fPbool
1059 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
1061 .BI sync \fR=\fPbool
1062 Use synchronous I/O for buffered writes. For the majority of I/O engines,
1063 this means using O_SYNC. Default: false.
1065 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
1066 Allocation method for I/O unit buffer. Allowed values are:
1071 Allocate memory with \fBmalloc\fR\|(3).
1074 Use shared memory buffers allocated through \fBshmget\fR\|(2).
1077 Same as \fBshm\fR, but use huge pages as backing.
1080 Use \fBmmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
1081 is given after the option in the format `:\fIfile\fR'.
1084 Same as \fBmmap\fR, but use huge files as backing.
1087 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
1088 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
1089 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
1090 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
1091 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
1092 and the documentation for that. Normally you just need to echo an appropriate
1093 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
1097 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
1098 This indicates the memory alignment of the IO memory buffers. Note that the
1099 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
1100 the alignment of the following buffers are given by the \fBbs\fR used. In
1101 other words, if using a \fBbs\fR that is a multiple of the page sized in the
1102 system, all buffers will be aligned to this value. If using a \fBbs\fR that
1103 is not page aligned, the alignment of subsequent IO memory buffers is the
1104 sum of the \fBiomem_align\fR and \fBbs\fR used.
1106 .BI hugepage\-size \fR=\fPint
1107 Defines the size of a huge page. Must be at least equal to the system setting.
1108 Should be a multiple of 1MB. Default: 4MB.
1111 Terminate all jobs when one finishes. Default: wait for each job to finish.
1113 .BI bwavgtime \fR=\fPint
1114 Average bandwidth calculations over the given time in milliseconds. Default:
1117 .BI iopsavgtime \fR=\fPint
1118 Average IOPS calculations over the given time in milliseconds. Default:
1121 .BI create_serialize \fR=\fPbool
1122 If true, serialize file creation for the jobs. Default: true.
1124 .BI create_fsync \fR=\fPbool
1125 \fBfsync\fR\|(2) data file after creation. Default: true.
1127 .BI create_on_open \fR=\fPbool
1128 If true, the files are not created until they are opened for IO by the job.
1130 .BI create_only \fR=\fPbool
1131 If true, fio will only run the setup phase of the job. If files need to be
1132 laid out or updated on disk, only that will be done. The actual job contents
1135 .BI allow_file_create \fR=\fPbool
1136 If true, fio is permitted to create files as part of its workload. This is
1137 the default behavior. If this option is false, then fio will error out if the
1138 files it needs to use don't already exist. Default: true.
1140 .BI allow_mounted_write \fR=\fPbool
1141 If this isn't set, fio will abort jobs that are destructive (eg that write)
1142 to what appears to be a mounted device or partition. This should help catch
1143 creating inadvertently destructive tests, not realizing that the test will
1144 destroy data on the mounted file system. Default: false.
1146 .BI pre_read \fR=\fPbool
1147 If this is given, files will be pre-read into memory before starting the given
1148 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
1149 pointless to pre-read and then drop the cache. This will only work for IO
1150 engines that are seekable, since they allow you to read the same data
1151 multiple times. Thus it will not work on eg network or splice IO.
1153 .BI unlink \fR=\fPbool
1154 Unlink job files when done. Default: false.
1156 .BI loops \fR=\fPint
1157 Specifies the number of iterations (runs of the same workload) of this job.
1160 .BI verify_only \fR=\fPbool
1161 Do not perform the specified workload, only verify data still matches previous
1162 invocation of this workload. This option allows one to check data multiple
1163 times at a later date without overwriting it. This option makes sense only for
1164 workloads that write data, and does not support workloads with the
1165 \fBtime_based\fR option set.
1167 .BI do_verify \fR=\fPbool
1168 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
1171 .BI verify \fR=\fPstr
1172 Method of verifying file contents after each iteration of the job. Each
1173 verification method also implies verification of special header, which is
1174 written to the beginning of each block. This header also includes meta
1175 information, like offset of the block, block number, timestamp when block
1176 was written, etc. \fBverify\fR=str can be combined with \fBverify_pattern\fR=str
1177 option. The allowed values are:
1181 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 xxhash
1182 Store appropriate checksum in the header of each block. crc32c-intel is
1183 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
1184 not supported by the system.
1187 This option is deprecated, since now meta information is included in generic
1188 verification header and meta verification happens by default. For detailed
1189 information see the description of the \fBverify\fR=str setting. This option
1190 is kept because of compatibility's sake with old configurations. Do not use it.
1193 Verify a strict pattern. Normally fio includes a header with some basic
1194 information and checksumming, but if this option is set, only the
1195 specific pattern set with \fBverify_pattern\fR is verified.
1198 Pretend to verify. Used for testing internals.
1201 This option can be used for repeated burn-in tests of a system to make sure
1202 that the written data is also correctly read back. If the data direction given
1203 is a read or random read, fio will assume that it should verify a previously
1204 written file. If the data direction includes any form of write, the verify will
1205 be of the newly written data.
1208 .BI verifysort \fR=\fPbool
1209 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
1210 read them back in a sorted manner. Default: true.
1212 .BI verifysort_nr \fR=\fPint
1213 Pre-load and sort verify blocks for a read workload.
1215 .BI verify_offset \fR=\fPint
1216 Swap the verification header with data somewhere else in the block before
1217 writing. It is swapped back before verifying.
1219 .BI verify_interval \fR=\fPint
1220 Write the verification header for this number of bytes, which should divide
1221 \fBblocksize\fR. Default: \fBblocksize\fR.
1223 .BI verify_pattern \fR=\fPstr
1224 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
1225 with totally random bytes, but sometimes it's interesting to fill with a known
1226 pattern for io verification purposes. Depending on the width of the pattern,
1227 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
1228 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
1229 has to be a hex number that starts with either "0x" or "0X". Use with
1230 \fBverify\fP=str. Also, verify_pattern supports %o format, which means that for
1231 each block offset will be written and then verifyied back, e.g.:
1234 \fBverify_pattern\fR=%o
1236 Or use combination of everything:
1239 \fBverify_pattern\fR=0xff%o"abcd"-21
1243 .BI verify_fatal \fR=\fPbool
1244 If true, exit the job on the first observed verification failure. Default:
1247 .BI verify_dump \fR=\fPbool
1248 If set, dump the contents of both the original data block and the data block we
1249 read off disk to files. This allows later analysis to inspect just what kind of
1250 data corruption occurred. Off by default.
1252 .BI verify_async \fR=\fPint
1253 Fio will normally verify IO inline from the submitting thread. This option
1254 takes an integer describing how many async offload threads to create for IO
1255 verification instead, causing fio to offload the duty of verifying IO contents
1256 to one or more separate threads. If using this offload option, even sync IO
1257 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
1258 allows them to have IO in flight while verifies are running.
1260 .BI verify_async_cpus \fR=\fPstr
1261 Tell fio to set the given CPU affinity on the async IO verification threads.
1262 See \fBcpus_allowed\fP for the format used.
1264 .BI verify_backlog \fR=\fPint
1265 Fio will normally verify the written contents of a job that utilizes verify
1266 once that job has completed. In other words, everything is written then
1267 everything is read back and verified. You may want to verify continually
1268 instead for a variety of reasons. Fio stores the meta data associated with an
1269 IO block in memory, so for large verify workloads, quite a bit of memory would
1270 be used up holding this meta data. If this option is enabled, fio will write
1271 only N blocks before verifying these blocks.
1273 .BI verify_backlog_batch \fR=\fPint
1274 Control how many blocks fio will verify if verify_backlog is set. If not set,
1275 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
1276 read back and verified). If \fBverify_backlog_batch\fR is less than
1277 \fBverify_backlog\fR then not all blocks will be verified, if
1278 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
1279 will be verified more than once.
1281 .BI trim_percentage \fR=\fPint
1282 Number of verify blocks to discard/trim.
1284 .BI trim_verify_zero \fR=\fPbool
1285 Verify that trim/discarded blocks are returned as zeroes.
1287 .BI trim_backlog \fR=\fPint
1288 Trim after this number of blocks are written.
1290 .BI trim_backlog_batch \fR=\fPint
1291 Trim this number of IO blocks.
1293 .BI experimental_verify \fR=\fPbool
1294 Enable experimental verification.
1296 .BI verify_state_save \fR=\fPbool
1297 When a job exits during the write phase of a verify workload, save its
1298 current state. This allows fio to replay up until that point, if the
1299 verify state is loaded for the verify read phase.
1301 .BI verify_state_load \fR=\fPbool
1302 If a verify termination trigger was used, fio stores the current write
1303 state of each thread. This can be used at verification time so that fio
1304 knows how far it should verify. Without this information, fio will run
1305 a full verification pass, according to the settings in the job file used.
1307 .B stonewall "\fR,\fP wait_for_previous"
1308 Wait for preceding jobs in the job file to exit before starting this one.
1309 \fBstonewall\fR implies \fBnew_group\fR.
1312 Start a new reporting group. If not given, all jobs in a file will be part
1313 of the same reporting group, unless separated by a stonewall.
1315 .BI numjobs \fR=\fPint
1316 Number of clones (processes/threads performing the same workload) of this job.
1320 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
1324 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
1325 with \fBfork\fR\|(2).
1327 .BI zonesize \fR=\fPint
1328 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
1330 .BI zonerange \fR=\fPint
1331 Give size of an IO zone. See \fBzoneskip\fR.
1333 .BI zoneskip \fR=\fPint
1334 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
1337 .BI write_iolog \fR=\fPstr
1338 Write the issued I/O patterns to the specified file. Specify a separate file
1339 for each job, otherwise the iologs will be interspersed and the file may be
1342 .BI read_iolog \fR=\fPstr
1343 Replay the I/O patterns contained in the specified file generated by
1344 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1346 .BI replay_no_stall \fR=\fPint
1347 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1348 attempts to respect timing information between I/Os. Enabling
1349 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1350 still respecting ordering.
1352 .BI replay_redirect \fR=\fPstr
1353 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1354 is to replay the IOPS onto the major/minor device that each IOP was recorded
1355 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1356 single specified device regardless of the device it was recorded from.
1358 .BI replay_align \fR=\fPint
1359 Force alignment of IO offsets and lengths in a trace to this power of 2 value.
1361 .BI replay_scale \fR=\fPint
1362 Scale sector offsets down by this factor when replaying traces.
1364 .BI per_job_logs \fR=\fPbool
1365 If set, this generates bw/clat/iops log with per file private filenames. If
1366 not set, jobs with identical names will share the log filename. Default: true.
1368 .BI write_bw_log \fR=\fPstr
1369 If given, write a bandwidth log of the jobs in this job file. Can be used to
1370 store data of the bandwidth of the jobs in their lifetime. The included
1371 fio_generate_plots script uses gnuplot to turn these text files into nice
1372 graphs. See \fBwrite_lat_log\fR for behaviour of given filename. For this
1373 option, the postfix is _bw.x.log, where x is the index of the job (1..N,
1374 where N is the number of jobs). If \fBper_job_logs\fR is false, then the
1375 filename will not include the job index.
1377 .BI write_lat_log \fR=\fPstr
1378 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1379 filename is given with this option, the default filename of
1380 "jobname_type.x.log" is used, where x is the index of the job (1..N, where
1381 N is the number of jobs). Even if the filename is given, fio will still
1382 append the type of log. If \fBper_job_logs\fR is false, then the filename will
1383 not include the job index.
1385 .BI write_iops_log \fR=\fPstr
1386 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1387 option, the default filename of "jobname_type.x.log" is used, where x is the
1388 index of the job (1..N, where N is the number of jobs). Even if the filename
1389 is given, fio will still append the type of log. If \fBper_job_logs\fR is false,
1390 then the filename will not include the job index.
1392 .BI log_avg_msec \fR=\fPint
1393 By default, fio will log an entry in the iops, latency, or bw log for every
1394 IO that completes. When writing to the disk log, that can quickly grow to a
1395 very large size. Setting this option makes fio average the each log entry
1396 over the specified period of time, reducing the resolution of the log.
1399 .BI log_offset \fR=\fPbool
1400 If this is set, the iolog options will include the byte offset for the IO
1401 entry as well as the other data values.
1403 .BI log_compression \fR=\fPint
1404 If this is set, fio will compress the IO logs as it goes, to keep the memory
1405 footprint lower. When a log reaches the specified size, that chunk is removed
1406 and compressed in the background. Given that IO logs are fairly highly
1407 compressible, this yields a nice memory savings for longer runs. The downside
1408 is that the compression will consume some background CPU cycles, so it may
1409 impact the run. This, however, is also true if the logging ends up consuming
1410 most of the system memory. So pick your poison. The IO logs are saved
1411 normally at the end of a run, by decompressing the chunks and storing them
1412 in the specified log file. This feature depends on the availability of zlib.
1414 .BI log_store_compressed \fR=\fPbool
1415 If set, and \fBlog\fR_compression is also set, fio will store the log files in
1416 a compressed format. They can be decompressed with fio, using the
1417 \fB\-\-inflate-log\fR command line parameter. The files will be stored with a
1420 .BI block_error_percentiles \fR=\fPbool
1421 If set, record errors in trim block-sized units from writes and trims and output
1422 a histogram of how many trims it took to get to errors, and what kind of error
1425 .BI disable_lat \fR=\fPbool
1426 Disable measurements of total latency numbers. Useful only for cutting
1427 back the number of calls to \fBgettimeofday\fR\|(2), as that does impact performance at
1428 really high IOPS rates. Note that to really get rid of a large amount of these
1429 calls, this option must be used with disable_slat and disable_bw as well.
1431 .BI disable_clat \fR=\fPbool
1432 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1434 .BI disable_slat \fR=\fPbool
1435 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1437 .BI disable_bw_measurement \fR=\fPbool
1438 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1440 .BI lockmem \fR=\fPint
1441 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1442 simulate a smaller amount of memory. The amount specified is per worker.
1444 .BI exec_prerun \fR=\fPstr
1445 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1447 Output is redirected in a file called \fBjobname.prerun.txt\fR
1450 .BI exec_postrun \fR=\fPstr
1451 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1453 Output is redirected in a file called \fBjobname.postrun.txt\fR
1456 .BI ioscheduler \fR=\fPstr
1457 Attempt to switch the device hosting the file to the specified I/O scheduler.
1459 .BI disk_util \fR=\fPbool
1460 Generate disk utilization statistics if the platform supports it. Default: true.
1462 .BI clocksource \fR=\fPstr
1463 Use the given clocksource as the base of timing. The supported options are:
1467 \fBgettimeofday\fR\|(2)
1470 \fBclock_gettime\fR\|(2)
1473 Internal CPU clock source
1477 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1478 (and fio is heavy on time calls). Fio will automatically use this clocksource
1479 if it's supported and considered reliable on the system it is running on,
1480 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1481 means supporting TSC Invariant.
1483 .BI gtod_reduce \fR=\fPbool
1484 Enable all of the \fBgettimeofday\fR\|(2) reducing options (disable_clat, disable_slat,
1485 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1486 \fBgettimeofday\fR\|(2) call count. With this option enabled, we only do about 0.4% of
1487 the gtod() calls we would have done if all time keeping was enabled.
1489 .BI gtod_cpu \fR=\fPint
1490 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1491 the current time. Fio (and databases, for instance) are very intensive on
1492 \fBgettimeofday\fR\|(2) calls. With this option, you can set one CPU aside for doing
1493 nothing but logging current time to a shared memory location. Then the other
1494 threads/processes that run IO workloads need only copy that segment, instead of
1495 entering the kernel with a \fBgettimeofday\fR\|(2) call. The CPU set aside for doing
1496 these time calls will be excluded from other uses. Fio will manually clear it
1497 from the CPU mask of other jobs.
1499 .BI ignore_error \fR=\fPstr
1500 Sometimes you want to ignore some errors during test in that case you can specify
1501 error list for each error type.
1503 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1505 errors for given error type is separated with ':'.
1506 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1508 Example: ignore_error=EAGAIN,ENOSPC:122 .
1510 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1512 .BI error_dump \fR=\fPbool
1513 If set dump every error even if it is non fatal, true by default. If disabled
1514 only fatal error will be dumped
1516 .BI profile \fR=\fPstr
1517 Select a specific builtin performance test.
1519 .BI cgroup \fR=\fPstr
1520 Add job to this control group. If it doesn't exist, it will be created.
1521 The system must have a mounted cgroup blkio mount point for this to work. If
1522 your system doesn't have it mounted, you can do so with:
1524 # mount \-t cgroup \-o blkio none /cgroup
1526 .BI cgroup_weight \fR=\fPint
1527 Set the weight of the cgroup to this value. See the documentation that comes
1528 with the kernel, allowed values are in the range of 100..1000.
1530 .BI cgroup_nodelete \fR=\fPbool
1531 Normally fio will delete the cgroups it has created after the job completion.
1532 To override this behavior and to leave cgroups around after the job completion,
1533 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1534 cgroup files after job completion. Default: false
1537 Instead of running as the invoking user, set the user ID to this value before
1538 the thread/process does any work.
1541 Set group ID, see \fBuid\fR.
1543 .BI unit_base \fR=\fPint
1544 Base unit for reporting. Allowed values are:
1548 Use auto-detection (default).
1558 .BI flow_id \fR=\fPint
1559 The ID of the flow. If not specified, it defaults to being a global flow. See
1563 Weight in token-based flow control. If this value is used, then there is a
1564 \fBflow counter\fR which is used to regulate the proportion of activity between
1565 two or more jobs. fio attempts to keep this flow counter near zero. The
1566 \fBflow\fR parameter stands for how much should be added or subtracted to the
1567 flow counter on each iteration of the main I/O loop. That is, if one job has
1568 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1569 1:8 ratio in how much one runs vs the other.
1571 .BI flow_watermark \fR=\fPint
1572 The maximum value that the absolute value of the flow counter is allowed to
1573 reach before the job must wait for a lower value of the counter.
1575 .BI flow_sleep \fR=\fPint
1576 The period of time, in microseconds, to wait after the flow watermark has been
1577 exceeded before retrying operations
1579 .BI clat_percentiles \fR=\fPbool
1580 Enable the reporting of percentiles of completion latencies.
1582 .BI percentile_list \fR=\fPfloat_list
1583 Overwrite the default list of percentiles for completion latencies and the
1584 block error histogram. Each number is a floating number in the range (0,100],
1585 and the maximum length of the list is 20. Use ':' to separate the
1586 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1587 report the values of completion latency below which 99.5% and 99.9% of
1588 the observed latencies fell, respectively.
1589 .SS "Ioengine Parameters List"
1590 Some parameters are only valid when a specific ioengine is in use. These are
1591 used identically to normal parameters, with the caveat that when used on the
1592 command line, they must come after the ioengine.
1594 .BI (cpu)cpuload \fR=\fPint
1595 Attempt to use the specified percentage of CPU cycles.
1597 .BI (cpu)cpuchunks \fR=\fPint
1598 Split the load into cycles of the given time. In microseconds.
1600 .BI (cpu)exit_on_io_done \fR=\fPbool
1601 Detect when IO threads are done, then exit.
1603 .BI (libaio)userspace_reap
1604 Normally, with the libaio engine in use, fio will use
1605 the io_getevents system call to reap newly returned events.
1606 With this flag turned on, the AIO ring will be read directly
1607 from user-space to reap events. The reaping mode is only
1608 enabled when polling for a minimum of 0 events (eg when
1609 iodepth_batch_complete=0).
1611 .BI (net,netsplice)hostname \fR=\fPstr
1612 The host name or IP address to use for TCP or UDP based IO.
1613 If the job is a TCP listener or UDP reader, the hostname is not
1614 used and must be omitted unless it is a valid UDP multicast address.
1616 .BI (net,netsplice)port \fR=\fPint
1617 The TCP or UDP port to bind to or connect to. If this is used with
1618 \fBnumjobs\fR to spawn multiple instances of the same job type, then
1619 this will be the starting port number since fio will use a range of ports.
1621 .BI (net,netsplice)interface \fR=\fPstr
1622 The IP address of the network interface used to send or receive UDP multicast
1625 .BI (net,netsplice)ttl \fR=\fPint
1626 Time-to-live value for outgoing UDP multicast packets. Default: 1
1628 .BI (net,netsplice)nodelay \fR=\fPbool
1629 Set TCP_NODELAY on TCP connections.
1631 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1632 The network protocol to use. Accepted values are:
1637 Transmission control protocol
1640 Transmission control protocol V6
1643 User datagram protocol
1646 User datagram protocol V6
1652 When the protocol is TCP or UDP, the port must also be given,
1653 as well as the hostname if the job is a TCP listener or UDP
1654 reader. For unix sockets, the normal filename option should be
1655 used and the port is invalid.
1658 .BI (net,netsplice)listen
1659 For TCP network connections, tell fio to listen for incoming
1660 connections rather than initiating an outgoing connection. The
1661 hostname must be omitted if this option is used.
1663 .BI (net, pingpong) \fR=\fPbool
1664 Normally a network writer will just continue writing data, and a network reader
1665 will just consume packets. If pingpong=1 is set, a writer will send its normal
1666 payload to the reader, then wait for the reader to send the same payload back.
1667 This allows fio to measure network latencies. The submission and completion
1668 latencies then measure local time spent sending or receiving, and the
1669 completion latency measures how long it took for the other end to receive and
1670 send back. For UDP multicast traffic pingpong=1 should only be set for a single
1671 reader when multiple readers are listening to the same address.
1673 .BI (net, window_size) \fR=\fPint
1674 Set the desired socket buffer size for the connection.
1676 .BI (net, mss) \fR=\fPint
1677 Set the TCP maximum segment size (TCP_MAXSEG).
1679 .BI (e4defrag,donorname) \fR=\fPstr
1680 File will be used as a block donor (swap extents between files)
1682 .BI (e4defrag,inplace) \fR=\fPint
1683 Configure donor file block allocation strategy
1686 Preallocate donor's file on init
1689 allocate space immediately inside defragment event, and free right after event
1692 .BI (rbd)rbdname \fR=\fPstr
1693 Specifies the name of the RBD.
1695 .BI (rbd)pool \fR=\fPstr
1696 Specifies the name of the Ceph pool containing the RBD.
1698 .BI (rbd)clientname \fR=\fPstr
1699 Specifies the username (without the 'client.' prefix) used to access the Ceph cluster.
1701 .BI (mtd)skipbad \fR=\fPbool
1702 Skip operations against known bad blocks.
1704 While running, \fBfio\fR will display the status of the created jobs. For
1708 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1711 The characters in the first set of brackets denote the current status of each
1712 threads. The possible values are:
1718 Setup but not started.
1724 Initialized, waiting.
1727 Running, doing sequential reads.
1730 Running, doing random reads.
1733 Running, doing sequential writes.
1736 Running, doing random writes.
1739 Running, doing mixed sequential reads/writes.
1742 Running, doing mixed random reads/writes.
1745 Running, currently waiting for \fBfsync\fR\|(2).
1748 Running, verifying written data.
1751 Exited, not reaped by main thread.
1754 Exited, thread reaped.
1758 The second set of brackets shows the estimated completion percentage of
1759 the current group. The third set shows the read and write I/O rate,
1760 respectively. Finally, the estimated run time of the job is displayed.
1762 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1763 for each thread, each group of threads, and each disk, in that order.
1765 Per-thread statistics first show the threads client number, group-id, and
1766 error code. The remaining figures are as follows:
1770 Number of megabytes of I/O performed.
1773 Average data rate (bandwidth).
1779 Submission latency minimum, maximum, average and standard deviation. This is
1780 the time it took to submit the I/O.
1783 Completion latency minimum, maximum, average and standard deviation. This
1784 is the time between submission and completion.
1787 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1788 and standard deviation.
1791 CPU usage statistics. Includes user and system time, number of context switches
1792 this thread went through and number of major and minor page faults.
1795 Distribution of I/O depths. Each depth includes everything less than (or equal)
1796 to it, but greater than the previous depth.
1799 Number of read/write requests issued, and number of short read/write requests.
1802 Distribution of I/O completion latencies. The numbers follow the same pattern
1806 The group statistics show:
1811 Number of megabytes I/O performed.
1814 Aggregate bandwidth of threads in the group.
1817 Minimum average bandwidth a thread saw.
1820 Maximum average bandwidth a thread saw.
1823 Shortest runtime of threads in the group.
1826 Longest runtime of threads in the group.
1830 Finally, disk statistics are printed with reads first:
1835 Number of I/Os performed by all groups.
1838 Number of merges in the I/O scheduler.
1841 Number of ticks we kept the disk busy.
1844 Total time spent in the disk queue.
1851 It is also possible to get fio to dump the current output while it is
1852 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1855 If the \fB\-\-minimal\fR / \fB\-\-append-terse\fR options are given, the
1856 results will be printed/appended in a semicolon-delimited format suitable for
1858 A job description (if provided) follows on a new line. Note that the first
1859 number in the line is the version number. If the output has to be changed
1860 for some reason, this number will be incremented by 1 to signify that
1861 change. The fields are:
1864 .B terse version, fio version, jobname, groupid, error
1868 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1872 .B min, max, mean, standard deviation
1876 .B min, max, mean, standard deviation
1878 Completion latency percentiles (20 fields):
1880 .B Xth percentile=usec
1884 .B min, max, mean, standard deviation
1888 .B min, max, aggregate percentage of total, mean, standard deviation
1894 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1898 .B min, max, mean, standard deviation
1902 .B min, max, mean, standard deviation
1904 Completion latency percentiles (20 fields):
1906 .B Xth percentile=usec
1910 .B min, max, mean, standard deviation
1914 .B min, max, aggregate percentage of total, mean, standard deviation
1920 .B user, system, context switches, major page faults, minor page faults
1923 IO depth distribution:
1925 .B <=1, 2, 4, 8, 16, 32, >=64
1928 IO latency distribution:
1932 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1936 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1940 Disk utilization (1 for each disk used):
1942 .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
1945 Error Info (dependent on continue_on_error, default off):
1947 .B total # errors, first error code
1950 .B text description (if provided in config - appears on newline)
1953 Normally you would run fio as a stand-alone application on the machine
1954 where the IO workload should be generated. However, it is also possible to
1955 run the frontend and backend of fio separately. This makes it possible to
1956 have a fio server running on the machine(s) where the IO workload should
1957 be running, while controlling it from another machine.
1959 To start the server, you would do:
1961 \fBfio \-\-server=args\fR
1963 on that machine, where args defines what fio listens to. The arguments
1964 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1965 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1966 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1967 listen to (only valid for TCP/IP, not a local socket). Some examples:
1971 Start a fio server, listening on all interfaces on the default port (8765).
1973 2) fio \-\-server=ip:hostname,4444
1975 Start a fio server, listening on IP belonging to hostname and on port 4444.
1977 3) fio \-\-server=ip6:::1,4444
1979 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1981 4) fio \-\-server=,4444
1983 Start a fio server, listening on all interfaces on port 4444.
1985 5) fio \-\-server=1.2.3.4
1987 Start a fio server, listening on IP 1.2.3.4 on the default port.
1989 6) fio \-\-server=sock:/tmp/fio.sock
1991 Start a fio server, listening on the local socket /tmp/fio.sock.
1993 When a server is running, you can connect to it from a client. The client
1996 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1998 where \-\-local-args are arguments that are local to the client where it is
1999 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
2000 are sent to the server. The 'server' string follows the same format as it
2001 does on the server side, to allow IP/hostname/socket and port strings.
2002 You can connect to multiple clients as well, to do that you could run:
2004 fio \-\-client=server2 \-\-client=server2 <job file(s)>
2006 If the job file is located on the fio server, then you can tell the server
2007 to load a local file as well. This is done by using \-\-remote-config:
2009 fio \-\-client=server \-\-remote-config /path/to/file.fio
2011 Then fio will open this local (to the server) job file instead
2012 of being passed one from the client.
2014 If you have many servers (example: 100 VMs/containers), you can input a pathname
2015 of a file containing host IPs/names as the parameter value for the \-\-client option.
2016 For example, here is an example "host.list" file containing 2 hostnames:
2018 host1.your.dns.domain
2020 host2.your.dns.domain
2022 The fio command would then be:
2024 fio \-\-client=host.list <job file>
2026 In this mode, you cannot input server-specific parameters or job files, and all
2027 servers receive the same job file.
2029 In order to enable fio \-\-client runs utilizing a shared filesystem from multiple hosts,
2030 fio \-\-client now prepends the IP address of the server to the filename. For example,
2031 if fio is using directory /mnt/nfs/fio and is writing filename fileio.tmp,
2032 with a \-\-client hostfile
2033 containing two hostnames h1 and h2 with IP addresses 192.168.10.120 and 192.168.10.121, then
2034 fio will create two files:
2036 /mnt/nfs/fio/192.168.10.120.fileio.tmp
2038 /mnt/nfs/fio/192.168.10.121.fileio.tmp
2043 was written by Jens Axboe <jens.axboe@oracle.com>,
2044 now Jens Axboe <axboe@fb.com>.
2046 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
2047 on documentation by Jens Axboe.
2048 .SH "REPORTING BUGS"
2049 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
2052 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
2054 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob