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.
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 For mixed I/O, the default split is 50/50. For certain types of io the result
292 may still be skewed a bit, since the speed may be different. It is possible to
293 specify a number of IO's to do before getting a new offset, this is done by
294 appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
295 would look like \fBrw=randread:8\fR for passing in an offset modifier with a
296 value of 8. If the postfix is used with a sequential IO pattern, then the value
297 specified will be added to the generated offset for each IO. For instance,
298 using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
299 into sequential IO with holes. See the \fBrw_sequencer\fR option.
302 .BI rw_sequencer \fR=\fPstr
303 If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
304 then this option controls how that number modifies the IO offset being
305 generated. Accepted values are:
310 Generate sequential offset
313 Generate the same offset
316 \fBsequential\fR is only useful for random IO, where fio would normally
317 generate a new random offset for every IO. If you append eg 8 to randread, you
318 would get a new random offset for every 8 IO's. The result would be a seek for
319 only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
320 that. As sequential IO is already sequential, setting \fBsequential\fR for that
321 would not result in any differences. \fBidentical\fR behaves in a similar
322 fashion, except it sends the same offset 8 number of times before generating a
327 .BI kb_base \fR=\fPint
328 The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
329 manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
330 reasons. Allowed values are 1024 or 1000, with 1024 being the default.
332 .BI unified_rw_reporting \fR=\fPbool
333 Fio normally reports statistics on a per data direction basis, meaning that
334 read, write, and trim are accounted and reported separately. If this option is
335 set fio sums the results and reports them as "mixed" instead.
337 .BI randrepeat \fR=\fPbool
338 Seed the random number generator used for random I/O patterns in a predictable
339 way so the pattern is repeatable across runs. Default: true.
341 .BI allrandrepeat \fR=\fPbool
342 Seed all random number generators in a predictable way so results are
343 repeatable across runs. Default: false.
345 .BI randseed \fR=\fPint
346 Seed the random number generators based on this seed value, to be able to
347 control what sequence of output is being generated. If not set, the random
348 sequence depends on the \fBrandrepeat\fR setting.
350 .BI fallocate \fR=\fPstr
351 Whether pre-allocation is performed when laying down files. Accepted values
357 Do not pre-allocate space.
360 Pre-allocate via \fBposix_fallocate\fR\|(3).
363 Pre-allocate via \fBfallocate\fR\|(2) with FALLOC_FL_KEEP_SIZE set.
366 Backward-compatible alias for 'none'.
369 Backward-compatible alias for 'posix'.
372 May not be available on all supported platforms. 'keep' is only
373 available on Linux. If using ZFS on Solaris this must be set to 'none'
374 because ZFS doesn't support it. Default: 'posix'.
377 .BI fadvise_hint \fR=\fPbool
378 Use \fBposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
379 are likely to be issued. Default: true.
382 Total size of I/O for this job. \fBfio\fR will run until this many bytes have
383 been transferred, unless limited by other options (\fBruntime\fR, for instance,
384 or increased/descreased by \fBio_size\fR). Unless \fBnrfiles\fR and
385 \fBfilesize\fR options are given, this amount will be divided between the
386 available files for the job. If not set, fio will use the full size of the
387 given files or devices. If the files do not exist, size must be given. It is
388 also possible to give size as a percentage between 1 and 100. If size=20% is
389 given, fio will use 20% of the full size of the given files or devices.
391 .BI io_size \fR=\fPint "\fR,\fB io_limit \fR=\fPint
392 Normally fio operates within the region set by \fBsize\fR, which means that
393 the \fBsize\fR option sets both the region and size of IO to be performed.
394 Sometimes that is not what you want. With this option, it is possible to
395 define just the amount of IO that fio should do. For instance, if \fBsize\fR
396 is set to 20G and \fBio_limit\fR is set to 5G, fio will perform IO within
397 the first 20G but exit when 5G have been done. The opposite is also
398 possible - if \fBsize\fR is set to 20G, and \fBio_size\fR is set to 40G, then
399 fio will do 40G of IO within the 0..20G region.
401 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
402 Sets size to something really large and waits for ENOSPC (no space left on
403 device) as the terminating condition. Only makes sense with sequential write.
404 For a read workload, the mount point will be filled first then IO started on
405 the result. This option doesn't make sense if operating on a raw device node,
406 since the size of that is already known by the file system. Additionally,
407 writing beyond end-of-device will not return ENOSPC there.
409 .BI filesize \fR=\fPirange
410 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
411 for files at random within the given range, limited to \fBsize\fR in total (if
412 that is given). If \fBfilesize\fR is not specified, each created file is the
415 .BI file_append \fR=\fPbool
416 Perform IO after the end of the file. Normally fio will operate within the
417 size of a file. If this option is set, then fio will append to the file
418 instead. This has identical behavior to setting \fRoffset\fP to the size
419 of a file. This option is ignored on non-regular files.
421 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
422 Block size for I/O units. Default: 4k. Values for reads, writes, and trims
423 can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR
424 either of which may be empty to leave that value at its default. If a trailing
425 comma isn't given, the remainder will inherit the last value set.
427 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
428 Specify a range of I/O block sizes. The issued I/O unit will always be a
429 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
430 to both reads and writes if only one range is given, but can be specified
431 separately with a comma separating the values. Example: bsrange=1k-4k,2k-8k.
432 Also (see \fBblocksize\fR).
434 .BI bssplit \fR=\fPstr
435 This option allows even finer grained control of the block sizes issued,
436 not just even splits between them. With this option, you can weight various
437 block sizes for exact control of the issued IO for a job that has mixed
438 block sizes. The format of the option is bssplit=blocksize/percentage,
439 optionally adding as many definitions as needed separated by a colon.
440 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
441 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
442 splits to reads and writes. The format is identical to what the
443 \fBbs\fR option accepts, the read and write parts are separated with a
446 .B blocksize_unaligned\fR,\fP bs_unaligned
447 If set, any size in \fBblocksize_range\fR may be used. This typically won't
448 work with direct I/O, as that normally requires sector alignment.
450 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
451 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
452 the minimum blocksize given. Minimum alignment is typically 512b
453 for using direct IO, though it usually depends on the hardware block size.
454 This option is mutually exclusive with using a random map for files, so it
455 will turn off that option.
457 .BI bs_is_seq_rand \fR=\fPbool
458 If this option is set, fio will use the normal read,write blocksize settings as
459 sequential,random instead. Any random read or write will use the WRITE
460 blocksize settings, and any sequential read or write will use the READ
464 Initialize buffers with all zeros. Default: fill buffers with random data.
467 If this option is given, fio will refill the IO buffers on every submit. The
468 default is to only fill it at init time and reuse that data. Only makes sense
469 if zero_buffers isn't specified, naturally. If data verification is enabled,
470 refill_buffers is also automatically enabled.
472 .BI scramble_buffers \fR=\fPbool
473 If \fBrefill_buffers\fR is too costly and the target is using data
474 deduplication, then setting this option will slightly modify the IO buffer
475 contents to defeat normal de-dupe attempts. This is not enough to defeat
476 more clever block compression attempts, but it will stop naive dedupe
477 of blocks. Default: true.
479 .BI buffer_compress_percentage \fR=\fPint
480 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
481 that compress to the specified level. Fio does this by providing a mix of
482 random data and a fixed pattern. The fixed pattern is either zeroes, or the
483 pattern specified by \fBbuffer_pattern\fR. If the pattern option is used, it
484 might skew the compression ratio slightly. Note that this is per block size
485 unit, for file/disk wide compression level that matches this setting. Note
486 that this is per block size unit, for file/disk wide compression level that
487 matches this setting, you'll also want to set refill_buffers.
489 .BI buffer_compress_chunk \fR=\fPint
490 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
491 big the ranges of random data and zeroed data is. Without this set, fio will
492 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
493 the remaining zeroed. With this set to some chunk size smaller than the block
494 size, fio can alternate random and zeroed data throughout the IO buffer.
496 .BI buffer_pattern \fR=\fPstr
497 If set, fio will fill the IO buffers with this pattern. If not set, the contents
498 of IO buffers is defined by the other options related to buffer contents. The
499 setting can be any pattern of bytes, and can be prefixed with 0x for hex
500 values. It may also be a string, where the string must then be wrapped with
503 .BI dedupe_percentage \fR=\fPint
504 If set, fio will generate this percentage of identical buffers when writing.
505 These buffers will be naturally dedupable. The contents of the buffers depend
506 on what other buffer compression settings have been set. It's possible to have
507 the individual buffers either fully compressible, or not at all. This option
508 only controls the distribution of unique buffers.
510 .BI nrfiles \fR=\fPint
511 Number of files to use for this job. Default: 1.
513 .BI openfiles \fR=\fPint
514 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
516 .BI file_service_type \fR=\fPstr
517 Defines how files to service are selected. The following types are defined:
522 Choose a file at random.
525 Round robin over opened files (default).
528 Do each file in the set sequentially.
531 The number of I/Os to issue before switching to a new file can be specified by
532 appending `:\fIint\fR' to the service type.
535 .BI ioengine \fR=\fPstr
536 Defines how the job issues I/O. The following types are defined:
541 Basic \fBread\fR\|(2) or \fBwrite\fR\|(2) I/O. \fBfseek\fR\|(2) is used to
542 position the I/O location.
545 Basic \fBpread\fR\|(2) or \fBpwrite\fR\|(2) I/O.
548 Basic \fBreadv\fR\|(2) or \fBwritev\fR\|(2) I/O. Will emulate queuing by
549 coalescing adjacent IOs into a single submission.
552 Basic \fBpreadv\fR\|(2) or \fBpwritev\fR\|(2) I/O.
555 Linux native asynchronous I/O. This ioengine defines engine specific options.
558 POSIX asynchronous I/O using \fBaio_read\fR\|(3) and \fBaio_write\fR\|(3).
561 Solaris native asynchronous I/O.
564 Windows native asynchronous I/O.
567 File is memory mapped with \fBmmap\fR\|(2) and data copied using
571 \fBsplice\fR\|(2) is used to transfer the data and \fBvmsplice\fR\|(2) to
572 transfer data from user-space to the kernel.
575 Use the syslet system calls to make regular read/write asynchronous.
578 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
579 the target is an sg character device, we use \fBread\fR\|(2) and
580 \fBwrite\fR\|(2) for asynchronous I/O.
583 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
584 itself and for debugging and testing purposes.
587 Transfer over the network. The protocol to be used can be defined with the
588 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
589 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
590 This ioengine defines engine specific options.
593 Like \fBnet\fR, but uses \fBsplice\fR\|(2) and \fBvmsplice\fR\|(2) to map data
594 and send/receive. This ioengine defines engine specific options.
597 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
598 \fBcpucycles\fR parameters.
601 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
602 approach to asynchronous I/O.
604 See <http://www.xmailserver.org/guasi\-lib.html>.
607 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
608 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
611 Loads an external I/O engine object file. Append the engine filename as
615 IO engine that does regular linux native fallocate call to simulate data
616 transfer as fio ioengine
618 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
620 DIR_WRITE does fallocate(,mode = 0)
622 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
625 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
626 request to DDIR_WRITE event
629 IO engine supporting direct access to Ceph Rados Block Devices (RBD) via librbd
630 without the need to use the kernel rbd driver. This ioengine defines engine specific
634 Using Glusterfs libgfapi sync interface to direct access to Glusterfs volumes without
635 having to go through FUSE. This ioengine defines engine specific
639 Using Glusterfs libgfapi async interface to direct access to Glusterfs volumes without
640 having to go through FUSE. This ioengine defines engine specific
644 Read and write through Hadoop (HDFS). The \fBfilename\fR option is used to
645 specify host,port of the hdfs name-node to connect. This engine interprets
646 offsets a little differently. In HDFS, files once created cannot be modified.
647 So random writes are not possible. To imitate this, libhdfs engine expects
648 bunch of small files to be created over HDFS, and engine will randomly pick a
649 file out of those files based on the offset generated by fio backend. (see the
650 example job file to create such files, use rw=write option). Please note, you
651 might want to set necessary environment variables to work with hdfs/libhdfs
657 .BI iodepth \fR=\fPint
658 Number of I/O units to keep in flight against the file. Note that increasing
659 iodepth beyond 1 will not affect synchronous ioengines (except for small
660 degress when verify_async is in use). Even async engines may impose OS
661 restrictions causing the desired depth not to be achieved. This may happen on
662 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
663 not async on that OS. Keep an eye on the IO depth distribution in the
664 fio output to verify that the achieved depth is as expected. Default: 1.
666 .BI iodepth_batch \fR=\fPint
667 Number of I/Os to submit at once. Default: \fBiodepth\fR.
669 .BI iodepth_batch_complete \fR=\fPint
670 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
671 means that we'll ask for a minimum of 1 IO in the retrieval process from the
672 kernel. The IO retrieval will go on until we hit the limit set by
673 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
674 completed events before queuing more IO. This helps reduce IO latency, at the
675 cost of more retrieval system calls.
677 .BI iodepth_low \fR=\fPint
678 Low watermark indicating when to start filling the queue again. Default:
681 .BI direct \fR=\fPbool
682 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
684 .BI atomic \fR=\fPbool
685 If value is true, attempt to use atomic direct IO. Atomic writes are guaranteed
686 to be stable once acknowledged by the operating system. Only Linux supports
689 .BI buffered \fR=\fPbool
690 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
693 .BI offset \fR=\fPint
694 Offset in the file to start I/O. Data before the offset will not be touched.
696 .BI offset_increment \fR=\fPint
697 If this is provided, then the real offset becomes the
698 offset + offset_increment * thread_number, where the thread number is a
699 counter that starts at 0 and is incremented for each sub-job (i.e. when
700 numjobs option is specified). This option is useful if there are several jobs
701 which are intended to operate on a file in parallel disjoint segments, with
702 even spacing between the starting points.
704 .BI number_ios \fR=\fPint
705 Fio will normally perform IOs until it has exhausted the size of the region
706 set by \fBsize\fR, or if it exhaust the allocated time (or hits an error
707 condition). With this setting, the range/size can be set independently of
708 the number of IOs to perform. When fio reaches this number, it will exit
709 normally and report status. Note that this does not extend the amount
710 of IO that will be done, it will only stop fio if this condition is met
711 before other end-of-job criteria.
714 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
715 0, don't sync. Default: 0.
717 .BI fdatasync \fR=\fPint
718 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
719 data parts of the file. Default: 0.
721 .BI write_barrier \fR=\fPint
722 Make every Nth write a barrier write.
724 .BI sync_file_range \fR=\fPstr:int
725 Use \fBsync_file_range\fR\|(2) for every \fRval\fP number of write operations. Fio will
726 track range of writes that have happened since the last \fBsync_file_range\fR\|(2) call.
727 \fRstr\fP can currently be one or more of:
731 SYNC_FILE_RANGE_WAIT_BEFORE
734 SYNC_FILE_RANGE_WRITE
737 SYNC_FILE_RANGE_WRITE
741 So if you do sync_file_range=wait_before,write:8, fio would use
742 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
743 Also see the \fBsync_file_range\fR\|(2) man page. This option is Linux specific.
745 .BI overwrite \fR=\fPbool
746 If writing, setup the file first and do overwrites. Default: false.
748 .BI end_fsync \fR=\fPbool
749 Sync file contents when a write stage has completed. Default: false.
751 .BI fsync_on_close \fR=\fPbool
752 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
753 it will happen on every close, not just at the end of the job. Default: false.
755 .BI rwmixread \fR=\fPint
756 Percentage of a mixed workload that should be reads. Default: 50.
758 .BI rwmixwrite \fR=\fPint
759 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
760 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
761 overrides the first. This may interfere with a given rate setting, if fio is
762 asked to limit reads or writes to a certain rate. If that is the case, then
763 the distribution may be skewed. Default: 50.
765 .BI random_distribution \fR=\fPstr:float
766 By default, fio will use a completely uniform random distribution when asked
767 to perform random IO. Sometimes it is useful to skew the distribution in
768 specific ways, ensuring that some parts of the data is more hot than others.
769 Fio includes the following distribution models:
773 Uniform random distribution
783 When using a zipf or pareto distribution, an input value is also needed to
784 define the access pattern. For zipf, this is the zipf theta. For pareto,
785 it's the pareto power. Fio includes a test program, genzipf, that can be
786 used visualize what the given input values will yield in terms of hit rates.
787 If you wanted to use zipf with a theta of 1.2, you would use
788 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
789 fio will disable use of the random map.
791 .BI percentage_random \fR=\fPint
792 For a random workload, set how big a percentage should be random. This defaults
793 to 100%, in which case the workload is fully random. It can be set from
794 anywhere from 0 to 100. Setting it to 0 would make the workload fully
795 sequential. It is possible to set different values for reads, writes, and
796 trim. To do so, simply use a comma separated list. See \fBblocksize\fR.
799 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
800 this parameter is given, a new offset will be chosen without looking at past
801 I/O history. This parameter is mutually exclusive with \fBverify\fR.
803 .BI softrandommap \fR=\fPbool
804 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
805 fails to allocate the map, if this option is set it will continue without a
806 random block map. As coverage will not be as complete as with random maps, this
807 option is disabled by default.
809 .BI random_generator \fR=\fPstr
810 Fio supports the following engines for generating IO offsets for random IO:
814 Strong 2^88 cycle random number generator
817 Linear feedback shift register generator
821 Tausworthe is a strong random number generator, but it requires tracking on the
822 side if we want to ensure that blocks are only read or written once. LFSR
823 guarantees that we never generate the same offset twice, and it's also less
824 computationally expensive. It's not a true random generator, however, though
825 for IO purposes it's typically good enough. LFSR only works with single block
826 sizes, not with workloads that use multiple block sizes. If used with such a
827 workload, fio may read or write some blocks multiple times.
830 Run job with given nice value. See \fBnice\fR\|(2).
833 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
836 .BI prioclass \fR=\fPint
837 Set I/O priority class. See \fBionice\fR\|(1).
839 .BI thinktime \fR=\fPint
840 Stall job for given number of microseconds between issuing I/Os.
842 .BI thinktime_spin \fR=\fPint
843 Pretend to spend CPU time for given number of microseconds, sleeping the rest
844 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
846 .BI thinktime_blocks \fR=\fPint
847 Only valid if thinktime is set - control how many blocks to issue, before
848 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
849 make fio wait \fBthinktime\fR microseconds after every block. This
850 effectively makes any queue depth setting redundant, since no more than 1 IO
851 will be queued before we have to complete it and do our thinktime. In other
852 words, this setting effectively caps the queue depth if the latter is larger.
856 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
857 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
858 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
859 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
860 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
861 limit writes (to 500KB/sec), the latter will only limit reads.
863 .BI ratemin \fR=\fPint
864 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
865 Failing to meet this requirement will cause the job to exit. The same format
866 as \fBrate\fR is used for read vs write separation.
868 .BI rate_iops \fR=\fPint
869 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
870 specified independently of bandwidth. The same format as \fBrate\fR is used for
871 read vs write separation. If \fBblocksize\fR is a range, the smallest block
872 size is used as the metric.
874 .BI rate_iops_min \fR=\fPint
875 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
876 is used for read vs write separation.
878 .BI ratecycle \fR=\fPint
879 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
880 milliseconds. Default: 1000ms.
882 .BI latency_target \fR=\fPint
883 If set, fio will attempt to find the max performance point that the given
884 workload will run at while maintaining a latency below this target. The
885 values is given in microseconds. See \fBlatency_window\fR and
886 \fBlatency_percentile\fR.
888 .BI latency_window \fR=\fPint
889 Used with \fBlatency_target\fR to specify the sample window that the job
890 is run at varying queue depths to test the performance. The value is given
893 .BI latency_percentile \fR=\fPfloat
894 The percentage of IOs that must fall within the criteria specified by
895 \fBlatency_target\fR and \fBlatency_window\fR. If not set, this defaults
896 to 100.0, meaning that all IOs must be equal or below to the value set
897 by \fBlatency_target\fR.
899 .BI max_latency \fR=\fPint
900 If set, fio will exit the job if it exceeds this maximum latency. It will exit
903 .BI cpumask \fR=\fPint
904 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
905 may run on. See \fBsched_setaffinity\fR\|(2).
907 .BI cpus_allowed \fR=\fPstr
908 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
910 .BI cpus_allowed_policy \fR=\fPstr
911 Set the policy of how fio distributes the CPUs specified by \fBcpus_allowed\fR
912 or \fBcpumask\fR. Two policies are supported:
917 All jobs will share the CPU set specified.
920 Each job will get a unique CPU from the CPU set.
923 \fBshared\fR is the default behaviour, if the option isn't specified. If
924 \fBsplit\fR is specified, then fio will assign one cpu per job. If not enough
925 CPUs are given for the jobs listed, then fio will roundrobin the CPUs in
930 .BI numa_cpu_nodes \fR=\fPstr
931 Set this job running on specified NUMA nodes' CPUs. The arguments allow
932 comma delimited list of cpu numbers, A-B ranges, or 'all'.
934 .BI numa_mem_policy \fR=\fPstr
935 Set this job's memory policy and corresponding NUMA nodes. Format of
939 .B <mode>[:<nodelist>]
942 is one of the following memory policy:
944 .B default, prefer, bind, interleave, local
947 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
948 needed to be specified. For \fBprefer\fR, only one node is
949 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
950 comma delimited list of numbers, A-B ranges, or 'all'.
952 .BI startdelay \fR=\fPirange
953 Delay start of job for the specified number of seconds. Supports all time
954 suffixes to allow specification of hours, minutes, seconds and
955 milliseconds - seconds are the default if a unit is omitted.
956 Can be given as a range which causes each thread to choose randomly out of the
959 .BI runtime \fR=\fPint
960 Terminate processing after the specified number of seconds.
963 If given, run for the specified \fBruntime\fR duration even if the files are
964 completely read or written. The same workload will be repeated as many times
965 as \fBruntime\fR allows.
967 .BI ramp_time \fR=\fPint
968 If set, fio will run the specified workload for this amount of time before
969 logging any performance numbers. Useful for letting performance settle before
970 logging results, thus minimizing the runtime required for stable results. Note
971 that the \fBramp_time\fR is considered lead in time for a job, thus it will
972 increase the total runtime if a special timeout or runtime is specified.
974 .BI invalidate \fR=\fPbool
975 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
978 Use synchronous I/O for buffered writes. For the majority of I/O engines,
979 this means using O_SYNC. Default: false.
981 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
982 Allocation method for I/O unit buffer. Allowed values are:
987 Allocate memory with \fBmalloc\fR\|(3).
990 Use shared memory buffers allocated through \fBshmget\fR\|(2).
993 Same as \fBshm\fR, but use huge pages as backing.
996 Use \fBmmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
997 is given after the option in the format `:\fIfile\fR'.
1000 Same as \fBmmap\fR, but use huge files as backing.
1003 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
1004 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
1005 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
1006 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
1007 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
1008 and the documentation for that. Normally you just need to echo an appropriate
1009 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
1013 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
1014 This indicates the memory alignment of the IO memory buffers. Note that the
1015 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
1016 the alignment of the following buffers are given by the \fBbs\fR used. In
1017 other words, if using a \fBbs\fR that is a multiple of the page sized in the
1018 system, all buffers will be aligned to this value. If using a \fBbs\fR that
1019 is not page aligned, the alignment of subsequent IO memory buffers is the
1020 sum of the \fBiomem_align\fR and \fBbs\fR used.
1022 .BI hugepage\-size \fR=\fPint
1023 Defines the size of a huge page. Must be at least equal to the system setting.
1024 Should be a multiple of 1MB. Default: 4MB.
1027 Terminate all jobs when one finishes. Default: wait for each job to finish.
1029 .BI bwavgtime \fR=\fPint
1030 Average bandwidth calculations over the given time in milliseconds. Default:
1033 .BI iopsavgtime \fR=\fPint
1034 Average IOPS calculations over the given time in milliseconds. Default:
1037 .BI create_serialize \fR=\fPbool
1038 If true, serialize file creation for the jobs. Default: true.
1040 .BI create_fsync \fR=\fPbool
1041 \fBfsync\fR\|(2) data file after creation. Default: true.
1043 .BI create_on_open \fR=\fPbool
1044 If true, the files are not created until they are opened for IO by the job.
1046 .BI create_only \fR=\fPbool
1047 If true, fio will only run the setup phase of the job. If files need to be
1048 laid out or updated on disk, only that will be done. The actual job contents
1051 .BI pre_read \fR=\fPbool
1052 If this is given, files will be pre-read into memory before starting the given
1053 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
1054 pointless to pre-read and then drop the cache. This will only work for IO
1055 engines that are seekable, since they allow you to read the same data
1056 multiple times. Thus it will not work on eg network or splice IO.
1058 .BI unlink \fR=\fPbool
1059 Unlink job files when done. Default: false.
1061 .BI loops \fR=\fPint
1062 Specifies the number of iterations (runs of the same workload) of this job.
1065 .BI verify_only \fR=\fPbool
1066 Do not perform the specified workload, only verify data still matches previous
1067 invocation of this workload. This option allows one to check data multiple
1068 times at a later date without overwriting it. This option makes sense only for
1069 workloads that write data, and does not support workloads with the
1070 \fBtime_based\fR option set.
1072 .BI do_verify \fR=\fPbool
1073 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
1076 .BI verify \fR=\fPstr
1077 Method of verifying file contents after each iteration of the job. Allowed
1082 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 xxhash
1083 Store appropriate checksum in the header of each block. crc32c-intel is
1084 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
1085 not supported by the system.
1088 Write extra information about each I/O (timestamp, block number, etc.). The
1089 block number is verified. See \fBverify_pattern\fR as well.
1092 Pretend to verify. Used for testing internals.
1095 This option can be used for repeated burn-in tests of a system to make sure
1096 that the written data is also correctly read back. If the data direction given
1097 is a read or random read, fio will assume that it should verify a previously
1098 written file. If the data direction includes any form of write, the verify will
1099 be of the newly written data.
1102 .BI verifysort \fR=\fPbool
1103 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
1104 read them back in a sorted manner. Default: true.
1106 .BI verifysort_nr \fR=\fPint
1107 Pre-load and sort verify blocks for a read workload.
1109 .BI verify_offset \fR=\fPint
1110 Swap the verification header with data somewhere else in the block before
1111 writing. It is swapped back before verifying.
1113 .BI verify_interval \fR=\fPint
1114 Write the verification header for this number of bytes, which should divide
1115 \fBblocksize\fR. Default: \fBblocksize\fR.
1117 .BI verify_pattern \fR=\fPstr
1118 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
1119 with totally random bytes, but sometimes it's interesting to fill with a known
1120 pattern for io verification purposes. Depending on the width of the pattern,
1121 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
1122 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
1123 has to be a hex number that starts with either "0x" or "0X". Use with
1126 .BI verify_fatal \fR=\fPbool
1127 If true, exit the job on the first observed verification failure. Default:
1130 .BI verify_dump \fR=\fPbool
1131 If set, dump the contents of both the original data block and the data block we
1132 read off disk to files. This allows later analysis to inspect just what kind of
1133 data corruption occurred. Off by default.
1135 .BI verify_async \fR=\fPint
1136 Fio will normally verify IO inline from the submitting thread. This option
1137 takes an integer describing how many async offload threads to create for IO
1138 verification instead, causing fio to offload the duty of verifying IO contents
1139 to one or more separate threads. If using this offload option, even sync IO
1140 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
1141 allows them to have IO in flight while verifies are running.
1143 .BI verify_async_cpus \fR=\fPstr
1144 Tell fio to set the given CPU affinity on the async IO verification threads.
1145 See \fBcpus_allowed\fP for the format used.
1147 .BI verify_backlog \fR=\fPint
1148 Fio will normally verify the written contents of a job that utilizes verify
1149 once that job has completed. In other words, everything is written then
1150 everything is read back and verified. You may want to verify continually
1151 instead for a variety of reasons. Fio stores the meta data associated with an
1152 IO block in memory, so for large verify workloads, quite a bit of memory would
1153 be used up holding this meta data. If this option is enabled, fio will write
1154 only N blocks before verifying these blocks.
1156 .BI verify_backlog_batch \fR=\fPint
1157 Control how many blocks fio will verify if verify_backlog is set. If not set,
1158 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
1159 read back and verified). If \fBverify_backlog_batch\fR is less than
1160 \fBverify_backlog\fR then not all blocks will be verified, if
1161 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
1162 will be verified more than once.
1164 .BI trim_percentage \fR=\fPint
1165 Number of verify blocks to discard/trim.
1167 .BI trim_verify_zero \fR=\fPbool
1168 Verify that trim/discarded blocks are returned as zeroes.
1170 .BI trim_backlog \fR=\fPint
1171 Trim after this number of blocks are written.
1173 .BI trim_backlog_batch \fR=\fPint
1174 Trim this number of IO blocks.
1176 .BI experimental_verify \fR=\fPbool
1177 Enable experimental verification.
1179 .BI verify_state_save \fR=\fPbool
1180 When a job exits during the write phase of a verify workload, save its
1181 current state. This allows fio to replay up until that point, if the
1182 verify state is loaded for the verify read phase.
1184 .BI verify_state_load \fR=\fPbool
1185 If a verify termination trigger was used, fio stores the current write
1186 state of each thread. This can be used at verification time so that fio
1187 knows how far it should verify. Without this information, fio will run
1188 a full verification pass, according to the settings in the job file used.
1190 .B stonewall "\fR,\fP wait_for_previous"
1191 Wait for preceding jobs in the job file to exit before starting this one.
1192 \fBstonewall\fR implies \fBnew_group\fR.
1195 Start a new reporting group. If not given, all jobs in a file will be part
1196 of the same reporting group, unless separated by a stonewall.
1198 .BI numjobs \fR=\fPint
1199 Number of clones (processes/threads performing the same workload) of this job.
1203 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
1207 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
1208 with \fBfork\fR\|(2).
1210 .BI zonesize \fR=\fPint
1211 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
1213 .BI zonerange \fR=\fPint
1214 Give size of an IO zone. See \fBzoneskip\fR.
1216 .BI zoneskip \fR=\fPint
1217 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
1220 .BI write_iolog \fR=\fPstr
1221 Write the issued I/O patterns to the specified file. Specify a separate file
1222 for each job, otherwise the iologs will be interspersed and the file may be
1225 .BI read_iolog \fR=\fPstr
1226 Replay the I/O patterns contained in the specified file generated by
1227 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1229 .BI replay_no_stall \fR=\fPint
1230 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1231 attempts to respect timing information between I/Os. Enabling
1232 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1233 still respecting ordering.
1235 .BI replay_redirect \fR=\fPstr
1236 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1237 is to replay the IOPS onto the major/minor device that each IOP was recorded
1238 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1239 single specified device regardless of the device it was recorded from.
1241 .BI write_bw_log \fR=\fPstr
1242 If given, write a bandwidth log of the jobs in this job file. Can be used to
1243 store data of the bandwidth of the jobs in their lifetime. The included
1244 fio_generate_plots script uses gnuplot to turn these text files into nice
1245 graphs. See \fBwrite_lat_log\fR for behaviour of given filename. For this
1246 option, the postfix is _bw.x.log, where x is the index of the job (1..N,
1247 where N is the number of jobs)
1249 .BI write_lat_log \fR=\fPstr
1250 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1251 filename is given with this option, the default filename of
1252 "jobname_type.x.log" is used, where x is the index of the job (1..N, where
1253 N is the number of jobs). Even if the filename is given, fio will still
1254 append the type of log.
1256 .BI write_iops_log \fR=\fPstr
1257 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1258 option, the default filename of "jobname_type.x.log" is used, where x is the
1259 index of the job (1..N, where N is the number of jobs). Even if the filename
1260 is given, fio will still append the type of log.
1262 .BI log_avg_msec \fR=\fPint
1263 By default, fio will log an entry in the iops, latency, or bw log for every
1264 IO that completes. When writing to the disk log, that can quickly grow to a
1265 very large size. Setting this option makes fio average the each log entry
1266 over the specified period of time, reducing the resolution of the log.
1269 .BI log_offset \fR=\fPbool
1270 If this is set, the iolog options will include the byte offset for the IO
1271 entry as well as the other data values.
1273 .BI log_compression \fR=\fPint
1274 If this is set, fio will compress the IO logs as it goes, to keep the memory
1275 footprint lower. When a log reaches the specified size, that chunk is removed
1276 and compressed in the background. Given that IO logs are fairly highly
1277 compressible, this yields a nice memory savings for longer runs. The downside
1278 is that the compression will consume some background CPU cycles, so it may
1279 impact the run. This, however, is also true if the logging ends up consuming
1280 most of the system memory. So pick your poison. The IO logs are saved
1281 normally at the end of a run, by decompressing the chunks and storing them
1282 in the specified log file. This feature depends on the availability of zlib.
1284 .BI log_store_compressed \fR=\fPbool
1285 If set, and \fBlog\fR_compression is also set, fio will store the log files in
1286 a compressed format. They can be decompressed with fio, using the
1287 \fB\-\-inflate-log\fR command line parameter. The files will be stored with a
1290 .BI disable_lat \fR=\fPbool
1291 Disable measurements of total latency numbers. Useful only for cutting
1292 back the number of calls to \fBgettimeofday\fR\|(2), as that does impact performance at
1293 really high IOPS rates. Note that to really get rid of a large amount of these
1294 calls, this option must be used with disable_slat and disable_bw as well.
1296 .BI disable_clat \fR=\fPbool
1297 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1299 .BI disable_slat \fR=\fPbool
1300 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1302 .BI disable_bw_measurement \fR=\fPbool
1303 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1305 .BI lockmem \fR=\fPint
1306 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1307 simulate a smaller amount of memory. The amount specified is per worker.
1309 .BI exec_prerun \fR=\fPstr
1310 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1312 Output is redirected in a file called \fBjobname.prerun.txt\fR
1315 .BI exec_postrun \fR=\fPstr
1316 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1318 Output is redirected in a file called \fBjobname.postrun.txt\fR
1321 .BI ioscheduler \fR=\fPstr
1322 Attempt to switch the device hosting the file to the specified I/O scheduler.
1324 .BI disk_util \fR=\fPbool
1325 Generate disk utilization statistics if the platform supports it. Default: true.
1327 .BI clocksource \fR=\fPstr
1328 Use the given clocksource as the base of timing. The supported options are:
1332 \fBgettimeofday\fR\|(2)
1335 \fBclock_gettime\fR\|(2)
1338 Internal CPU clock source
1342 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1343 (and fio is heavy on time calls). Fio will automatically use this clocksource
1344 if it's supported and considered reliable on the system it is running on,
1345 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1346 means supporting TSC Invariant.
1348 .BI gtod_reduce \fR=\fPbool
1349 Enable all of the \fBgettimeofday\fR\|(2) reducing options (disable_clat, disable_slat,
1350 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1351 \fBgettimeofday\fR\|(2) call count. With this option enabled, we only do about 0.4% of
1352 the gtod() calls we would have done if all time keeping was enabled.
1354 .BI gtod_cpu \fR=\fPint
1355 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1356 the current time. Fio (and databases, for instance) are very intensive on
1357 \fBgettimeofday\fR\|(2) calls. With this option, you can set one CPU aside for doing
1358 nothing but logging current time to a shared memory location. Then the other
1359 threads/processes that run IO workloads need only copy that segment, instead of
1360 entering the kernel with a \fBgettimeofday\fR\|(2) call. The CPU set aside for doing
1361 these time calls will be excluded from other uses. Fio will manually clear it
1362 from the CPU mask of other jobs.
1364 .BI ignore_error \fR=\fPstr
1365 Sometimes you want to ignore some errors during test in that case you can specify
1366 error list for each error type.
1368 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1370 errors for given error type is separated with ':'.
1371 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1373 Example: ignore_error=EAGAIN,ENOSPC:122 .
1375 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1377 .BI error_dump \fR=\fPbool
1378 If set dump every error even if it is non fatal, true by default. If disabled
1379 only fatal error will be dumped
1381 .BI profile \fR=\fPstr
1382 Select a specific builtin performance test.
1384 .BI cgroup \fR=\fPstr
1385 Add job to this control group. If it doesn't exist, it will be created.
1386 The system must have a mounted cgroup blkio mount point for this to work. If
1387 your system doesn't have it mounted, you can do so with:
1389 # mount \-t cgroup \-o blkio none /cgroup
1391 .BI cgroup_weight \fR=\fPint
1392 Set the weight of the cgroup to this value. See the documentation that comes
1393 with the kernel, allowed values are in the range of 100..1000.
1395 .BI cgroup_nodelete \fR=\fPbool
1396 Normally fio will delete the cgroups it has created after the job completion.
1397 To override this behavior and to leave cgroups around after the job completion,
1398 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1399 cgroup files after job completion. Default: false
1402 Instead of running as the invoking user, set the user ID to this value before
1403 the thread/process does any work.
1406 Set group ID, see \fBuid\fR.
1408 .BI unit_base \fR=\fPint
1409 Base unit for reporting. Allowed values are:
1413 Use auto-detection (default).
1423 .BI flow_id \fR=\fPint
1424 The ID of the flow. If not specified, it defaults to being a global flow. See
1428 Weight in token-based flow control. If this value is used, then there is a
1429 \fBflow counter\fR which is used to regulate the proportion of activity between
1430 two or more jobs. fio attempts to keep this flow counter near zero. The
1431 \fBflow\fR parameter stands for how much should be added or subtracted to the
1432 flow counter on each iteration of the main I/O loop. That is, if one job has
1433 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1434 1:8 ratio in how much one runs vs the other.
1436 .BI flow_watermark \fR=\fPint
1437 The maximum value that the absolute value of the flow counter is allowed to
1438 reach before the job must wait for a lower value of the counter.
1440 .BI flow_sleep \fR=\fPint
1441 The period of time, in microseconds, to wait after the flow watermark has been
1442 exceeded before retrying operations
1444 .BI clat_percentiles \fR=\fPbool
1445 Enable the reporting of percentiles of completion latencies.
1447 .BI percentile_list \fR=\fPfloat_list
1448 Overwrite the default list of percentiles for completion
1449 latencies. Each number is a floating number in the range (0,100], and
1450 the maximum length of the list is 20. Use ':' to separate the
1451 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1452 report the values of completion latency below which 99.5% and 99.9% of
1453 the observed latencies fell, respectively.
1454 .SS "Ioengine Parameters List"
1455 Some parameters are only valid when a specific ioengine is in use. These are
1456 used identically to normal parameters, with the caveat that when used on the
1457 command line, they must come after the ioengine.
1459 .BI (cpu)cpuload \fR=\fPint
1460 Attempt to use the specified percentage of CPU cycles.
1462 .BI (cpu)cpuchunks \fR=\fPint
1463 Split the load into cycles of the given time. In microseconds.
1465 .BI (cpu)exit_on_io_done \fR=\fPbool
1466 Detect when IO threads are done, then exit.
1468 .BI (libaio)userspace_reap
1469 Normally, with the libaio engine in use, fio will use
1470 the io_getevents system call to reap newly returned events.
1471 With this flag turned on, the AIO ring will be read directly
1472 from user-space to reap events. The reaping mode is only
1473 enabled when polling for a minimum of 0 events (eg when
1474 iodepth_batch_complete=0).
1476 .BI (net,netsplice)hostname \fR=\fPstr
1477 The host name or IP address to use for TCP or UDP based IO.
1478 If the job is a TCP listener or UDP reader, the hostname is not
1479 used and must be omitted unless it is a valid UDP multicast address.
1481 .BI (net,netsplice)port \fR=\fPint
1482 The TCP or UDP port to bind to or connect to. If this is used with
1483 \fBnumjobs\fR to spawn multiple instances of the same job type, then
1484 this will be the starting port number since fio will use a range of ports.
1486 .BI (net,netsplice)interface \fR=\fPstr
1487 The IP address of the network interface used to send or receive UDP multicast
1490 .BI (net,netsplice)ttl \fR=\fPint
1491 Time-to-live value for outgoing UDP multicast packets. Default: 1
1493 .BI (net,netsplice)nodelay \fR=\fPbool
1494 Set TCP_NODELAY on TCP connections.
1496 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1497 The network protocol to use. Accepted values are:
1502 Transmission control protocol
1505 Transmission control protocol V6
1508 User datagram protocol
1511 User datagram protocol V6
1517 When the protocol is TCP or UDP, the port must also be given,
1518 as well as the hostname if the job is a TCP listener or UDP
1519 reader. For unix sockets, the normal filename option should be
1520 used and the port is invalid.
1523 .BI (net,netsplice)listen
1524 For TCP network connections, tell fio to listen for incoming
1525 connections rather than initiating an outgoing connection. The
1526 hostname must be omitted if this option is used.
1528 .BI (net, pingpong) \fR=\fPbool
1529 Normally a network writer will just continue writing data, and a network reader
1530 will just consume packets. If pingpong=1 is set, a writer will send its normal
1531 payload to the reader, then wait for the reader to send the same payload back.
1532 This allows fio to measure network latencies. The submission and completion
1533 latencies then measure local time spent sending or receiving, and the
1534 completion latency measures how long it took for the other end to receive and
1535 send back. For UDP multicast traffic pingpong=1 should only be set for a single
1536 reader when multiple readers are listening to the same address.
1538 .BI (net, window_size) \fR=\fPint
1539 Set the desired socket buffer size for the connection.
1541 .BI (net, mss) \fR=\fPint
1542 Set the TCP maximum segment size (TCP_MAXSEG).
1544 .BI (e4defrag,donorname) \fR=\fPstr
1545 File will be used as a block donor (swap extents between files)
1547 .BI (e4defrag,inplace) \fR=\fPint
1548 Configure donor file block allocation strategy
1551 Preallocate donor's file on init
1554 allocate space immediately inside defragment event, and free right after event
1557 .BI (rbd)rbdname \fR=\fPstr
1558 Specifies the name of the RBD.
1560 .BI (rbd)pool \fR=\fPstr
1561 Specifies the name of the Ceph pool containing the RBD.
1563 .BI (rbd)clientname \fR=\fPstr
1564 Specifies the username (without the 'client.' prefix) used to access the Ceph cluster.
1566 While running, \fBfio\fR will display the status of the created jobs. For
1570 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1573 The characters in the first set of brackets denote the current status of each
1574 threads. The possible values are:
1580 Setup but not started.
1586 Initialized, waiting.
1589 Running, doing sequential reads.
1592 Running, doing random reads.
1595 Running, doing sequential writes.
1598 Running, doing random writes.
1601 Running, doing mixed sequential reads/writes.
1604 Running, doing mixed random reads/writes.
1607 Running, currently waiting for \fBfsync\fR\|(2).
1610 Running, verifying written data.
1613 Exited, not reaped by main thread.
1616 Exited, thread reaped.
1620 The second set of brackets shows the estimated completion percentage of
1621 the current group. The third set shows the read and write I/O rate,
1622 respectively. Finally, the estimated run time of the job is displayed.
1624 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1625 for each thread, each group of threads, and each disk, in that order.
1627 Per-thread statistics first show the threads client number, group-id, and
1628 error code. The remaining figures are as follows:
1632 Number of megabytes of I/O performed.
1635 Average data rate (bandwidth).
1641 Submission latency minimum, maximum, average and standard deviation. This is
1642 the time it took to submit the I/O.
1645 Completion latency minimum, maximum, average and standard deviation. This
1646 is the time between submission and completion.
1649 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1650 and standard deviation.
1653 CPU usage statistics. Includes user and system time, number of context switches
1654 this thread went through and number of major and minor page faults.
1657 Distribution of I/O depths. Each depth includes everything less than (or equal)
1658 to it, but greater than the previous depth.
1661 Number of read/write requests issued, and number of short read/write requests.
1664 Distribution of I/O completion latencies. The numbers follow the same pattern
1668 The group statistics show:
1673 Number of megabytes I/O performed.
1676 Aggregate bandwidth of threads in the group.
1679 Minimum average bandwidth a thread saw.
1682 Maximum average bandwidth a thread saw.
1685 Shortest runtime of threads in the group.
1688 Longest runtime of threads in the group.
1692 Finally, disk statistics are printed with reads first:
1697 Number of I/Os performed by all groups.
1700 Number of merges in the I/O scheduler.
1703 Number of ticks we kept the disk busy.
1706 Total time spent in the disk queue.
1713 It is also possible to get fio to dump the current output while it is
1714 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1717 If the \fB\-\-minimal\fR / \fB\-\-append-terse\fR options are given, the
1718 results will be printed/appended in a semicolon-delimited format suitable for
1720 A job description (if provided) follows on a new line. Note that the first
1721 number in the line is the version number. If the output has to be changed
1722 for some reason, this number will be incremented by 1 to signify that
1723 change. The fields are:
1726 .B terse version, fio version, jobname, groupid, error
1730 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1734 .B min, max, mean, standard deviation
1738 .B min, max, mean, standard deviation
1740 Completion latency percentiles (20 fields):
1742 .B Xth percentile=usec
1746 .B min, max, mean, standard deviation
1750 .B min, max, aggregate percentage of total, mean, standard deviation
1756 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1760 .B min, max, mean, standard deviation
1764 .B min, max, mean, standard deviation
1766 Completion latency percentiles (20 fields):
1768 .B Xth percentile=usec
1772 .B min, max, mean, standard deviation
1776 .B min, max, aggregate percentage of total, mean, standard deviation
1782 .B user, system, context switches, major page faults, minor page faults
1785 IO depth distribution:
1787 .B <=1, 2, 4, 8, 16, 32, >=64
1790 IO latency distribution:
1794 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1798 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1802 Disk utilization (1 for each disk used):
1804 .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
1807 Error Info (dependent on continue_on_error, default off):
1809 .B total # errors, first error code
1812 .B text description (if provided in config - appears on newline)
1815 Normally you would run fio as a stand-alone application on the machine
1816 where the IO workload should be generated. However, it is also possible to
1817 run the frontend and backend of fio separately. This makes it possible to
1818 have a fio server running on the machine(s) where the IO workload should
1819 be running, while controlling it from another machine.
1821 To start the server, you would do:
1823 \fBfio \-\-server=args\fR
1825 on that machine, where args defines what fio listens to. The arguments
1826 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1827 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1828 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1829 listen to (only valid for TCP/IP, not a local socket). Some examples:
1833 Start a fio server, listening on all interfaces on the default port (8765).
1835 2) fio \-\-server=ip:hostname,4444
1837 Start a fio server, listening on IP belonging to hostname and on port 4444.
1839 3) fio \-\-server=ip6:::1,4444
1841 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1843 4) fio \-\-server=,4444
1845 Start a fio server, listening on all interfaces on port 4444.
1847 5) fio \-\-server=1.2.3.4
1849 Start a fio server, listening on IP 1.2.3.4 on the default port.
1851 6) fio \-\-server=sock:/tmp/fio.sock
1853 Start a fio server, listening on the local socket /tmp/fio.sock.
1855 When a server is running, you can connect to it from a client. The client
1858 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1860 where \-\-local-args are arguments that are local to the client where it is
1861 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1862 are sent to the server. The 'server' string follows the same format as it
1863 does on the server side, to allow IP/hostname/socket and port strings.
1864 You can connect to multiple clients as well, to do that you could run:
1866 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1868 If the job file is located on the fio server, then you can tell the server
1869 to load a local file as well. This is done by using \-\-remote-config:
1871 fio \-\-client=server \-\-remote-config /path/to/file.fio
1873 Then the fio serer will open this local (to the server) job file instead
1874 of being passed one from the client.
1878 was written by Jens Axboe <jens.axboe@oracle.com>,
1879 now Jens Axboe <axboe@fb.com>.
1881 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1882 on documentation by Jens Axboe.
1883 .SH "REPORTING BUGS"
1884 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1887 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1889 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob