1 .TH fio 1 "October 2013" "User Manual"
3 fio \- flexible I/O tester
6 [\fIoptions\fR] [\fIjobfile\fR]...
9 is a tool that will spawn a number of threads or processes doing a
10 particular type of I/O action as specified by the user.
11 The typical use of fio is to write a job file matching the I/O load
12 one wants to simulate.
15 .BI \-\-debug \fR=\fPtype
16 Enable verbose tracing of various fio actions. May be `all' for all types
17 or individual types separated by a comma (eg \-\-debug=io,file). `help' will
18 list all available tracing options.
20 .BI \-\-output \fR=\fPfilename
21 Write output to \fIfilename\fR.
23 .BI \-\-runtime \fR=\fPruntime
24 Limit run time to \fIruntime\fR seconds.
27 Generate per-job latency logs.
30 Generate per-job bandwidth logs.
33 Print statistics in a terse, semicolon-delimited format.
36 Print statistics in selected mode AND terse, semicolon-delimited format.
39 Display version information and exit.
41 .BI \-\-terse\-version \fR=\fPversion
42 Set terse version output format (Current version 3, or older version 2).
45 Display usage information and exit.
48 Perform test and validation of internal CPU clock
50 .BI \-\-crctest[\fR=\fPtest]
51 Test the speed of the builtin checksumming functions. If no argument is given,
52 all of them are tested. Or a comma separated list can be passed, in which
53 case the given ones are tested.
55 .BI \-\-cmdhelp \fR=\fPcommand
56 Print help information for \fIcommand\fR. May be `all' for all commands.
58 .BI \-\-enghelp \fR=\fPioengine[,command]
59 List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR.
61 .BI \-\-showcmd \fR=\fPjobfile
62 Convert \fIjobfile\fR to a set of command-line options.
64 .BI \-\-eta \fR=\fPwhen
65 Specifies when real-time ETA estimate should be printed. \fIwhen\fR may
66 be one of `always', `never' or `auto'.
68 .BI \-\-eta\-newline \fR=\fPtime
69 Force an ETA newline for every `time` period passed.
71 .BI \-\-status\-interval \fR=\fPtime
72 Report full output status every `time` period passed.
75 Turn on safety read-only checks, preventing any attempted write.
77 .BI \-\-section \fR=\fPsec
78 Only run section \fIsec\fR from job file. This option can be used multiple times to add more sections to run.
80 .BI \-\-alloc\-size \fR=\fPkb
81 Set the internal smalloc pool size to \fIkb\fP kilobytes.
83 .BI \-\-warnings\-fatal
84 All fio parser warnings are fatal, causing fio to exit with an error.
86 .BI \-\-max\-jobs \fR=\fPnr
87 Set the maximum allowed number of jobs (threads/processes) to support.
89 .BI \-\-server \fR=\fPargs
90 Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section.
92 .BI \-\-daemonize \fR=\fPpidfile
93 Background a fio server, writing the pid to the given pid file.
95 .BI \-\-client \fR=\fPhost
96 Instead of running the jobs locally, send and run them on the given host.
98 .BI \-\-idle\-prof \fR=\fPoption
99 Report cpu idleness on a system or percpu basis (\fIoption\fP=system,percpu) or run unit work calibration only (\fIoption\fP=calibrate).
100 .SH "JOB FILE FORMAT"
101 Job files are in `ini' format. They consist of one or more
102 job definitions, which begin with a job name in square brackets and
103 extend to the next job name. The job name can be any ASCII string
104 except `global', which has a special meaning. Following the job name is
105 a sequence of zero or more parameters, one per line, that define the
106 behavior of the job. Any line starting with a `;' or `#' character is
107 considered a comment and ignored.
109 If \fIjobfile\fR is specified as `-', the job file will be read from
112 The global section contains default parameters for jobs specified in the
113 job file. A job is only affected by global sections residing above it,
114 and there may be any number of global sections. Specific job definitions
115 may override any parameter set in global sections.
118 Some parameters may take arguments of a specific type.
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 Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
385 divided between the available files for the job. If not set, fio will use the
386 full size of the given files or devices. If the files do not exist, size
387 must be given. It is also possible to give size as a percentage between 1 and
388 100. If size=20% is given, fio will use 20% of the full size of the given
391 .BI 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.
399 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
400 Sets size to something really large and waits for ENOSPC (no space left on
401 device) as the terminating condition. Only makes sense with sequential write.
402 For a read workload, the mount point will be filled first then IO started on
403 the result. This option doesn't make sense if operating on a raw device node,
404 since the size of that is already known by the file system. Additionally,
405 writing beyond end-of-device will not return ENOSPC there.
407 .BI filesize \fR=\fPirange
408 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
409 for files at random within the given range, limited to \fBsize\fR in total (if
410 that is given). If \fBfilesize\fR is not specified, each created file is the
413 .BI file_append \fR=\fPbool
414 Perform IO after the end of the file. Normally fio will operate within the
415 size of a file. If this option is set, then fio will append to the file
416 instead. This has identical behavior to setting \fRoffset\fP to the size
417 of a file. This option is ignored on non-regular files.
419 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
420 Block size for I/O units. Default: 4k. Values for reads, writes, and trims
421 can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR
422 either of which may be empty to leave that value at its default. If a trailing
423 comma isn't given, the remainder will inherit the last value set.
425 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
426 Specify a range of I/O block sizes. The issued I/O unit will always be a
427 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
428 to both reads and writes if only one range is given, but can be specified
429 separately with a comma separating the values. Example: bsrange=1k-4k,2k-8k.
430 Also (see \fBblocksize\fR).
432 .BI bssplit \fR=\fPstr
433 This option allows even finer grained control of the block sizes issued,
434 not just even splits between them. With this option, you can weight various
435 block sizes for exact control of the issued IO for a job that has mixed
436 block sizes. The format of the option is bssplit=blocksize/percentage,
437 optionally adding as many definitions as needed separated by a colon.
438 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
439 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
440 splits to reads and writes. The format is identical to what the
441 \fBbs\fR option accepts, the read and write parts are separated with a
444 .B blocksize_unaligned\fR,\fP bs_unaligned
445 If set, any size in \fBblocksize_range\fR may be used. This typically won't
446 work with direct I/O, as that normally requires sector alignment.
448 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
449 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
450 the minimum blocksize given. Minimum alignment is typically 512b
451 for using direct IO, though it usually depends on the hardware block size.
452 This option is mutually exclusive with using a random map for files, so it
453 will turn off that option.
455 .BI bs_is_seq_rand \fR=\fPbool
456 If this option is set, fio will use the normal read,write blocksize settings as
457 sequential,random instead. Any random read or write will use the WRITE
458 blocksize settings, and any sequential read or write will use the READ
462 Initialize buffers with all zeros. Default: fill buffers with random data.
463 The resulting IO buffers will not be completely zeroed, unless
464 \fPscramble_buffers\fR is also turned off.
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 zeroes. Note that this is per block size unit, for file/disk
483 wide compression level that matches this setting, you'll also want to set
484 \fBrefill_buffers\fR.
486 .BI buffer_compress_chunk \fR=\fPint
487 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
488 big the ranges of random data and zeroed data is. Without this set, fio will
489 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
490 the remaining zeroed. With this set to some chunk size smaller than the block
491 size, fio can alternate random and zeroed data throughout the IO buffer.
493 .BI buffer_pattern \fR=\fPstr
494 If set, fio will fill the IO buffers with this pattern. If not set, the contents
495 of IO buffers is defined by the other options related to buffer contents. The
496 setting can be any pattern of bytes, and can be prefixed with 0x for hex
497 values. It may also be a string, where the string must then be wrapped with
500 .BI dedupe_percentage \fR=\fPint
501 If set, fio will generate this percentage of identical buffers when writing.
502 These buffers will be naturally dedupable. The contents of the buffers depend
503 on what other buffer compression settings have been set. It's possible to have
504 the individual buffers either fully compressible, or not at all. This option
505 only controls the distribution of unique buffers.
507 .BI nrfiles \fR=\fPint
508 Number of files to use for this job. Default: 1.
510 .BI openfiles \fR=\fPint
511 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
513 .BI file_service_type \fR=\fPstr
514 Defines how files to service are selected. The following types are defined:
519 Choose a file at random.
522 Round robin over opened files (default).
525 Do each file in the set sequentially.
528 The number of I/Os to issue before switching to a new file can be specified by
529 appending `:\fIint\fR' to the service type.
532 .BI ioengine \fR=\fPstr
533 Defines how the job issues I/O. The following types are defined:
538 Basic \fBread\fR\|(2) or \fBwrite\fR\|(2) I/O. \fBfseek\fR\|(2) is used to
539 position the I/O location.
542 Basic \fBpread\fR\|(2) or \fBpwrite\fR\|(2) I/O.
545 Basic \fBreadv\fR\|(2) or \fBwritev\fR\|(2) I/O. Will emulate queuing by
546 coalescing adjacent IOs into a single submission.
549 Basic \fBpreadv\fR\|(2) or \fBpwritev\fR\|(2) I/O.
552 Linux native asynchronous I/O. This ioengine defines engine specific options.
555 POSIX asynchronous I/O using \fBaio_read\fR\|(3) and \fBaio_write\fR\|(3).
558 Solaris native asynchronous I/O.
561 Windows native asynchronous I/O.
564 File is memory mapped with \fBmmap\fR\|(2) and data copied using
568 \fBsplice\fR\|(2) is used to transfer the data and \fBvmsplice\fR\|(2) to
569 transfer data from user-space to the kernel.
572 Use the syslet system calls to make regular read/write asynchronous.
575 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
576 the target is an sg character device, we use \fBread\fR\|(2) and
577 \fBwrite\fR\|(2) for asynchronous I/O.
580 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
581 itself and for debugging and testing purposes.
584 Transfer over the network. The protocol to be used can be defined with the
585 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
586 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
587 This ioengine defines engine specific options.
590 Like \fBnet\fR, but uses \fBsplice\fR\|(2) and \fBvmsplice\fR\|(2) to map data
591 and send/receive. This ioengine defines engine specific options.
594 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
595 \fBcpucycles\fR parameters.
598 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
599 approach to asynchronous I/O.
601 See <http://www.xmailserver.org/guasi\-lib.html>.
604 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
605 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
608 Loads an external I/O engine object file. Append the engine filename as
612 IO engine that does regular linux native fallocate call to simulate data
613 transfer as fio ioengine
615 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
617 DIR_WRITE does fallocate(,mode = 0)
619 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
622 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
623 request to DDIR_WRITE event
626 IO engine supporting direct access to Ceph Rados Block Devices (RBD) via librbd
627 without the need to use the kernel rbd driver. This ioengine defines engine specific
631 Using Glusterfs libgfapi sync interface to direct access to Glusterfs volumes without
632 having to go through FUSE. This ioengine defines engine specific
636 Using Glusterfs libgfapi async interface to direct access to Glusterfs volumes without
637 having to go through FUSE. This ioengine defines engine specific
641 Read and write through Hadoop (HDFS). The \fBfilename\fR option is used to
642 specify host,port of the hdfs name-node to connect. This engine interprets
643 offsets a little differently. In HDFS, files once created cannot be modified.
644 So random writes are not possible. To imitate this, libhdfs engine expects
645 bunch of small files to be created over HDFS, and engine will randomly pick a
646 file out of those files based on the offset generated by fio backend. (see the
647 example job file to create such files, use rw=write option). Please note, you
648 might want to set necessary environment variables to work with hdfs/libhdfs
654 .BI iodepth \fR=\fPint
655 Number of I/O units to keep in flight against the file. Note that increasing
656 iodepth beyond 1 will not affect synchronous ioengines (except for small
657 degress when verify_async is in use). Even async engines may impose OS
658 restrictions causing the desired depth not to be achieved. This may happen on
659 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
660 not async on that OS. Keep an eye on the IO depth distribution in the
661 fio output to verify that the achieved depth is as expected. Default: 1.
663 .BI iodepth_batch \fR=\fPint
664 Number of I/Os to submit at once. Default: \fBiodepth\fR.
666 .BI iodepth_batch_complete \fR=\fPint
667 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
668 means that we'll ask for a minimum of 1 IO in the retrieval process from the
669 kernel. The IO retrieval will go on until we hit the limit set by
670 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
671 completed events before queuing more IO. This helps reduce IO latency, at the
672 cost of more retrieval system calls.
674 .BI iodepth_low \fR=\fPint
675 Low watermark indicating when to start filling the queue again. Default:
678 .BI direct \fR=\fPbool
679 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
681 .BI atomic \fR=\fPbool
682 If value is true, attempt to use atomic direct IO. Atomic writes are guaranteed
683 to be stable once acknowledged by the operating system. Only Linux supports
686 .BI buffered \fR=\fPbool
687 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
690 .BI offset \fR=\fPint
691 Offset in the file to start I/O. Data before the offset will not be touched.
693 .BI offset_increment \fR=\fPint
694 If this is provided, then the real offset becomes the
695 offset + offset_increment * thread_number, where the thread number is a
696 counter that starts at 0 and is incremented for each sub-job (i.e. when
697 numjobs option is specified). This option is useful if there are several jobs
698 which are intended to operate on a file in parallel disjoint segments, with
699 even spacing between the starting points.
701 .BI number_ios \fR=\fPint
702 Fio will normally perform IOs until it has exhausted the size of the region
703 set by \fBsize\fR, or if it exhaust the allocated time (or hits an error
704 condition). With this setting, the range/size can be set independently of
705 the number of IOs to perform. When fio reaches this number, it will exit
706 normally and report status. Note that this does not extend the amount
707 of IO that will be done, it will only stop fio if this condition is met
708 before other end-of-job criteria.
711 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
712 0, don't sync. Default: 0.
714 .BI fdatasync \fR=\fPint
715 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
716 data parts of the file. Default: 0.
718 .BI write_barrier \fR=\fPint
719 Make every Nth write a barrier write.
721 .BI sync_file_range \fR=\fPstr:int
722 Use \fBsync_file_range\fR\|(2) for every \fRval\fP number of write operations. Fio will
723 track range of writes that have happened since the last \fBsync_file_range\fR\|(2) call.
724 \fRstr\fP can currently be one or more of:
728 SYNC_FILE_RANGE_WAIT_BEFORE
731 SYNC_FILE_RANGE_WRITE
734 SYNC_FILE_RANGE_WRITE
738 So if you do sync_file_range=wait_before,write:8, fio would use
739 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
740 Also see the \fBsync_file_range\fR\|(2) man page. This option is Linux specific.
742 .BI overwrite \fR=\fPbool
743 If writing, setup the file first and do overwrites. Default: false.
745 .BI end_fsync \fR=\fPbool
746 Sync file contents when a write stage has completed. Default: false.
748 .BI fsync_on_close \fR=\fPbool
749 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
750 it will happen on every close, not just at the end of the job. Default: false.
752 .BI rwmixread \fR=\fPint
753 Percentage of a mixed workload that should be reads. Default: 50.
755 .BI rwmixwrite \fR=\fPint
756 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
757 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
758 overrides the first. This may interfere with a given rate setting, if fio is
759 asked to limit reads or writes to a certain rate. If that is the case, then
760 the distribution may be skewed. Default: 50.
762 .BI random_distribution \fR=\fPstr:float
763 By default, fio will use a completely uniform random distribution when asked
764 to perform random IO. Sometimes it is useful to skew the distribution in
765 specific ways, ensuring that some parts of the data is more hot than others.
766 Fio includes the following distribution models:
770 Uniform random distribution
780 When using a zipf or pareto distribution, an input value is also needed to
781 define the access pattern. For zipf, this is the zipf theta. For pareto,
782 it's the pareto power. Fio includes a test program, genzipf, that can be
783 used visualize what the given input values will yield in terms of hit rates.
784 If you wanted to use zipf with a theta of 1.2, you would use
785 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
786 fio will disable use of the random map.
788 .BI percentage_random \fR=\fPint
789 For a random workload, set how big a percentage should be random. This defaults
790 to 100%, in which case the workload is fully random. It can be set from
791 anywhere from 0 to 100. Setting it to 0 would make the workload fully
792 sequential. It is possible to set different values for reads, writes, and
793 trim. To do so, simply use a comma separated list. See \fBblocksize\fR.
796 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
797 this parameter is given, a new offset will be chosen without looking at past
798 I/O history. This parameter is mutually exclusive with \fBverify\fR.
800 .BI softrandommap \fR=\fPbool
801 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
802 fails to allocate the map, if this option is set it will continue without a
803 random block map. As coverage will not be as complete as with random maps, this
804 option is disabled by default.
806 .BI random_generator \fR=\fPstr
807 Fio supports the following engines for generating IO offsets for random IO:
811 Strong 2^88 cycle random number generator
814 Linear feedback shift register generator
818 Tausworthe is a strong random number generator, but it requires tracking on the
819 side if we want to ensure that blocks are only read or written once. LFSR
820 guarantees that we never generate the same offset twice, and it's also less
821 computationally expensive. It's not a true random generator, however, though
822 for IO purposes it's typically good enough. LFSR only works with single block
823 sizes, not with workloads that use multiple block sizes. If used with such a
824 workload, fio may read or write some blocks multiple times.
827 Run job with given nice value. See \fBnice\fR\|(2).
830 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
833 .BI prioclass \fR=\fPint
834 Set I/O priority class. See \fBionice\fR\|(1).
836 .BI thinktime \fR=\fPint
837 Stall job for given number of microseconds between issuing I/Os.
839 .BI thinktime_spin \fR=\fPint
840 Pretend to spend CPU time for given number of microseconds, sleeping the rest
841 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
843 .BI thinktime_blocks \fR=\fPint
844 Only valid if thinktime is set - control how many blocks to issue, before
845 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
846 make fio wait \fBthinktime\fR microseconds after every block. This
847 effectively makes any queue depth setting redundant, since no more than 1 IO
848 will be queued before we have to complete it and do our thinktime. In other
849 words, this setting effectively caps the queue depth if the latter is larger.
853 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
854 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
855 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
856 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
857 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
858 limit writes (to 500KB/sec), the latter will only limit reads.
860 .BI ratemin \fR=\fPint
861 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
862 Failing to meet this requirement will cause the job to exit. The same format
863 as \fBrate\fR is used for read vs write separation.
865 .BI rate_iops \fR=\fPint
866 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
867 specified independently of bandwidth. The same format as \fBrate\fR is used for
868 read vs write separation. If \fBblocksize\fR is a range, the smallest block
869 size is used as the metric.
871 .BI rate_iops_min \fR=\fPint
872 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
873 is used for read vs write separation.
875 .BI ratecycle \fR=\fPint
876 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
877 milliseconds. Default: 1000ms.
879 .BI latency_target \fR=\fPint
880 If set, fio will attempt to find the max performance point that the given
881 workload will run at while maintaining a latency below this target. The
882 values is given in microseconds. See \fBlatency_window\fR and
883 \fBlatency_percentile\fR.
885 .BI latency_window \fR=\fPint
886 Used with \fBlatency_target\fR to specify the sample window that the job
887 is run at varying queue depths to test the performance. The value is given
890 .BI latency_percentile \fR=\fPfloat
891 The percentage of IOs that must fall within the criteria specified by
892 \fBlatency_target\fR and \fBlatency_window\fR. If not set, this defaults
893 to 100.0, meaning that all IOs must be equal or below to the value set
894 by \fBlatency_target\fR.
896 .BI max_latency \fR=\fPint
897 If set, fio will exit the job if it exceeds this maximum latency. It will exit
900 .BI cpumask \fR=\fPint
901 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
902 may run on. See \fBsched_setaffinity\fR\|(2).
904 .BI cpus_allowed \fR=\fPstr
905 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
907 .BI cpus_allowed_policy \fR=\fPstr
908 Set the policy of how fio distributes the CPUs specified by \fBcpus_allowed\fR
909 or \fBcpumask\fR. Two policies are supported:
914 All jobs will share the CPU set specified.
917 Each job will get a unique CPU from the CPU set.
920 \fBshared\fR is the default behaviour, if the option isn't specified. If
921 \fBsplit\fR is specified, then fio will assign one cpu per job. If not enough
922 CPUs are given for the jobs listed, then fio will roundrobin the CPUs in
927 .BI numa_cpu_nodes \fR=\fPstr
928 Set this job running on specified NUMA nodes' CPUs. The arguments allow
929 comma delimited list of cpu numbers, A-B ranges, or 'all'.
931 .BI numa_mem_policy \fR=\fPstr
932 Set this job's memory policy and corresponding NUMA nodes. Format of
936 .B <mode>[:<nodelist>]
939 is one of the following memory policy:
941 .B default, prefer, bind, interleave, local
944 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
945 needed to be specified. For \fBprefer\fR, only one node is
946 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
947 comma delimited list of numbers, A-B ranges, or 'all'.
949 .BI startdelay \fR=\fPirange
950 Delay start of job for the specified number of seconds. Supports all time
951 suffixes to allow specification of hours, minutes, seconds and
952 milliseconds - seconds are the default if a unit is ommited.
953 Can be given as a range which causes each thread to choose randomly out of the
956 .BI runtime \fR=\fPint
957 Terminate processing after the specified number of seconds.
960 If given, run for the specified \fBruntime\fR duration even if the files are
961 completely read or written. The same workload will be repeated as many times
962 as \fBruntime\fR allows.
964 .BI ramp_time \fR=\fPint
965 If set, fio will run the specified workload for this amount of time before
966 logging any performance numbers. Useful for letting performance settle before
967 logging results, thus minimizing the runtime required for stable results. Note
968 that the \fBramp_time\fR is considered lead in time for a job, thus it will
969 increase the total runtime if a special timeout or runtime is specified.
971 .BI invalidate \fR=\fPbool
972 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
975 Use synchronous I/O for buffered writes. For the majority of I/O engines,
976 this means using O_SYNC. Default: false.
978 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
979 Allocation method for I/O unit buffer. Allowed values are:
984 Allocate memory with \fBmalloc\fR\|(3).
987 Use shared memory buffers allocated through \fBshmget\fR\|(2).
990 Same as \fBshm\fR, but use huge pages as backing.
993 Use \fBmmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
994 is given after the option in the format `:\fIfile\fR'.
997 Same as \fBmmap\fR, but use huge files as backing.
1000 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
1001 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
1002 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
1003 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
1004 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
1005 and the documentation for that. Normally you just need to echo an appropriate
1006 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
1010 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
1011 This indicates the memory alignment of the IO memory buffers. Note that the
1012 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
1013 the alignment of the following buffers are given by the \fBbs\fR used. In
1014 other words, if using a \fBbs\fR that is a multiple of the page sized in the
1015 system, all buffers will be aligned to this value. If using a \fBbs\fR that
1016 is not page aligned, the alignment of subsequent IO memory buffers is the
1017 sum of the \fBiomem_align\fR and \fBbs\fR used.
1019 .BI hugepage\-size \fR=\fPint
1020 Defines the size of a huge page. Must be at least equal to the system setting.
1021 Should be a multiple of 1MB. Default: 4MB.
1024 Terminate all jobs when one finishes. Default: wait for each job to finish.
1026 .BI bwavgtime \fR=\fPint
1027 Average bandwidth calculations over the given time in milliseconds. Default:
1030 .BI iopsavgtime \fR=\fPint
1031 Average IOPS calculations over the given time in milliseconds. Default:
1034 .BI create_serialize \fR=\fPbool
1035 If true, serialize file creation for the jobs. Default: true.
1037 .BI create_fsync \fR=\fPbool
1038 \fBfsync\fR\|(2) data file after creation. Default: true.
1040 .BI create_on_open \fR=\fPbool
1041 If true, the files are not created until they are opened for IO by the job.
1043 .BI create_only \fR=\fPbool
1044 If true, fio will only run the setup phase of the job. If files need to be
1045 laid out or updated on disk, only that will be done. The actual job contents
1048 .BI pre_read \fR=\fPbool
1049 If this is given, files will be pre-read into memory before starting the given
1050 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
1051 pointless to pre-read and then drop the cache. This will only work for IO
1052 engines that are seekable, since they allow you to read the same data
1053 multiple times. Thus it will not work on eg network or splice IO.
1055 .BI unlink \fR=\fPbool
1056 Unlink job files when done. Default: false.
1058 .BI loops \fR=\fPint
1059 Specifies the number of iterations (runs of the same workload) of this job.
1062 .BI verify_only \fR=\fPbool
1063 Do not perform the specified workload, only verify data still matches previous
1064 invocation of this workload. This option allows one to check data multiple
1065 times at a later date without overwriting it. This option makes sense only for
1066 workloads that write data, and does not support workloads with the
1067 \fBtime_based\fR option set.
1069 .BI do_verify \fR=\fPbool
1070 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
1073 .BI verify \fR=\fPstr
1074 Method of verifying file contents after each iteration of the job. Allowed
1079 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 xxhash
1080 Store appropriate checksum in the header of each block. crc32c-intel is
1081 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
1082 not supported by the system.
1085 Write extra information about each I/O (timestamp, block number, etc.). The
1086 block number is verified. See \fBverify_pattern\fR as well.
1089 Pretend to verify. Used for testing internals.
1092 This option can be used for repeated burn-in tests of a system to make sure
1093 that the written data is also correctly read back. If the data direction given
1094 is a read or random read, fio will assume that it should verify a previously
1095 written file. If the data direction includes any form of write, the verify will
1096 be of the newly written data.
1099 .BI verifysort \fR=\fPbool
1100 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
1101 read them back in a sorted manner. Default: true.
1103 .BI verifysort_nr \fR=\fPint
1104 Pre-load and sort verify blocks for a read workload.
1106 .BI verify_offset \fR=\fPint
1107 Swap the verification header with data somewhere else in the block before
1108 writing. It is swapped back before verifying.
1110 .BI verify_interval \fR=\fPint
1111 Write the verification header for this number of bytes, which should divide
1112 \fBblocksize\fR. Default: \fBblocksize\fR.
1114 .BI verify_pattern \fR=\fPstr
1115 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
1116 with totally random bytes, but sometimes it's interesting to fill with a known
1117 pattern for io verification purposes. Depending on the width of the pattern,
1118 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
1119 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
1120 has to be a hex number that starts with either "0x" or "0X". Use with
1123 .BI verify_fatal \fR=\fPbool
1124 If true, exit the job on the first observed verification failure. Default:
1127 .BI verify_dump \fR=\fPbool
1128 If set, dump the contents of both the original data block and the data block we
1129 read off disk to files. This allows later analysis to inspect just what kind of
1130 data corruption occurred. Off by default.
1132 .BI verify_async \fR=\fPint
1133 Fio will normally verify IO inline from the submitting thread. This option
1134 takes an integer describing how many async offload threads to create for IO
1135 verification instead, causing fio to offload the duty of verifying IO contents
1136 to one or more separate threads. If using this offload option, even sync IO
1137 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
1138 allows them to have IO in flight while verifies are running.
1140 .BI verify_async_cpus \fR=\fPstr
1141 Tell fio to set the given CPU affinity on the async IO verification threads.
1142 See \fBcpus_allowed\fP for the format used.
1144 .BI verify_backlog \fR=\fPint
1145 Fio will normally verify the written contents of a job that utilizes verify
1146 once that job has completed. In other words, everything is written then
1147 everything is read back and verified. You may want to verify continually
1148 instead for a variety of reasons. Fio stores the meta data associated with an
1149 IO block in memory, so for large verify workloads, quite a bit of memory would
1150 be used up holding this meta data. If this option is enabled, fio will write
1151 only N blocks before verifying these blocks.
1153 .BI verify_backlog_batch \fR=\fPint
1154 Control how many blocks fio will verify if verify_backlog is set. If not set,
1155 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
1156 read back and verified). If \fBverify_backlog_batch\fR is less than
1157 \fBverify_backlog\fR then not all blocks will be verified, if
1158 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
1159 will be verified more than once.
1161 .BI trim_percentage \fR=\fPint
1162 Number of verify blocks to discard/trim.
1164 .BI trim_verify_zero \fR=\fPbool
1165 Verify that trim/discarded blocks are returned as zeroes.
1167 .BI trim_backlog \fR=\fPint
1168 Trim after this number of blocks are written.
1170 .BI trim_backlog_batch \fR=\fPint
1171 Trim this number of IO blocks.
1173 .BI experimental_verify \fR=\fPbool
1174 Enable experimental verification.
1176 .B stonewall "\fR,\fP wait_for_previous"
1177 Wait for preceding jobs in the job file to exit before starting this one.
1178 \fBstonewall\fR implies \fBnew_group\fR.
1181 Start a new reporting group. If not given, all jobs in a file will be part
1182 of the same reporting group, unless separated by a stonewall.
1184 .BI numjobs \fR=\fPint
1185 Number of clones (processes/threads performing the same workload) of this job.
1189 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
1193 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
1194 with \fBfork\fR\|(2).
1196 .BI zonesize \fR=\fPint
1197 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
1199 .BI zonerange \fR=\fPint
1200 Give size of an IO zone. See \fBzoneskip\fR.
1202 .BI zoneskip \fR=\fPint
1203 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
1206 .BI write_iolog \fR=\fPstr
1207 Write the issued I/O patterns to the specified file. Specify a separate file
1208 for each job, otherwise the iologs will be interspersed and the file may be
1211 .BI read_iolog \fR=\fPstr
1212 Replay the I/O patterns contained in the specified file generated by
1213 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1215 .BI replay_no_stall \fR=\fPint
1216 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1217 attempts to respect timing information between I/Os. Enabling
1218 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1219 still respecting ordering.
1221 .BI replay_redirect \fR=\fPstr
1222 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1223 is to replay the IOPS onto the major/minor device that each IOP was recorded
1224 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1225 single specified device regardless of the device it was recorded from.
1227 .BI write_bw_log \fR=\fPstr
1228 If given, write a bandwidth log of the jobs in this job file. Can be used to
1229 store data of the bandwidth of the jobs in their lifetime. The included
1230 fio_generate_plots script uses gnuplot to turn these text files into nice
1231 graphs. See \fBwrite_lat_log\fR for behaviour of given filename. For this
1232 option, the postfix is _bw.x.log, where x is the index of the job (1..N,
1233 where N is the number of jobs)
1235 .BI write_lat_log \fR=\fPstr
1236 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1237 filename is given with this option, the default filename of
1238 "jobname_type.x.log" is used, where x is the index of the job (1..N, where
1239 N is the number of jobs). Even if the filename is given, fio will still
1240 append the type of log.
1242 .BI write_iops_log \fR=\fPstr
1243 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1244 option, the default filename of "jobname_type.x.log" is used, where x is the
1245 index of the job (1..N, where N is the number of jobs). Even if the filename
1246 is given, fio will still append the type of log.
1248 .BI log_avg_msec \fR=\fPint
1249 By default, fio will log an entry in the iops, latency, or bw log for every
1250 IO that completes. When writing to the disk log, that can quickly grow to a
1251 very large size. Setting this option makes fio average the each log entry
1252 over the specified period of time, reducing the resolution of the log.
1255 .BI log_offset \fR=\fPbool
1256 If this is set, the iolog options will include the byte offset for the IO
1257 entry as well as the other data values.
1259 .BI log_compression \fR=\fPint
1260 If this is set, fio will compress the IO logs as it goes, to keep the memory
1261 footprint lower. When a log reaches the specified size, that chunk is removed
1262 and compressed in the background. Given that IO logs are fairly highly
1263 compressible, this yields a nice memory savings for longer runs. The downside
1264 is that the compression will consume some background CPU cycles, so it may
1265 impact the run. This, however, is also true if the logging ends up consuming
1266 most of the system memory. So pick your poison. The IO logs are saved
1267 normally at the end of a run, by decompressing the chunks and storing them
1268 in the specified log file. This feature depends on the availability of zlib.
1270 .BI log_store_compressed \fR=\fPbool
1271 If set, and \fBlog\fR_compression is also set, fio will store the log files in
1272 a compressed format. They can be decompressed with fio, using the
1273 \fB\-\-inflate-log\fR command line parameter. The files will be stored with a
1276 .BI disable_lat \fR=\fPbool
1277 Disable measurements of total latency numbers. Useful only for cutting
1278 back the number of calls to \fBgettimeofday\fR\|(2), as that does impact performance at
1279 really high IOPS rates. Note that to really get rid of a large amount of these
1280 calls, this option must be used with disable_slat and disable_bw as well.
1282 .BI disable_clat \fR=\fPbool
1283 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1285 .BI disable_slat \fR=\fPbool
1286 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1288 .BI disable_bw_measurement \fR=\fPbool
1289 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1291 .BI lockmem \fR=\fPint
1292 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1293 simulate a smaller amount of memory. The amount specified is per worker.
1295 .BI exec_prerun \fR=\fPstr
1296 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1298 Output is redirected in a file called \fBjobname.prerun.txt\fR
1301 .BI exec_postrun \fR=\fPstr
1302 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1304 Output is redirected in a file called \fBjobname.postrun.txt\fR
1307 .BI ioscheduler \fR=\fPstr
1308 Attempt to switch the device hosting the file to the specified I/O scheduler.
1310 .BI disk_util \fR=\fPbool
1311 Generate disk utilization statistics if the platform supports it. Default: true.
1313 .BI clocksource \fR=\fPstr
1314 Use the given clocksource as the base of timing. The supported options are:
1318 \fBgettimeofday\fR\|(2)
1321 \fBclock_gettime\fR\|(2)
1324 Internal CPU clock source
1328 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1329 (and fio is heavy on time calls). Fio will automatically use this clocksource
1330 if it's supported and considered reliable on the system it is running on,
1331 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1332 means supporting TSC Invariant.
1334 .BI gtod_reduce \fR=\fPbool
1335 Enable all of the \fBgettimeofday\fR\|(2) reducing options (disable_clat, disable_slat,
1336 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1337 \fBgettimeofday\fR\|(2) call count. With this option enabled, we only do about 0.4% of
1338 the gtod() calls we would have done if all time keeping was enabled.
1340 .BI gtod_cpu \fR=\fPint
1341 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1342 the current time. Fio (and databases, for instance) are very intensive on
1343 \fBgettimeofday\fR\|(2) calls. With this option, you can set one CPU aside for doing
1344 nothing but logging current time to a shared memory location. Then the other
1345 threads/processes that run IO workloads need only copy that segment, instead of
1346 entering the kernel with a \fBgettimeofday\fR\|(2) call. The CPU set aside for doing
1347 these time calls will be excluded from other uses. Fio will manually clear it
1348 from the CPU mask of other jobs.
1350 .BI ignore_error \fR=\fPstr
1351 Sometimes you want to ignore some errors during test in that case you can specify
1352 error list for each error type.
1354 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1356 errors for given error type is separated with ':'.
1357 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1359 Example: ignore_error=EAGAIN,ENOSPC:122 .
1361 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1363 .BI error_dump \fR=\fPbool
1364 If set dump every error even if it is non fatal, true by default. If disabled
1365 only fatal error will be dumped
1367 .BI profile \fR=\fPstr
1368 Select a specific builtin performance test.
1370 .BI cgroup \fR=\fPstr
1371 Add job to this control group. If it doesn't exist, it will be created.
1372 The system must have a mounted cgroup blkio mount point for this to work. If
1373 your system doesn't have it mounted, you can do so with:
1375 # mount \-t cgroup \-o blkio none /cgroup
1377 .BI cgroup_weight \fR=\fPint
1378 Set the weight of the cgroup to this value. See the documentation that comes
1379 with the kernel, allowed values are in the range of 100..1000.
1381 .BI cgroup_nodelete \fR=\fPbool
1382 Normally fio will delete the cgroups it has created after the job completion.
1383 To override this behavior and to leave cgroups around after the job completion,
1384 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1385 cgroup files after job completion. Default: false
1388 Instead of running as the invoking user, set the user ID to this value before
1389 the thread/process does any work.
1392 Set group ID, see \fBuid\fR.
1394 .BI unit_base \fR=\fPint
1395 Base unit for reporting. Allowed values are:
1399 Use auto-detection (default).
1409 .BI flow_id \fR=\fPint
1410 The ID of the flow. If not specified, it defaults to being a global flow. See
1414 Weight in token-based flow control. If this value is used, then there is a
1415 \fBflow counter\fR which is used to regulate the proportion of activity between
1416 two or more jobs. fio attempts to keep this flow counter near zero. The
1417 \fBflow\fR parameter stands for how much should be added or subtracted to the
1418 flow counter on each iteration of the main I/O loop. That is, if one job has
1419 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1420 1:8 ratio in how much one runs vs the other.
1422 .BI flow_watermark \fR=\fPint
1423 The maximum value that the absolute value of the flow counter is allowed to
1424 reach before the job must wait for a lower value of the counter.
1426 .BI flow_sleep \fR=\fPint
1427 The period of time, in microseconds, to wait after the flow watermark has been
1428 exceeded before retrying operations
1430 .BI clat_percentiles \fR=\fPbool
1431 Enable the reporting of percentiles of completion latencies.
1433 .BI percentile_list \fR=\fPfloat_list
1434 Overwrite the default list of percentiles for completion
1435 latencies. Each number is a floating number in the range (0,100], and
1436 the maximum length of the list is 20. Use ':' to separate the
1437 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1438 report the values of completion latency below which 99.5% and 99.9% of
1439 the observed latencies fell, respectively.
1440 .SS "Ioengine Parameters List"
1441 Some parameters are only valid when a specific ioengine is in use. These are
1442 used identically to normal parameters, with the caveat that when used on the
1443 command line, they must come after the ioengine.
1445 .BI (cpu)cpuload \fR=\fPint
1446 Attempt to use the specified percentage of CPU cycles.
1448 .BI (cpu)cpuchunks \fR=\fPint
1449 Split the load into cycles of the given time. In microseconds.
1451 .BI (cpu)exit_on_io_done \fR=\fPbool
1452 Detect when IO threads are done, then exit.
1454 .BI (libaio)userspace_reap
1455 Normally, with the libaio engine in use, fio will use
1456 the io_getevents system call to reap newly returned events.
1457 With this flag turned on, the AIO ring will be read directly
1458 from user-space to reap events. The reaping mode is only
1459 enabled when polling for a minimum of 0 events (eg when
1460 iodepth_batch_complete=0).
1462 .BI (net,netsplice)hostname \fR=\fPstr
1463 The host name or IP address to use for TCP or UDP based IO.
1464 If the job is a TCP listener or UDP reader, the hostname is not
1465 used and must be omitted unless it is a valid UDP multicast address.
1467 .BI (net,netsplice)port \fR=\fPint
1468 The TCP or UDP port to bind to or connect to. If this is used with
1469 \fBnumjobs\fR to spawn multiple instances of the same job type, then
1470 this will be the starting port number since fio will use a range of ports.
1472 .BI (net,netsplice)interface \fR=\fPstr
1473 The IP address of the network interface used to send or receive UDP multicast
1476 .BI (net,netsplice)ttl \fR=\fPint
1477 Time-to-live value for outgoing UDP multicast packets. Default: 1
1479 .BI (net,netsplice)nodelay \fR=\fPbool
1480 Set TCP_NODELAY on TCP connections.
1482 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1483 The network protocol to use. Accepted values are:
1488 Transmission control protocol
1491 Transmission control protocol V6
1494 User datagram protocol
1497 User datagram protocol V6
1503 When the protocol is TCP or UDP, the port must also be given,
1504 as well as the hostname if the job is a TCP listener or UDP
1505 reader. For unix sockets, the normal filename option should be
1506 used and the port is invalid.
1509 .BI (net,netsplice)listen
1510 For TCP network connections, tell fio to listen for incoming
1511 connections rather than initiating an outgoing connection. The
1512 hostname must be omitted if this option is used.
1514 .BI (net, pingpong) \fR=\fPbool
1515 Normally a network writer will just continue writing data, and a network reader
1516 will just consume packets. If pingpong=1 is set, a writer will send its normal
1517 payload to the reader, then wait for the reader to send the same payload back.
1518 This allows fio to measure network latencies. The submission and completion
1519 latencies then measure local time spent sending or receiving, and the
1520 completion latency measures how long it took for the other end to receive and
1521 send back. For UDP multicast traffic pingpong=1 should only be set for a single
1522 reader when multiple readers are listening to the same address.
1524 .BI (net, window_size) \fR=\fPint
1525 Set the desired socket buffer size for the connection.
1527 .BI (net, mss) \fR=\fPint
1528 Set the TCP maximum segment size (TCP_MAXSEG).
1530 .BI (e4defrag,donorname) \fR=\fPstr
1531 File will be used as a block donor (swap extents between files)
1533 .BI (e4defrag,inplace) \fR=\fPint
1534 Configure donor file block allocation strategy
1537 Preallocate donor's file on init
1540 allocate space immediately inside defragment event, and free right after event
1543 .BI (rbd)rbdname \fR=\fPstr
1544 Specifies the name of the RBD.
1546 .BI (rbd)pool \fR=\fPstr
1547 Specifies the name of the Ceph pool containing the RBD.
1549 .BI (rbd)clientname \fR=\fPstr
1550 Specifies the username (without the 'client.' prefix) used to access the Ceph cluster.
1552 While running, \fBfio\fR will display the status of the created jobs. For
1556 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1559 The characters in the first set of brackets denote the current status of each
1560 threads. The possible values are:
1566 Setup but not started.
1572 Initialized, waiting.
1575 Running, doing sequential reads.
1578 Running, doing random reads.
1581 Running, doing sequential writes.
1584 Running, doing random writes.
1587 Running, doing mixed sequential reads/writes.
1590 Running, doing mixed random reads/writes.
1593 Running, currently waiting for \fBfsync\fR\|(2).
1596 Running, verifying written data.
1599 Exited, not reaped by main thread.
1602 Exited, thread reaped.
1606 The second set of brackets shows the estimated completion percentage of
1607 the current group. The third set shows the read and write I/O rate,
1608 respectively. Finally, the estimated run time of the job is displayed.
1610 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1611 for each thread, each group of threads, and each disk, in that order.
1613 Per-thread statistics first show the threads client number, group-id, and
1614 error code. The remaining figures are as follows:
1618 Number of megabytes of I/O performed.
1621 Average data rate (bandwidth).
1627 Submission latency minimum, maximum, average and standard deviation. This is
1628 the time it took to submit the I/O.
1631 Completion latency minimum, maximum, average and standard deviation. This
1632 is the time between submission and completion.
1635 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1636 and standard deviation.
1639 CPU usage statistics. Includes user and system time, number of context switches
1640 this thread went through and number of major and minor page faults.
1643 Distribution of I/O depths. Each depth includes everything less than (or equal)
1644 to it, but greater than the previous depth.
1647 Number of read/write requests issued, and number of short read/write requests.
1650 Distribution of I/O completion latencies. The numbers follow the same pattern
1654 The group statistics show:
1659 Number of megabytes I/O performed.
1662 Aggregate bandwidth of threads in the group.
1665 Minimum average bandwidth a thread saw.
1668 Maximum average bandwidth a thread saw.
1671 Shortest runtime of threads in the group.
1674 Longest runtime of threads in the group.
1678 Finally, disk statistics are printed with reads first:
1683 Number of I/Os performed by all groups.
1686 Number of merges in the I/O scheduler.
1689 Number of ticks we kept the disk busy.
1692 Total time spent in the disk queue.
1699 It is also possible to get fio to dump the current output while it is
1700 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1703 If the \fB\-\-minimal\fR / \fB\-\-append-terse\fR options are given, the
1704 results will be printed/appended in a semicolon-delimited format suitable for
1706 A job description (if provided) follows on a new line. Note that the first
1707 number in the line is the version number. If the output has to be changed
1708 for some reason, this number will be incremented by 1 to signify that
1709 change. The fields are:
1712 .B terse version, fio version, jobname, groupid, error
1716 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1720 .B min, max, mean, standard deviation
1724 .B min, max, mean, standard deviation
1726 Completion latency percentiles (20 fields):
1728 .B Xth percentile=usec
1732 .B min, max, mean, standard deviation
1736 .B min, max, aggregate percentage of total, mean, standard deviation
1742 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1746 .B min, max, mean, standard deviation
1750 .B min, max, mean, standard deviation
1752 Completion latency percentiles (20 fields):
1754 .B Xth percentile=usec
1758 .B min, max, mean, standard deviation
1762 .B min, max, aggregate percentage of total, mean, standard deviation
1768 .B user, system, context switches, major page faults, minor page faults
1771 IO depth distribution:
1773 .B <=1, 2, 4, 8, 16, 32, >=64
1776 IO latency distribution:
1780 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1784 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1788 Disk utilization (1 for each disk used):
1790 .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
1793 Error Info (dependent on continue_on_error, default off):
1795 .B total # errors, first error code
1798 .B text description (if provided in config - appears on newline)
1801 Normally you would run fio as a stand-alone application on the machine
1802 where the IO workload should be generated. However, it is also possible to
1803 run the frontend and backend of fio separately. This makes it possible to
1804 have a fio server running on the machine(s) where the IO workload should
1805 be running, while controlling it from another machine.
1807 To start the server, you would do:
1809 \fBfio \-\-server=args\fR
1811 on that machine, where args defines what fio listens to. The arguments
1812 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1813 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1814 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1815 listen to (only valid for TCP/IP, not a local socket). Some examples:
1819 Start a fio server, listening on all interfaces on the default port (8765).
1821 2) fio \-\-server=ip:hostname,4444
1823 Start a fio server, listening on IP belonging to hostname and on port 4444.
1825 3) fio \-\-server=ip6:::1,4444
1827 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1829 4) fio \-\-server=,4444
1831 Start a fio server, listening on all interfaces on port 4444.
1833 5) fio \-\-server=1.2.3.4
1835 Start a fio server, listening on IP 1.2.3.4 on the default port.
1837 6) fio \-\-server=sock:/tmp/fio.sock
1839 Start a fio server, listening on the local socket /tmp/fio.sock.
1841 When a server is running, you can connect to it from a client. The client
1844 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1846 where \-\-local-args are arguments that are local to the client where it is
1847 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1848 are sent to the server. The 'server' string follows the same format as it
1849 does on the server side, to allow IP/hostname/socket and port strings.
1850 You can connect to multiple clients as well, to do that you could run:
1852 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1854 If the job file is located on the fio server, then you can tell the server
1855 to load a local file as well. This is done by using \-\-remote-config:
1857 fio \-\-client=server \-\-remote-config /path/to/file.fio
1859 Then the fio serer will open this local (to the server) job file instead
1860 of being passed one from the client.
1864 was written by Jens Axboe <jens.axboe@oracle.com>,
1865 now Jens Axboe <jaxboe@fusionio.com>.
1867 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1868 on documentation by Jens Axboe.
1869 .SH "REPORTING BUGS"
1870 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1873 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1875 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob