1 .TH fio 1 "September 2007" "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 Display version information and exit.
38 .BI \-\-terse\-version \fR=\fPversion
39 Set terse version output format (Current version 3, or older version 2).
42 Display usage information and exit.
44 .BI \-\-cmdhelp \fR=\fPcommand
45 Print help information for \fIcommand\fR. May be `all' for all commands.
47 .BI \-\-enghelp \fR=\fPioengine[,command]
48 List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR.
50 .BI \-\-showcmd \fR=\fPjobfile
51 Convert \fIjobfile\fR to a set of command-line options.
53 .BI \-\-eta \fR=\fPwhen
54 Specifies when real-time ETA estimate should be printed. \fIwhen\fR may
55 be one of `always', `never' or `auto'.
57 .BI \-\-eta\-newline \fR=\fPtime
58 Force an ETA newline for every `time` period passed.
60 .BI \-\-status\-interval \fR=\fPtime
61 Report full output status every `time` period passed.
64 Turn on safety read-only checks, preventing any attempted write.
66 .BI \-\-section \fR=\fPsec
67 Only run section \fIsec\fR from job file. Multiple of these options can be given, adding more sections to run.
69 .BI \-\-alloc\-size \fR=\fPkb
70 Set the internal smalloc pool size to \fIkb\fP kilobytes.
72 .BI \-\-warnings\-fatal
73 All fio parser warnings are fatal, causing fio to exit with an error.
75 .BI \-\-max\-jobs \fR=\fPnr
76 Set the maximum allowed number of jobs (threads/processes) to support.
78 .BI \-\-server \fR=\fPargs
79 Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section.
81 .BI \-\-daemonize \fR=\fPpidfile
82 Background a fio server, writing the pid to the given pid file.
84 .BI \-\-client \fR=\fPhost
85 Instead of running the jobs locally, send and run them on the given host.
87 .BI \-\-idle\-prof \fR=\fPoption
88 Report cpu idleness on a system or percpu basis (\fIoption\fP=system,percpu) or run unit work calibration only (\fIoption\fP=calibrate).
90 Job files are in `ini' format. They consist of one or more
91 job definitions, which begin with a job name in square brackets and
92 extend to the next job name. The job name can be any ASCII string
93 except `global', which has a special meaning. Following the job name is
94 a sequence of zero or more parameters, one per line, that define the
95 behavior of the job. Any line starting with a `;' or `#' character is
96 considered a comment and ignored.
98 If \fIjobfile\fR is specified as `-', the job file will be read from
101 The global section contains default parameters for jobs specified in the
102 job file. A job is only affected by global sections residing above it,
103 and there may be any number of global sections. Specific job definitions
104 may override any parameter set in global sections.
107 Some parameters may take arguments of a specific type. The types used are:
110 String: a sequence of alphanumeric characters.
113 SI integer: a whole number, possibly containing a suffix denoting the base unit
114 of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
115 kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
116 respectively. The suffix is not case sensitive. If prefixed with '0x', the
117 value is assumed to be base 16 (hexadecimal). A suffix may include a trailing 'b',
118 for instance 'kb' is identical to 'k'. You can specify a base 10 value
119 by using 'KiB', 'MiB', 'GiB', etc. This is useful for disk drives where
120 values are often given in base 10 values. Specifying '30GiB' will get you
124 Boolean: a true or false value. `0' denotes false, `1' denotes true.
127 Integer range: a range of integers specified in the format
128 \fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
129 \fIupper\fR may contain a suffix as described above. If an option allows two
130 sets of ranges, they are separated with a `,' or `/' character. For example:
134 List of floating numbers: A list of floating numbers, separated by
139 May be used to override the job name. On the command line, this parameter
140 has the special purpose of signalling the start of a new job.
142 .BI description \fR=\fPstr
143 Human-readable description of the job. It is printed when the job is run, but
144 otherwise has no special purpose.
146 .BI directory \fR=\fPstr
147 Prefix filenames with this directory. Used to place files in a location other
150 .BI filename \fR=\fPstr
152 normally makes up a file name based on the job name, thread number, and file
153 number. If you want to share files between threads in a job or several jobs,
154 specify a \fIfilename\fR for each of them to override the default.
155 If the I/O engine is file-based, you can specify
156 a number of files by separating the names with a `:' character. `\-' is a
157 reserved name, meaning stdin or stdout, depending on the read/write direction
160 .BI filename_format \fR=\fPstr
161 If sharing multiple files between jobs, it is usually necessary to have
162 fio generate the exact names that you want. By default, fio will name a file
163 based on the default file format specification of
164 \fBjobname.jobnumber.filenumber\fP. With this option, that can be
165 customized. Fio will recognize and replace the following keywords in this
171 The name of the worker thread or process.
174 The incremental number of the worker thread or process.
177 The incremental number of the file for that worker thread or process.
180 To have dependent jobs share a set of files, this option can be set to
181 have fio generate filenames that are shared between the two. For instance,
182 if \fBtestfiles.$filenum\fR is specified, file number 4 for any job will
183 be named \fBtestfiles.4\fR. The default of \fB$jobname.$jobnum.$filenum\fR
184 will be used if no other format specifier is given.
188 .BI lockfile \fR=\fPstr
189 Fio defaults to not locking any files before it does IO to them. If a file or
190 file descriptor is shared, fio can serialize IO to that file to make the end
191 result consistent. This is usual for emulating real workloads that share files.
197 No locking. This is the default.
200 Only one thread or process may do IO at the time, excluding all others.
203 Read-write locking on the file. Many readers may access the file at the same
204 time, but writes get exclusive access.
208 .BI opendir \fR=\fPstr
209 Recursively open any files below directory \fIstr\fR.
211 .BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
212 Type of I/O pattern. Accepted values are:
229 Mixed sequential reads and writes.
232 Mixed random reads and writes.
235 For mixed I/O, the default split is 50/50. For certain types of io the result
236 may still be skewed a bit, since the speed may be different. It is possible to
237 specify a number of IO's to do before getting a new offset, this is done by
238 appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
239 would look like \fBrw=randread:8\fR for passing in an offset modifier with a
240 value of 8. If the postfix is used with a sequential IO pattern, then the value
241 specified will be added to the generated offset for each IO. For instance,
242 using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
243 into sequential IO with holes. See the \fBrw_sequencer\fR option.
246 .BI rw_sequencer \fR=\fPstr
247 If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
248 then this option controls how that number modifies the IO offset being
249 generated. Accepted values are:
254 Generate sequential offset
257 Generate the same offset
260 \fBsequential\fR is only useful for random IO, where fio would normally
261 generate a new random offset for every IO. If you append eg 8 to randread, you
262 would get a new random offset for every 8 IO's. The result would be a seek for
263 only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
264 that. As sequential IO is already sequential, setting \fBsequential\fR for that
265 would not result in any differences. \fBidentical\fR behaves in a similar
266 fashion, except it sends the same offset 8 number of times before generating a
271 .BI kb_base \fR=\fPint
272 The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
273 manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
274 reasons. Allow values are 1024 or 1000, with 1024 being the default.
276 .BI unified_rw_reporting \fR=\fPbool
277 Fio normally reports statistics on a per data direction basis, meaning that
278 read, write, and trim are accounted and reported separately. If this option is
279 set, the fio will sum the results and report them as "mixed" instead.
281 .BI randrepeat \fR=\fPbool
282 Seed the random number generator in a predictable way so results are repeatable
283 across runs. Default: true.
285 .BI use_os_rand \fR=\fPbool
286 Fio can either use the random generator supplied by the OS to generator random
287 offsets, or it can use it's own internal generator (based on Tausworthe).
288 Default is to use the internal generator, which is often of better quality and
289 faster. Default: false.
291 .BI fallocate \fR=\fPstr
292 Whether pre-allocation is performed when laying down files. Accepted values
298 Do not pre-allocate space.
301 Pre-allocate via posix_fallocate().
304 Pre-allocate via fallocate() with FALLOC_FL_KEEP_SIZE set.
307 Backward-compatible alias for 'none'.
310 Backward-compatible alias for 'posix'.
313 May not be available on all supported platforms. 'keep' is only
314 available on Linux. If using ZFS on Solaris this must be set to 'none'
315 because ZFS doesn't support it. Default: 'posix'.
318 .BI fadvise_hint \fR=\fPbool
319 Use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
320 are likely to be issued. Default: true.
323 Total size of I/O for this job. \fBfio\fR will run until this many bytes have
324 been transfered, unless limited by other options (\fBruntime\fR, for instance).
325 Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
326 divided between the available files for the job. If not set, fio will use the
327 full size of the given files or devices. If the the files do not exist, size
328 must be given. It is also possible to give size as a percentage between 1 and
329 100. If size=20% is given, fio will use 20% of the full size of the given files
332 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
333 Sets size to something really large and waits for ENOSPC (no space left on
334 device) as the terminating condition. Only makes sense with sequential write.
335 For a read workload, the mount point will be filled first then IO started on
336 the result. This option doesn't make sense if operating on a raw device node,
337 since the size of that is already known by the file system. Additionally,
338 writing beyond end-of-device will not return ENOSPC there.
340 .BI filesize \fR=\fPirange
341 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
342 for files at random within the given range, limited to \fBsize\fR in total (if
343 that is given). If \fBfilesize\fR is not specified, each created file is the
346 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
347 Block size for I/O units. Default: 4k. Values for reads and writes can be
348 specified separately in the format \fIread\fR,\fIwrite\fR, either of
349 which may be empty to leave that value at its default.
351 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
352 Specify a range of I/O block sizes. The issued I/O unit will always be a
353 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
354 to both reads and writes if only one range is given, but can be specified
355 separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
356 Also (see \fBblocksize\fR).
358 .BI bssplit \fR=\fPstr
359 This option allows even finer grained control of the block sizes issued,
360 not just even splits between them. With this option, you can weight various
361 block sizes for exact control of the issued IO for a job that has mixed
362 block sizes. The format of the option is bssplit=blocksize/percentage,
363 optionally adding as many definitions as needed separated by a colon.
364 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
365 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
366 splits to reads and writes. The format is identical to what the
367 \fBbs\fR option accepts, the read and write parts are separated with a
370 .B blocksize_unaligned\fR,\fP bs_unaligned
371 If set, any size in \fBblocksize_range\fR may be used. This typically won't
372 work with direct I/O, as that normally requires sector alignment.
374 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
375 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
376 the minimum blocksize given. Minimum alignment is typically 512b
377 for using direct IO, though it usually depends on the hardware block size.
378 This option is mutually exclusive with using a random map for files, so it
379 will turn off that option.
382 Initialise buffers with all zeros. Default: fill buffers with random data.
385 If this option is given, fio will refill the IO buffers on every submit. The
386 default is to only fill it at init time and reuse that data. Only makes sense
387 if zero_buffers isn't specified, naturally. If data verification is enabled,
388 refill_buffers is also automatically enabled.
390 .BI scramble_buffers \fR=\fPbool
391 If \fBrefill_buffers\fR is too costly and the target is using data
392 deduplication, then setting this option will slightly modify the IO buffer
393 contents to defeat normal de-dupe attempts. This is not enough to defeat
394 more clever block compression attempts, but it will stop naive dedupe
395 of blocks. Default: true.
397 .BI buffer_compress_percentage \fR=\fPint
398 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
399 that compress to the specified level. Fio does this by providing a mix of
400 random data and zeroes. Note that this is per block size unit, for file/disk
401 wide compression level that matches this setting, you'll also want to set
402 \fBrefill_buffers\fR.
404 .BI buffer_compress_chunk \fR=\fPint
405 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
406 big the ranges of random data and zeroed data is. Without this set, fio will
407 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
408 the remaining zeroed. With this set to some chunk size smaller than the block
409 size, fio can alternate random and zeroed data throughout the IO buffer.
411 .BI nrfiles \fR=\fPint
412 Number of files to use for this job. Default: 1.
414 .BI openfiles \fR=\fPint
415 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
417 .BI file_service_type \fR=\fPstr
418 Defines how files to service are selected. The following types are defined:
423 Choose a file at random
426 Round robin over open files (default).
428 Do each file in the set sequentially.
431 The number of I/Os to issue before switching a new file can be specified by
432 appending `:\fIint\fR' to the service type.
435 .BI ioengine \fR=\fPstr
436 Defines how the job issues I/O. The following types are defined:
441 Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
442 position the I/O location.
445 Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
448 Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
449 coalescing adjacents IOs into a single submission.
452 Basic \fIpreadv\fR\|(2) or \fIpwritev\fR\|(2) I/O.
455 Linux native asynchronous I/O. This ioengine defines engine specific options.
458 POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
461 Solaris native asynchronous I/O.
464 Windows native asynchronous I/O.
467 File is memory mapped with \fImmap\fR\|(2) and data copied using
471 \fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
472 transfer data from user-space to the kernel.
475 Use the syslet system calls to make regular read/write asynchronous.
478 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
479 the target is an sg character device, we use \fIread\fR\|(2) and
480 \fIwrite\fR\|(2) for asynchronous I/O.
483 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
484 itself and for debugging and testing purposes.
487 Transfer over the network. The protocol to be used can be defined with the
488 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
489 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
490 This ioengine defines engine specific options.
493 Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
494 and send/receive. This ioengine defines engine specific options.
497 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
498 \fBcpucycles\fR parameters.
501 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
502 approach to asycnronous I/O.
504 See <http://www.xmailserver.org/guasi\-lib.html>.
507 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
508 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
511 Loads an external I/O engine object file. Append the engine filename as
515 IO engine that does regular linux native fallocate callt to simulate data
516 transfer as fio ioengine
518 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
520 DIR_WRITE does fallocate(,mode = 0)
522 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
525 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
526 request to DDIR_WRITE event
531 .BI iodepth \fR=\fPint
532 Number of I/O units to keep in flight against the file. Note that increasing
533 iodepth beyond 1 will not affect synchronous ioengines (except for small
534 degress when verify_async is in use). Even async engines my impose OS
535 restrictions causing the desired depth not to be achieved. This may happen on
536 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
537 not async on that OS. Keep an eye on the IO depth distribution in the
538 fio output to verify that the achieved depth is as expected. Default: 1.
540 .BI iodepth_batch \fR=\fPint
541 Number of I/Os to submit at once. Default: \fBiodepth\fR.
543 .BI iodepth_batch_complete \fR=\fPint
544 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
545 means that we'll ask for a minimum of 1 IO in the retrieval process from the
546 kernel. The IO retrieval will go on until we hit the limit set by
547 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
548 completed events before queuing more IO. This helps reduce IO latency, at the
549 cost of more retrieval system calls.
551 .BI iodepth_low \fR=\fPint
552 Low watermark indicating when to start filling the queue again. Default:
555 .BI direct \fR=\fPbool
556 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
558 .BI buffered \fR=\fPbool
559 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
562 .BI offset \fR=\fPint
563 Offset in the file to start I/O. Data before the offset will not be touched.
565 .BI offset_increment \fR=\fPint
566 If this is provided, then the real offset becomes the
567 offset + offset_increment * thread_number, where the thread number is a counter
568 that starts at 0 and is incremented for each job. This option is useful if
569 there are several jobs which are intended to operate on a file in parallel in
570 disjoint segments, with even spacing between the starting points.
573 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
574 0, don't sync. Default: 0.
576 .BI fdatasync \fR=\fPint
577 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
578 data parts of the file. Default: 0.
580 .BI sync_file_range \fR=\fPstr:int
581 Use sync_file_range() for every \fRval\fP number of write operations. Fio will
582 track range of writes that have happened since the last sync_file_range() call.
583 \fRstr\fP can currently be one or more of:
587 SYNC_FILE_RANGE_WAIT_BEFORE
590 SYNC_FILE_RANGE_WRITE
593 SYNC_FILE_RANGE_WRITE
597 So if you do sync_file_range=wait_before,write:8, fio would use
598 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
599 Also see the sync_file_range(2) man page. This option is Linux specific.
601 .BI overwrite \fR=\fPbool
602 If writing, setup the file first and do overwrites. Default: false.
604 .BI end_fsync \fR=\fPbool
605 Sync file contents when a write stage has completed. Default: false.
607 .BI fsync_on_close \fR=\fPbool
608 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
609 it will happen on every close, not just at the end of the job. Default: false.
611 .BI rwmixread \fR=\fPint
612 Percentage of a mixed workload that should be reads. Default: 50.
614 .BI rwmixwrite \fR=\fPint
615 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
616 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
617 overrides the first. This may interfere with a given rate setting, if fio is
618 asked to limit reads or writes to a certain rate. If that is the case, then
619 the distribution may be skewed. Default: 50.
621 .BI random_distribution \fR=\fPstr:float
622 By default, fio will use a completely uniform random distribution when asked
623 to perform random IO. Sometimes it is useful to skew the distribution in
624 specific ways, ensuring that some parts of the data is more hot than others.
625 Fio includes the following distribution models:
629 Uniform random distribution
639 When using a zipf or pareto distribution, an input value is also needed to
640 define the access pattern. For zipf, this is the zipf theta. For pareto,
641 it's the pareto power. Fio includes a test program, genzipf, that can be
642 used visualize what the given input values will yield in terms of hit rates.
643 If you wanted to use zipf with a theta of 1.2, you would use
644 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
645 fio will disable use of the random map.
647 .BI percentage_random \fR=\fPint
648 For a random workload, set how big a percentage should be random. This defaults
649 to 100%, in which case the workload is fully random. It can be set from
650 anywhere from 0 to 100. Setting it to 0 would make the workload fully
653 .BI percentage_sequential \fR=\fPint
654 See \fBpercentage_random\fR.
657 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
658 this parameter is given, a new offset will be chosen without looking at past
659 I/O history. This parameter is mutually exclusive with \fBverify\fR.
661 .BI softrandommap \fR=\fPbool
662 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
663 fails to allocate the map, if this option is set it will continue without a
664 random block map. As coverage will not be as complete as with random maps, this
665 option is disabled by default.
667 .BI random_generator \fR=\fPstr
668 Fio supports the following engines for generating IO offsets for random IO:
672 Strong 2^88 cycle random number generator
675 Linear feedback shift register generator
679 Tausworthe is a strong random number generator, but it requires tracking on the
680 side if we want to ensure that blocks are only read or written once. LFSR
681 guarantees that we never generate the same offset twice, and it's also less
682 computationally expensive. It's not a true random generator, however, though
683 for IO purposes it's typically good enough. LFSR only works with single block
684 sizes, not with workloads that use multiple block sizes. If used with such a
685 workload, fio may read or write some blocks multiple times.
688 Run job with given nice value. See \fInice\fR\|(2).
691 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
694 .BI prioclass \fR=\fPint
695 Set I/O priority class. See \fIionice\fR\|(1).
697 .BI thinktime \fR=\fPint
698 Stall job for given number of microseconds between issuing I/Os.
700 .BI thinktime_spin \fR=\fPint
701 Pretend to spend CPU time for given number of microseconds, sleeping the rest
702 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
704 .BI thinktime_blocks \fR=\fPint
705 Only valid if thinktime is set - control how many blocks to issue, before
706 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
707 make fio wait \fBthinktime\fR microseconds after every block. This
708 effectively makes any queue depth setting redundant, since no more than 1 IO
709 will be queued before we have to complete it and do our thinktime. In other
710 words, this setting effectively caps the queue depth if the latter is larger.
714 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
715 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
716 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
717 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
718 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
719 limit writes (to 500KB/sec), the latter will only limit reads.
721 .BI ratemin \fR=\fPint
722 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
723 Failing to meet this requirement will cause the job to exit. The same format
724 as \fBrate\fR is used for read vs write separation.
726 .BI rate_iops \fR=\fPint
727 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
728 specified independently of bandwidth. The same format as \fBrate\fR is used for
729 read vs write seperation. If \fBblocksize\fR is a range, the smallest block
730 size is used as the metric.
732 .BI rate_iops_min \fR=\fPint
733 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
734 is used for read vs write seperation.
736 .BI ratecycle \fR=\fPint
737 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
738 milliseconds. Default: 1000ms.
740 .BI max_latency \fR=\fPint
741 If set, fio will exit the job if it exceeds this maximum latency. It will exit
744 .BI cpumask \fR=\fPint
745 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
746 may run on. See \fBsched_setaffinity\fR\|(2).
748 .BI cpus_allowed \fR=\fPstr
749 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
751 .BI numa_cpu_nodes \fR=\fPstr
752 Set this job running on spcified NUMA nodes' CPUs. The arguments allow
753 comma delimited list of cpu numbers, A-B ranges, or 'all'.
755 .BI numa_mem_policy \fR=\fPstr
756 Set this job's memory policy and corresponding NUMA nodes. Format of
760 .B <mode>[:<nodelist>]
763 is one of the following memory policy:
765 .B default, prefer, bind, interleave, local
768 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
769 needed to be specified. For \fBprefer\fR, only one node is
770 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
771 comma delimited list of numbers, A-B ranges, or 'all'.
773 .BI startdelay \fR=\fPint
774 Delay start of job for the specified number of seconds.
776 .BI runtime \fR=\fPint
777 Terminate processing after the specified number of seconds.
780 If given, run for the specified \fBruntime\fR duration even if the files are
781 completely read or written. The same workload will be repeated as many times
782 as \fBruntime\fR allows.
784 .BI ramp_time \fR=\fPint
785 If set, fio will run the specified workload for this amount of time before
786 logging any performance numbers. Useful for letting performance settle before
787 logging results, thus minimizing the runtime required for stable results. Note
788 that the \fBramp_time\fR is considered lead in time for a job, thus it will
789 increase the total runtime if a special timeout or runtime is specified.
791 .BI invalidate \fR=\fPbool
792 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
795 Use synchronous I/O for buffered writes. For the majority of I/O engines,
796 this means using O_SYNC. Default: false.
798 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
799 Allocation method for I/O unit buffer. Allowed values are:
804 Allocate memory with \fImalloc\fR\|(3).
807 Use shared memory buffers allocated through \fIshmget\fR\|(2).
810 Same as \fBshm\fR, but use huge pages as backing.
813 Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
814 is given after the option in the format `:\fIfile\fR'.
817 Same as \fBmmap\fR, but use huge files as backing.
820 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
821 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
822 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
823 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
824 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
825 and the documentation for that. Normally you just need to echo an appropriate
826 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
830 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
831 This indiciates the memory alignment of the IO memory buffers. Note that the
832 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
833 the alignment of the following buffers are given by the \fBbs\fR used. In
834 other words, if using a \fBbs\fR that is a multiple of the page sized in the
835 system, all buffers will be aligned to this value. If using a \fBbs\fR that
836 is not page aligned, the alignment of subsequent IO memory buffers is the
837 sum of the \fBiomem_align\fR and \fBbs\fR used.
839 .BI hugepage\-size \fR=\fPint
840 Defines the size of a huge page. Must be at least equal to the system setting.
841 Should be a multiple of 1MB. Default: 4MB.
844 Terminate all jobs when one finishes. Default: wait for each job to finish.
846 .BI bwavgtime \fR=\fPint
847 Average bandwidth calculations over the given time in milliseconds. Default:
850 .BI iopsavgtime \fR=\fPint
851 Average IOPS calculations over the given time in milliseconds. Default:
854 .BI create_serialize \fR=\fPbool
855 If true, serialize file creation for the jobs. Default: true.
857 .BI create_fsync \fR=\fPbool
858 \fIfsync\fR\|(2) data file after creation. Default: true.
860 .BI create_on_open \fR=\fPbool
861 If true, the files are not created until they are opened for IO by the job.
863 .BI create_only \fR=\fPbool
864 If true, fio will only run the setup phase of the job. If files need to be
865 laid out or updated on disk, only that will be done. The actual job contents
868 .BI pre_read \fR=\fPbool
869 If this is given, files will be pre-read into memory before starting the given
870 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
871 pointless to pre-read and then drop the cache. This will only work for IO
872 engines that are seekable, since they allow you to read the same data
873 multiple times. Thus it will not work on eg network or splice IO.
875 .BI unlink \fR=\fPbool
876 Unlink job files when done. Default: false.
879 Specifies the number of iterations (runs of the same workload) of this job.
882 .BI do_verify \fR=\fPbool
883 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
886 .BI verify \fR=\fPstr
887 Method of verifying file contents after each iteration of the job. Allowed
892 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
893 Store appropriate checksum in the header of each block. crc32c-intel is
894 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
895 not supported by the system.
898 Write extra information about each I/O (timestamp, block number, etc.). The
899 block number is verified. See \fBverify_pattern\fR as well.
902 Pretend to verify. Used for testing internals.
905 This option can be used for repeated burn-in tests of a system to make sure
906 that the written data is also correctly read back. If the data direction given
907 is a read or random read, fio will assume that it should verify a previously
908 written file. If the data direction includes any form of write, the verify will
909 be of the newly written data.
912 .BI verify_sort \fR=\fPbool
913 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
914 read them back in a sorted manner. Default: true.
916 .BI verify_offset \fR=\fPint
917 Swap the verification header with data somewhere else in the block before
918 writing. It is swapped back before verifying.
920 .BI verify_interval \fR=\fPint
921 Write the verification header for this number of bytes, which should divide
922 \fBblocksize\fR. Default: \fBblocksize\fR.
924 .BI verify_pattern \fR=\fPstr
925 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
926 with totally random bytes, but sometimes it's interesting to fill with a known
927 pattern for io verification purposes. Depending on the width of the pattern,
928 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
929 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
930 has to be a hex number that starts with either "0x" or "0X". Use with
933 .BI verify_fatal \fR=\fPbool
934 If true, exit the job on the first observed verification failure. Default:
937 .BI verify_dump \fR=\fPbool
938 If set, dump the contents of both the original data block and the data block we
939 read off disk to files. This allows later analysis to inspect just what kind of
940 data corruption occurred. Off by default.
942 .BI verify_async \fR=\fPint
943 Fio will normally verify IO inline from the submitting thread. This option
944 takes an integer describing how many async offload threads to create for IO
945 verification instead, causing fio to offload the duty of verifying IO contents
946 to one or more separate threads. If using this offload option, even sync IO
947 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
948 allows them to have IO in flight while verifies are running.
950 .BI verify_async_cpus \fR=\fPstr
951 Tell fio to set the given CPU affinity on the async IO verification threads.
952 See \fBcpus_allowed\fP for the format used.
954 .BI verify_backlog \fR=\fPint
955 Fio will normally verify the written contents of a job that utilizes verify
956 once that job has completed. In other words, everything is written then
957 everything is read back and verified. You may want to verify continually
958 instead for a variety of reasons. Fio stores the meta data associated with an
959 IO block in memory, so for large verify workloads, quite a bit of memory would
960 be used up holding this meta data. If this option is enabled, fio will write
961 only N blocks before verifying these blocks.
963 .BI verify_backlog_batch \fR=\fPint
964 Control how many blocks fio will verify if verify_backlog is set. If not set,
965 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
966 read back and verified). If \fBverify_backlog_batch\fR is less than
967 \fBverify_backlog\fR then not all blocks will be verified, if
968 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
969 will be verified more than once.
971 .B stonewall "\fR,\fP wait_for_previous"
972 Wait for preceding jobs in the job file to exit before starting this one.
973 \fBstonewall\fR implies \fBnew_group\fR.
976 Start a new reporting group. If not given, all jobs in a file will be part
977 of the same reporting group, unless separated by a stonewall.
979 .BI numjobs \fR=\fPint
980 Number of clones (processes/threads performing the same workload) of this job.
984 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
988 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
989 with \fBfork\fR\|(2).
991 .BI zonesize \fR=\fPint
992 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
994 .BI zoneskip \fR=\fPint
995 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
998 .BI write_iolog \fR=\fPstr
999 Write the issued I/O patterns to the specified file. Specify a separate file
1000 for each job, otherwise the iologs will be interspersed and the file may be
1003 .BI read_iolog \fR=\fPstr
1004 Replay the I/O patterns contained in the specified file generated by
1005 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1007 .BI replay_no_stall \fR=\fPint
1008 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1009 attempts to respect timing information between I/Os. Enabling
1010 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1011 still respecting ordering.
1013 .BI replay_redirect \fR=\fPstr
1014 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1015 is to replay the IOPS onto the major/minor device that each IOP was recorded
1016 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1017 single specified device regardless of the device it was recorded from.
1019 .BI write_bw_log \fR=\fPstr
1020 If given, write a bandwidth log of the jobs in this job file. Can be used to
1021 store data of the bandwidth of the jobs in their lifetime. The included
1022 fio_generate_plots script uses gnuplot to turn these text files into nice
1023 graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
1024 option, the postfix is _bw.log.
1026 .BI write_lat_log \fR=\fPstr
1027 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1028 filename is given with this option, the default filename of "jobname_type.log"
1029 is used. Even if the filename is given, fio will still append the type of log.
1031 .BI write_iops_log \fR=\fPstr
1032 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1033 option, the default filename of "jobname_type.log" is used. Even if the
1034 filename is given, fio will still append the type of log.
1036 .BI log_avg_msec \fR=\fPint
1037 By default, fio will log an entry in the iops, latency, or bw log for every
1038 IO that completes. When writing to the disk log, that can quickly grow to a
1039 very large size. Setting this option makes fio average the each log entry
1040 over the specified period of time, reducing the resolution of the log.
1043 .BI disable_lat \fR=\fPbool
1044 Disable measurements of total latency numbers. Useful only for cutting
1045 back the number of calls to gettimeofday, as that does impact performance at
1046 really high IOPS rates. Note that to really get rid of a large amount of these
1047 calls, this option must be used with disable_slat and disable_bw as well.
1049 .BI disable_clat \fR=\fPbool
1050 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1052 .BI disable_slat \fR=\fPbool
1053 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1055 .BI disable_bw_measurement \fR=\fPbool
1056 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1058 .BI lockmem \fR=\fPint
1059 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1060 simulate a smaller amount of memory. The amount specified is per worker.
1062 .BI exec_prerun \fR=\fPstr
1063 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1065 Output is redirected in a file called \fBjobname.prerun.txt\fR
1068 .BI exec_postrun \fR=\fPstr
1069 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1071 Output is redirected in a file called \fBjobname.postrun.txt\fR
1074 .BI ioscheduler \fR=\fPstr
1075 Attempt to switch the device hosting the file to the specified I/O scheduler.
1077 .BI cpuload \fR=\fPint
1078 If the job is a CPU cycle-eater, attempt to use the specified percentage of
1081 .BI cpuchunks \fR=\fPint
1082 If the job is a CPU cycle-eater, split the load into cycles of the
1083 given time in milliseconds.
1085 .BI disk_util \fR=\fPbool
1086 Generate disk utilization statistics if the platform supports it. Default: true.
1088 .BI clocksource \fR=\fPstr
1089 Use the given clocksource as the base of timing. The supported options are:
1099 Internal CPU clock source
1103 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1104 (and fio is heavy on time calls). Fio will automatically use this clocksource
1105 if it's supported and considered reliable on the system it is running on,
1106 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1107 means supporting TSC Invariant.
1109 .BI gtod_reduce \fR=\fPbool
1110 Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
1111 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1112 gettimeofday() call count. With this option enabled, we only do about 0.4% of
1113 the gtod() calls we would have done if all time keeping was enabled.
1115 .BI gtod_cpu \fR=\fPint
1116 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1117 the current time. Fio (and databases, for instance) are very intensive on
1118 gettimeofday() calls. With this option, you can set one CPU aside for doing
1119 nothing but logging current time to a shared memory location. Then the other
1120 threads/processes that run IO workloads need only copy that segment, instead of
1121 entering the kernel with a gettimeofday() call. The CPU set aside for doing
1122 these time calls will be excluded from other uses. Fio will manually clear it
1123 from the CPU mask of other jobs.
1125 .BI ignore_error \fR=\fPstr
1126 Sometimes you want to ignore some errors during test in that case you can specify
1127 error list for each error type.
1129 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1131 errors for given error type is separated with ':'.
1132 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1134 Example: ignore_error=EAGAIN,ENOSPC:122 .
1136 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1138 .BI error_dump \fR=\fPbool
1139 If set dump every error even if it is non fatal, true by default. If disabled
1140 only fatal error will be dumped
1142 .BI cgroup \fR=\fPstr
1143 Add job to this control group. If it doesn't exist, it will be created.
1144 The system must have a mounted cgroup blkio mount point for this to work. If
1145 your system doesn't have it mounted, you can do so with:
1147 # mount \-t cgroup \-o blkio none /cgroup
1149 .BI cgroup_weight \fR=\fPint
1150 Set the weight of the cgroup to this value. See the documentation that comes
1151 with the kernel, allowed values are in the range of 100..1000.
1153 .BI cgroup_nodelete \fR=\fPbool
1154 Normally fio will delete the cgroups it has created after the job completion.
1155 To override this behavior and to leave cgroups around after the job completion,
1156 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1157 cgroup files after job completion. Default: false
1160 Instead of running as the invoking user, set the user ID to this value before
1161 the thread/process does any work.
1164 Set group ID, see \fBuid\fR.
1166 .BI flow_id \fR=\fPint
1167 The ID of the flow. If not specified, it defaults to being a global flow. See
1171 Weight in token-based flow control. If this value is used, then there is a
1172 \fBflow counter\fR which is used to regulate the proportion of activity between
1173 two or more jobs. fio attempts to keep this flow counter near zero. The
1174 \fBflow\fR parameter stands for how much should be added or subtracted to the
1175 flow counter on each iteration of the main I/O loop. That is, if one job has
1176 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1177 1:8 ratio in how much one runs vs the other.
1179 .BI flow_watermark \fR=\fPint
1180 The maximum value that the absolute value of the flow counter is allowed to
1181 reach before the job must wait for a lower value of the counter.
1183 .BI flow_sleep \fR=\fPint
1184 The period of time, in microseconds, to wait after the flow watermark has been
1185 exceeded before retrying operations
1187 .BI clat_percentiles \fR=\fPbool
1188 Enable the reporting of percentiles of completion latencies.
1190 .BI percentile_list \fR=\fPfloat_list
1191 Overwrite the default list of percentiles for completion
1192 latencies. Each number is a floating number in the range (0,100], and
1193 the maximum length of the list is 20. Use ':' to separate the
1194 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1195 report the values of completion latency below which 99.5% and 99.9% of
1196 the observed latencies fell, respectively.
1197 .SS "Ioengine Parameters List"
1198 Some parameters are only valid when a specific ioengine is in use. These are
1199 used identically to normal parameters, with the caveat that when used on the
1200 command line, the must come after the ioengine that defines them is selected.
1202 .BI (cpu)cpuload \fR=\fPint
1203 Attempt to use the specified percentage of CPU cycles.
1205 .BI (cpu)cpuchunks \fR=\fPint
1206 Split the load into cycles of the given time. In microseconds.
1208 .BI (libaio)userspace_reap
1209 Normally, with the libaio engine in use, fio will use
1210 the io_getevents system call to reap newly returned events.
1211 With this flag turned on, the AIO ring will be read directly
1212 from user-space to reap events. The reaping mode is only
1213 enabled when polling for a minimum of 0 events (eg when
1214 iodepth_batch_complete=0).
1216 .BI (net,netsplice)hostname \fR=\fPstr
1217 The host name or IP address to use for TCP or UDP based IO.
1218 If the job is a TCP listener or UDP reader, the hostname is not
1219 used and must be omitted.
1221 .BI (net,netsplice)port \fR=\fPint
1222 The TCP or UDP port to bind to or connect to.
1224 .BI (net,netsplice)nodelay \fR=\fPbool
1225 Set TCP_NODELAY on TCP connections.
1227 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1228 The network protocol to use. Accepted values are:
1233 Transmission control protocol
1236 User datagram protocol
1242 When the protocol is TCP or UDP, the port must also be given,
1243 as well as the hostname if the job is a TCP listener or UDP
1244 reader. For unix sockets, the normal filename option should be
1245 used and the port is invalid.
1248 .BI (net,netsplice)listen
1249 For TCP network connections, tell fio to listen for incoming
1250 connections rather than initiating an outgoing connection. The
1251 hostname must be omitted if this option is used.
1253 .BI (net, pingpong) \fR=\fPbool
1254 Normal a network writer will just continue writing data, and a network reader
1255 will just consume packages. If pingpong=1 is set, a writer will send its normal
1256 payload to the reader, then wait for the reader to send the same payload back.
1257 This allows fio to measure network latencies. The submission and completion
1258 latencies then measure local time spent sending or receiving, and the
1259 completion latency measures how long it took for the other end to receive and
1262 .BI (e4defrag,donorname) \fR=\fPstr
1263 File will be used as a block donor (swap extents between files)
1265 .BI (e4defrag,inplace) \fR=\fPint
1266 Configure donor file block allocation strategy
1269 Preallocate donor's file on init
1272 allocate space immidietly inside defragment event, and free right after event
1276 While running, \fBfio\fR will display the status of the created jobs. For
1280 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1283 The characters in the first set of brackets denote the current status of each
1284 threads. The possible values are:
1290 Setup but not started.
1296 Initialized, waiting.
1299 Running, doing sequential reads.
1302 Running, doing random reads.
1305 Running, doing sequential writes.
1308 Running, doing random writes.
1311 Running, doing mixed sequential reads/writes.
1314 Running, doing mixed random reads/writes.
1317 Running, currently waiting for \fBfsync\fR\|(2).
1320 Running, verifying written data.
1323 Exited, not reaped by main thread.
1326 Exited, thread reaped.
1330 The second set of brackets shows the estimated completion percentage of
1331 the current group. The third set shows the read and write I/O rate,
1332 respectively. Finally, the estimated run time of the job is displayed.
1334 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1335 for each thread, each group of threads, and each disk, in that order.
1337 Per-thread statistics first show the threads client number, group-id, and
1338 error code. The remaining figures are as follows:
1342 Number of megabytes of I/O performed.
1345 Average data rate (bandwidth).
1351 Submission latency minimum, maximum, average and standard deviation. This is
1352 the time it took to submit the I/O.
1355 Completion latency minimum, maximum, average and standard deviation. This
1356 is the time between submission and completion.
1359 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1360 and standard deviation.
1363 CPU usage statistics. Includes user and system time, number of context switches
1364 this thread went through and number of major and minor page faults.
1367 Distribution of I/O depths. Each depth includes everything less than (or equal)
1368 to it, but greater than the previous depth.
1371 Number of read/write requests issued, and number of short read/write requests.
1374 Distribution of I/O completion latencies. The numbers follow the same pattern
1378 The group statistics show:
1383 Number of megabytes I/O performed.
1386 Aggregate bandwidth of threads in the group.
1389 Minimum average bandwidth a thread saw.
1392 Maximum average bandwidth a thread saw.
1395 Shortest runtime of threads in the group.
1398 Longest runtime of threads in the group.
1402 Finally, disk statistics are printed with reads first:
1407 Number of I/Os performed by all groups.
1410 Number of merges in the I/O scheduler.
1413 Number of ticks we kept the disk busy.
1416 Total time spent in the disk queue.
1423 It is also possible to get fio to dump the current output while it is
1424 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1427 If the \fB\-\-minimal\fR option is given, the results will be printed in a
1428 semicolon-delimited format suitable for scripted use - a job description
1429 (if provided) follows on a new line. Note that the first
1430 number in the line is the version number. If the output has to be changed
1431 for some reason, this number will be incremented by 1 to signify that
1432 change. The fields are:
1435 .B terse version, fio version, jobname, groupid, error
1439 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1443 .B min, max, mean, standard deviation
1447 .B min, max, mean, standard deviation
1449 Completion latency percentiles (20 fields):
1451 .B Xth percentile=usec
1455 .B min, max, mean, standard deviation
1459 .B min, max, aggregate percentage of total, mean, standard deviation
1465 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1469 .B min, max, mean, standard deviation
1473 .B min, max, mean, standard deviation
1475 Completion latency percentiles (20 fields):
1477 .B Xth percentile=usec
1481 .B min, max, mean, standard deviation
1485 .B min, max, aggregate percentage of total, mean, standard deviation
1491 .B user, system, context switches, major page faults, minor page faults
1494 IO depth distribution:
1496 .B <=1, 2, 4, 8, 16, 32, >=64
1499 IO latency distribution:
1503 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1507 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1511 Disk utilization (1 for each disk used):
1513 .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
1516 Error Info (dependent on continue_on_error, default off):
1518 .B total # errors, first error code
1521 .B text description (if provided in config - appears on newline)
1524 Normally you would run fio as a stand-alone application on the machine
1525 where the IO workload should be generated. However, it is also possible to
1526 run the frontend and backend of fio separately. This makes it possible to
1527 have a fio server running on the machine(s) where the IO workload should
1528 be running, while controlling it from another machine.
1530 To start the server, you would do:
1532 \fBfio \-\-server=args\fR
1534 on that machine, where args defines what fio listens to. The arguments
1535 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1536 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1537 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1538 listen to (only valid for TCP/IP, not a local socket). Some examples:
1542 Start a fio server, listening on all interfaces on the default port (8765).
1544 2) fio \-\-server=ip:hostname,4444
1546 Start a fio server, listening on IP belonging to hostname and on port 4444.
1548 3) fio \-\-server=ip6:::1,4444
1550 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1552 4) fio \-\-server=,4444
1554 Start a fio server, listening on all interfaces on port 4444.
1556 5) fio \-\-server=1.2.3.4
1558 Start a fio server, listening on IP 1.2.3.4 on the default port.
1560 6) fio \-\-server=sock:/tmp/fio.sock
1562 Start a fio server, listening on the local socket /tmp/fio.sock.
1564 When a server is running, you can connect to it from a client. The client
1567 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1569 where \-\-local-args are arguments that are local to the client where it is
1570 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1571 are sent to the server. The 'server' string follows the same format as it
1572 does on the server side, to allow IP/hostname/socket and port strings.
1573 You can connect to multiple clients as well, to do that you could run:
1575 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1579 was written by Jens Axboe <jens.axboe@oracle.com>,
1580 now Jens Axboe <jaxboe@fusionio.com>.
1582 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1583 on documentation by Jens Axboe.
1584 .SH "REPORTING BUGS"
1585 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1588 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1590 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob