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 transferred, 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, writes, and trims
348 can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR
349 either of which may be empty to leave that value at its default. If a trailing
350 comma isn't given, the remainder will inherit the last value set.
352 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
353 Specify a range of I/O block sizes. The issued I/O unit will always be a
354 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
355 to both reads and writes if only one range is given, but can be specified
356 separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
357 Also (see \fBblocksize\fR).
359 .BI bssplit \fR=\fPstr
360 This option allows even finer grained control of the block sizes issued,
361 not just even splits between them. With this option, you can weight various
362 block sizes for exact control of the issued IO for a job that has mixed
363 block sizes. The format of the option is bssplit=blocksize/percentage,
364 optionally adding as many definitions as needed separated by a colon.
365 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
366 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
367 splits to reads and writes. The format is identical to what the
368 \fBbs\fR option accepts, the read and write parts are separated with a
371 .B blocksize_unaligned\fR,\fP bs_unaligned
372 If set, any size in \fBblocksize_range\fR may be used. This typically won't
373 work with direct I/O, as that normally requires sector alignment.
375 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
376 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
377 the minimum blocksize given. Minimum alignment is typically 512b
378 for using direct IO, though it usually depends on the hardware block size.
379 This option is mutually exclusive with using a random map for files, so it
380 will turn off that option.
382 .BI bs_is_seq_rand \fR=\fPbool
383 If this option is set, fio will use the normal read,write blocksize settings as
384 sequential,random instead. Any random read or write will use the WRITE
385 blocksize settings, and any sequential read or write will use the READ
389 Initialise buffers with all zeros. Default: fill buffers with random data.
392 If this option is given, fio will refill the IO buffers on every submit. The
393 default is to only fill it at init time and reuse that data. Only makes sense
394 if zero_buffers isn't specified, naturally. If data verification is enabled,
395 refill_buffers is also automatically enabled.
397 .BI scramble_buffers \fR=\fPbool
398 If \fBrefill_buffers\fR is too costly and the target is using data
399 deduplication, then setting this option will slightly modify the IO buffer
400 contents to defeat normal de-dupe attempts. This is not enough to defeat
401 more clever block compression attempts, but it will stop naive dedupe
402 of blocks. Default: true.
404 .BI buffer_compress_percentage \fR=\fPint
405 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
406 that compress to the specified level. Fio does this by providing a mix of
407 random data and zeroes. Note that this is per block size unit, for file/disk
408 wide compression level that matches this setting, you'll also want to set
409 \fBrefill_buffers\fR.
411 .BI buffer_compress_chunk \fR=\fPint
412 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
413 big the ranges of random data and zeroed data is. Without this set, fio will
414 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
415 the remaining zeroed. With this set to some chunk size smaller than the block
416 size, fio can alternate random and zeroed data throughout the IO buffer.
418 .BI nrfiles \fR=\fPint
419 Number of files to use for this job. Default: 1.
421 .BI openfiles \fR=\fPint
422 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
424 .BI file_service_type \fR=\fPstr
425 Defines how files to service are selected. The following types are defined:
430 Choose a file at random
433 Round robin over open files (default).
435 Do each file in the set sequentially.
438 The number of I/Os to issue before switching a new file can be specified by
439 appending `:\fIint\fR' to the service type.
442 .BI ioengine \fR=\fPstr
443 Defines how the job issues I/O. The following types are defined:
448 Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
449 position the I/O location.
452 Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
455 Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
456 coalescing adjacents IOs into a single submission.
459 Basic \fIpreadv\fR\|(2) or \fIpwritev\fR\|(2) I/O.
462 Linux native asynchronous I/O. This ioengine defines engine specific options.
465 POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
468 Solaris native asynchronous I/O.
471 Windows native asynchronous I/O.
474 File is memory mapped with \fImmap\fR\|(2) and data copied using
478 \fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
479 transfer data from user-space to the kernel.
482 Use the syslet system calls to make regular read/write asynchronous.
485 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
486 the target is an sg character device, we use \fIread\fR\|(2) and
487 \fIwrite\fR\|(2) for asynchronous I/O.
490 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
491 itself and for debugging and testing purposes.
494 Transfer over the network. The protocol to be used can be defined with the
495 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
496 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
497 This ioengine defines engine specific options.
500 Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
501 and send/receive. This ioengine defines engine specific options.
504 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
505 \fBcpucycles\fR parameters.
508 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
509 approach to asycnronous I/O.
511 See <http://www.xmailserver.org/guasi\-lib.html>.
514 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
515 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
518 Loads an external I/O engine object file. Append the engine filename as
522 IO engine that does regular linux native fallocate callt to simulate data
523 transfer as fio ioengine
525 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
527 DIR_WRITE does fallocate(,mode = 0)
529 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
532 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
533 request to DDIR_WRITE event
538 .BI iodepth \fR=\fPint
539 Number of I/O units to keep in flight against the file. Note that increasing
540 iodepth beyond 1 will not affect synchronous ioengines (except for small
541 degress when verify_async is in use). Even async engines my impose OS
542 restrictions causing the desired depth not to be achieved. This may happen on
543 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
544 not async on that OS. Keep an eye on the IO depth distribution in the
545 fio output to verify that the achieved depth is as expected. Default: 1.
547 .BI iodepth_batch \fR=\fPint
548 Number of I/Os to submit at once. Default: \fBiodepth\fR.
550 .BI iodepth_batch_complete \fR=\fPint
551 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
552 means that we'll ask for a minimum of 1 IO in the retrieval process from the
553 kernel. The IO retrieval will go on until we hit the limit set by
554 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
555 completed events before queuing more IO. This helps reduce IO latency, at the
556 cost of more retrieval system calls.
558 .BI iodepth_low \fR=\fPint
559 Low watermark indicating when to start filling the queue again. Default:
562 .BI direct \fR=\fPbool
563 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
565 .BI buffered \fR=\fPbool
566 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
569 .BI offset \fR=\fPint
570 Offset in the file to start I/O. Data before the offset will not be touched.
572 .BI offset_increment \fR=\fPint
573 If this is provided, then the real offset becomes the
574 offset + offset_increment * thread_number, where the thread number is a counter
575 that starts at 0 and is incremented for each job. This option is useful if
576 there are several jobs which are intended to operate on a file in parallel in
577 disjoint segments, with even spacing between the starting points.
579 .BI number_ios \fR=\fPint
580 Fio will normally perform IOs until it has exhausted the size of the region
581 set by \fBsize\fR, or if it exhaust the allocated time (or hits an error
582 condition). With this setting, the range/size can be set independently of
583 the number of IOs to perform. When fio reaches this number, it will exit
584 normally and report status.
587 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
588 0, don't sync. Default: 0.
590 .BI fdatasync \fR=\fPint
591 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
592 data parts of the file. Default: 0.
594 .BI sync_file_range \fR=\fPstr:int
595 Use sync_file_range() for every \fRval\fP number of write operations. Fio will
596 track range of writes that have happened since the last sync_file_range() call.
597 \fRstr\fP can currently be one or more of:
601 SYNC_FILE_RANGE_WAIT_BEFORE
604 SYNC_FILE_RANGE_WRITE
607 SYNC_FILE_RANGE_WRITE
611 So if you do sync_file_range=wait_before,write:8, fio would use
612 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
613 Also see the sync_file_range(2) man page. This option is Linux specific.
615 .BI overwrite \fR=\fPbool
616 If writing, setup the file first and do overwrites. Default: false.
618 .BI end_fsync \fR=\fPbool
619 Sync file contents when a write stage has completed. Default: false.
621 .BI fsync_on_close \fR=\fPbool
622 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
623 it will happen on every close, not just at the end of the job. Default: false.
625 .BI rwmixread \fR=\fPint
626 Percentage of a mixed workload that should be reads. Default: 50.
628 .BI rwmixwrite \fR=\fPint
629 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
630 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
631 overrides the first. This may interfere with a given rate setting, if fio is
632 asked to limit reads or writes to a certain rate. If that is the case, then
633 the distribution may be skewed. Default: 50.
635 .BI random_distribution \fR=\fPstr:float
636 By default, fio will use a completely uniform random distribution when asked
637 to perform random IO. Sometimes it is useful to skew the distribution in
638 specific ways, ensuring that some parts of the data is more hot than others.
639 Fio includes the following distribution models:
643 Uniform random distribution
653 When using a zipf or pareto distribution, an input value is also needed to
654 define the access pattern. For zipf, this is the zipf theta. For pareto,
655 it's the pareto power. Fio includes a test program, genzipf, that can be
656 used visualize what the given input values will yield in terms of hit rates.
657 If you wanted to use zipf with a theta of 1.2, you would use
658 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
659 fio will disable use of the random map.
661 .BI percentage_random \fR=\fPint
662 For a random workload, set how big a percentage should be random. This defaults
663 to 100%, in which case the workload is fully random. It can be set from
664 anywhere from 0 to 100. Setting it to 0 would make the workload fully
665 sequential. It is possible to set different values for reads, writes, and
666 trim. To do so, simply use a comma separated list. See \fBblocksize\fR.
669 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
670 this parameter is given, a new offset will be chosen without looking at past
671 I/O history. This parameter is mutually exclusive with \fBverify\fR.
673 .BI softrandommap \fR=\fPbool
674 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
675 fails to allocate the map, if this option is set it will continue without a
676 random block map. As coverage will not be as complete as with random maps, this
677 option is disabled by default.
679 .BI random_generator \fR=\fPstr
680 Fio supports the following engines for generating IO offsets for random IO:
684 Strong 2^88 cycle random number generator
687 Linear feedback shift register generator
691 Tausworthe is a strong random number generator, but it requires tracking on the
692 side if we want to ensure that blocks are only read or written once. LFSR
693 guarantees that we never generate the same offset twice, and it's also less
694 computationally expensive. It's not a true random generator, however, though
695 for IO purposes it's typically good enough. LFSR only works with single block
696 sizes, not with workloads that use multiple block sizes. If used with such a
697 workload, fio may read or write some blocks multiple times.
700 Run job with given nice value. See \fInice\fR\|(2).
703 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
706 .BI prioclass \fR=\fPint
707 Set I/O priority class. See \fIionice\fR\|(1).
709 .BI thinktime \fR=\fPint
710 Stall job for given number of microseconds between issuing I/Os.
712 .BI thinktime_spin \fR=\fPint
713 Pretend to spend CPU time for given number of microseconds, sleeping the rest
714 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
716 .BI thinktime_blocks \fR=\fPint
717 Only valid if thinktime is set - control how many blocks to issue, before
718 waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will
719 make fio wait \fBthinktime\fR microseconds after every block. This
720 effectively makes any queue depth setting redundant, since no more than 1 IO
721 will be queued before we have to complete it and do our thinktime. In other
722 words, this setting effectively caps the queue depth if the latter is larger.
726 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
727 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
728 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
729 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
730 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
731 limit writes (to 500KB/sec), the latter will only limit reads.
733 .BI ratemin \fR=\fPint
734 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
735 Failing to meet this requirement will cause the job to exit. The same format
736 as \fBrate\fR is used for read vs write separation.
738 .BI rate_iops \fR=\fPint
739 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
740 specified independently of bandwidth. The same format as \fBrate\fR is used for
741 read vs write seperation. If \fBblocksize\fR is a range, the smallest block
742 size is used as the metric.
744 .BI rate_iops_min \fR=\fPint
745 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
746 is used for read vs write seperation.
748 .BI ratecycle \fR=\fPint
749 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
750 milliseconds. Default: 1000ms.
752 .BI max_latency \fR=\fPint
753 If set, fio will exit the job if it exceeds this maximum latency. It will exit
756 .BI cpumask \fR=\fPint
757 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
758 may run on. See \fBsched_setaffinity\fR\|(2).
760 .BI cpus_allowed \fR=\fPstr
761 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
763 .BI numa_cpu_nodes \fR=\fPstr
764 Set this job running on spcified NUMA nodes' CPUs. The arguments allow
765 comma delimited list of cpu numbers, A-B ranges, or 'all'.
767 .BI numa_mem_policy \fR=\fPstr
768 Set this job's memory policy and corresponding NUMA nodes. Format of
772 .B <mode>[:<nodelist>]
775 is one of the following memory policy:
777 .B default, prefer, bind, interleave, local
780 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
781 needed to be specified. For \fBprefer\fR, only one node is
782 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
783 comma delimited list of numbers, A-B ranges, or 'all'.
785 .BI startdelay \fR=\fPint
786 Delay start of job for the specified number of seconds.
788 .BI runtime \fR=\fPint
789 Terminate processing after the specified number of seconds.
792 If given, run for the specified \fBruntime\fR duration even if the files are
793 completely read or written. The same workload will be repeated as many times
794 as \fBruntime\fR allows.
796 .BI ramp_time \fR=\fPint
797 If set, fio will run the specified workload for this amount of time before
798 logging any performance numbers. Useful for letting performance settle before
799 logging results, thus minimizing the runtime required for stable results. Note
800 that the \fBramp_time\fR is considered lead in time for a job, thus it will
801 increase the total runtime if a special timeout or runtime is specified.
803 .BI invalidate \fR=\fPbool
804 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
807 Use synchronous I/O for buffered writes. For the majority of I/O engines,
808 this means using O_SYNC. Default: false.
810 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
811 Allocation method for I/O unit buffer. Allowed values are:
816 Allocate memory with \fImalloc\fR\|(3).
819 Use shared memory buffers allocated through \fIshmget\fR\|(2).
822 Same as \fBshm\fR, but use huge pages as backing.
825 Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
826 is given after the option in the format `:\fIfile\fR'.
829 Same as \fBmmap\fR, but use huge files as backing.
832 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
833 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
834 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
835 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
836 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
837 and the documentation for that. Normally you just need to echo an appropriate
838 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
842 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
843 This indiciates the memory alignment of the IO memory buffers. Note that the
844 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
845 the alignment of the following buffers are given by the \fBbs\fR used. In
846 other words, if using a \fBbs\fR that is a multiple of the page sized in the
847 system, all buffers will be aligned to this value. If using a \fBbs\fR that
848 is not page aligned, the alignment of subsequent IO memory buffers is the
849 sum of the \fBiomem_align\fR and \fBbs\fR used.
851 .BI hugepage\-size \fR=\fPint
852 Defines the size of a huge page. Must be at least equal to the system setting.
853 Should be a multiple of 1MB. Default: 4MB.
856 Terminate all jobs when one finishes. Default: wait for each job to finish.
858 .BI bwavgtime \fR=\fPint
859 Average bandwidth calculations over the given time in milliseconds. Default:
862 .BI iopsavgtime \fR=\fPint
863 Average IOPS calculations over the given time in milliseconds. Default:
866 .BI create_serialize \fR=\fPbool
867 If true, serialize file creation for the jobs. Default: true.
869 .BI create_fsync \fR=\fPbool
870 \fIfsync\fR\|(2) data file after creation. Default: true.
872 .BI create_on_open \fR=\fPbool
873 If true, the files are not created until they are opened for IO by the job.
875 .BI create_only \fR=\fPbool
876 If true, fio will only run the setup phase of the job. If files need to be
877 laid out or updated on disk, only that will be done. The actual job contents
880 .BI pre_read \fR=\fPbool
881 If this is given, files will be pre-read into memory before starting the given
882 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
883 pointless to pre-read and then drop the cache. This will only work for IO
884 engines that are seekable, since they allow you to read the same data
885 multiple times. Thus it will not work on eg network or splice IO.
887 .BI unlink \fR=\fPbool
888 Unlink job files when done. Default: false.
891 Specifies the number of iterations (runs of the same workload) of this job.
894 .BI do_verify \fR=\fPbool
895 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
898 .BI verify \fR=\fPstr
899 Method of verifying file contents after each iteration of the job. Allowed
904 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
905 Store appropriate checksum in the header of each block. crc32c-intel is
906 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
907 not supported by the system.
910 Write extra information about each I/O (timestamp, block number, etc.). The
911 block number is verified. See \fBverify_pattern\fR as well.
914 Pretend to verify. Used for testing internals.
917 This option can be used for repeated burn-in tests of a system to make sure
918 that the written data is also correctly read back. If the data direction given
919 is a read or random read, fio will assume that it should verify a previously
920 written file. If the data direction includes any form of write, the verify will
921 be of the newly written data.
924 .BI verify_sort \fR=\fPbool
925 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
926 read them back in a sorted manner. Default: true.
928 .BI verify_offset \fR=\fPint
929 Swap the verification header with data somewhere else in the block before
930 writing. It is swapped back before verifying.
932 .BI verify_interval \fR=\fPint
933 Write the verification header for this number of bytes, which should divide
934 \fBblocksize\fR. Default: \fBblocksize\fR.
936 .BI verify_pattern \fR=\fPstr
937 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
938 with totally random bytes, but sometimes it's interesting to fill with a known
939 pattern for io verification purposes. Depending on the width of the pattern,
940 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
941 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
942 has to be a hex number that starts with either "0x" or "0X". Use with
945 .BI verify_fatal \fR=\fPbool
946 If true, exit the job on the first observed verification failure. Default:
949 .BI verify_dump \fR=\fPbool
950 If set, dump the contents of both the original data block and the data block we
951 read off disk to files. This allows later analysis to inspect just what kind of
952 data corruption occurred. Off by default.
954 .BI verify_async \fR=\fPint
955 Fio will normally verify IO inline from the submitting thread. This option
956 takes an integer describing how many async offload threads to create for IO
957 verification instead, causing fio to offload the duty of verifying IO contents
958 to one or more separate threads. If using this offload option, even sync IO
959 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
960 allows them to have IO in flight while verifies are running.
962 .BI verify_async_cpus \fR=\fPstr
963 Tell fio to set the given CPU affinity on the async IO verification threads.
964 See \fBcpus_allowed\fP for the format used.
966 .BI verify_backlog \fR=\fPint
967 Fio will normally verify the written contents of a job that utilizes verify
968 once that job has completed. In other words, everything is written then
969 everything is read back and verified. You may want to verify continually
970 instead for a variety of reasons. Fio stores the meta data associated with an
971 IO block in memory, so for large verify workloads, quite a bit of memory would
972 be used up holding this meta data. If this option is enabled, fio will write
973 only N blocks before verifying these blocks.
975 .BI verify_backlog_batch \fR=\fPint
976 Control how many blocks fio will verify if verify_backlog is set. If not set,
977 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
978 read back and verified). If \fBverify_backlog_batch\fR is less than
979 \fBverify_backlog\fR then not all blocks will be verified, if
980 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
981 will be verified more than once.
983 .B stonewall "\fR,\fP wait_for_previous"
984 Wait for preceding jobs in the job file to exit before starting this one.
985 \fBstonewall\fR implies \fBnew_group\fR.
988 Start a new reporting group. If not given, all jobs in a file will be part
989 of the same reporting group, unless separated by a stonewall.
991 .BI numjobs \fR=\fPint
992 Number of clones (processes/threads performing the same workload) of this job.
996 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
1000 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
1001 with \fBfork\fR\|(2).
1003 .BI zonesize \fR=\fPint
1004 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
1006 .BI zoneskip \fR=\fPint
1007 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
1010 .BI write_iolog \fR=\fPstr
1011 Write the issued I/O patterns to the specified file. Specify a separate file
1012 for each job, otherwise the iologs will be interspersed and the file may be
1015 .BI read_iolog \fR=\fPstr
1016 Replay the I/O patterns contained in the specified file generated by
1017 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
1019 .BI replay_no_stall \fR=\fPint
1020 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1021 attempts to respect timing information between I/Os. Enabling
1022 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
1023 still respecting ordering.
1025 .BI replay_redirect \fR=\fPstr
1026 While replaying I/O patterns using \fBread_iolog\fR the default behavior
1027 is to replay the IOPS onto the major/minor device that each IOP was recorded
1028 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
1029 single specified device regardless of the device it was recorded from.
1031 .BI write_bw_log \fR=\fPstr
1032 If given, write a bandwidth log of the jobs in this job file. Can be used to
1033 store data of the bandwidth of the jobs in their lifetime. The included
1034 fio_generate_plots script uses gnuplot to turn these text files into nice
1035 graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
1036 option, the postfix is _bw.log.
1038 .BI write_lat_log \fR=\fPstr
1039 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
1040 filename is given with this option, the default filename of "jobname_type.log"
1041 is used. Even if the filename is given, fio will still append the type of log.
1043 .BI write_iops_log \fR=\fPstr
1044 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
1045 option, the default filename of "jobname_type.log" is used. Even if the
1046 filename is given, fio will still append the type of log.
1048 .BI log_avg_msec \fR=\fPint
1049 By default, fio will log an entry in the iops, latency, or bw log for every
1050 IO that completes. When writing to the disk log, that can quickly grow to a
1051 very large size. Setting this option makes fio average the each log entry
1052 over the specified period of time, reducing the resolution of the log.
1055 .BI disable_lat \fR=\fPbool
1056 Disable measurements of total latency numbers. Useful only for cutting
1057 back the number of calls to gettimeofday, as that does impact performance at
1058 really high IOPS rates. Note that to really get rid of a large amount of these
1059 calls, this option must be used with disable_slat and disable_bw as well.
1061 .BI disable_clat \fR=\fPbool
1062 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1064 .BI disable_slat \fR=\fPbool
1065 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1067 .BI disable_bw_measurement \fR=\fPbool
1068 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1070 .BI lockmem \fR=\fPint
1071 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1072 simulate a smaller amount of memory. The amount specified is per worker.
1074 .BI exec_prerun \fR=\fPstr
1075 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1077 Output is redirected in a file called \fBjobname.prerun.txt\fR
1080 .BI exec_postrun \fR=\fPstr
1081 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1083 Output is redirected in a file called \fBjobname.postrun.txt\fR
1086 .BI ioscheduler \fR=\fPstr
1087 Attempt to switch the device hosting the file to the specified I/O scheduler.
1089 .BI cpuload \fR=\fPint
1090 If the job is a CPU cycle-eater, attempt to use the specified percentage of
1093 .BI cpuchunks \fR=\fPint
1094 If the job is a CPU cycle-eater, split the load into cycles of the
1095 given time in milliseconds.
1097 .BI disk_util \fR=\fPbool
1098 Generate disk utilization statistics if the platform supports it. Default: true.
1100 .BI clocksource \fR=\fPstr
1101 Use the given clocksource as the base of timing. The supported options are:
1111 Internal CPU clock source
1115 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1116 (and fio is heavy on time calls). Fio will automatically use this clocksource
1117 if it's supported and considered reliable on the system it is running on,
1118 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1119 means supporting TSC Invariant.
1121 .BI gtod_reduce \fR=\fPbool
1122 Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
1123 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1124 gettimeofday() call count. With this option enabled, we only do about 0.4% of
1125 the gtod() calls we would have done if all time keeping was enabled.
1127 .BI gtod_cpu \fR=\fPint
1128 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1129 the current time. Fio (and databases, for instance) are very intensive on
1130 gettimeofday() calls. With this option, you can set one CPU aside for doing
1131 nothing but logging current time to a shared memory location. Then the other
1132 threads/processes that run IO workloads need only copy that segment, instead of
1133 entering the kernel with a gettimeofday() call. The CPU set aside for doing
1134 these time calls will be excluded from other uses. Fio will manually clear it
1135 from the CPU mask of other jobs.
1137 .BI ignore_error \fR=\fPstr
1138 Sometimes you want to ignore some errors during test in that case you can specify
1139 error list for each error type.
1141 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1143 errors for given error type is separated with ':'.
1144 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1146 Example: ignore_error=EAGAIN,ENOSPC:122 .
1148 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1150 .BI error_dump \fR=\fPbool
1151 If set dump every error even if it is non fatal, true by default. If disabled
1152 only fatal error will be dumped
1154 .BI cgroup \fR=\fPstr
1155 Add job to this control group. If it doesn't exist, it will be created.
1156 The system must have a mounted cgroup blkio mount point for this to work. If
1157 your system doesn't have it mounted, you can do so with:
1159 # mount \-t cgroup \-o blkio none /cgroup
1161 .BI cgroup_weight \fR=\fPint
1162 Set the weight of the cgroup to this value. See the documentation that comes
1163 with the kernel, allowed values are in the range of 100..1000.
1165 .BI cgroup_nodelete \fR=\fPbool
1166 Normally fio will delete the cgroups it has created after the job completion.
1167 To override this behavior and to leave cgroups around after the job completion,
1168 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1169 cgroup files after job completion. Default: false
1172 Instead of running as the invoking user, set the user ID to this value before
1173 the thread/process does any work.
1176 Set group ID, see \fBuid\fR.
1178 .BI flow_id \fR=\fPint
1179 The ID of the flow. If not specified, it defaults to being a global flow. See
1183 Weight in token-based flow control. If this value is used, then there is a
1184 \fBflow counter\fR which is used to regulate the proportion of activity between
1185 two or more jobs. fio attempts to keep this flow counter near zero. The
1186 \fBflow\fR parameter stands for how much should be added or subtracted to the
1187 flow counter on each iteration of the main I/O loop. That is, if one job has
1188 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1189 1:8 ratio in how much one runs vs the other.
1191 .BI flow_watermark \fR=\fPint
1192 The maximum value that the absolute value of the flow counter is allowed to
1193 reach before the job must wait for a lower value of the counter.
1195 .BI flow_sleep \fR=\fPint
1196 The period of time, in microseconds, to wait after the flow watermark has been
1197 exceeded before retrying operations
1199 .BI clat_percentiles \fR=\fPbool
1200 Enable the reporting of percentiles of completion latencies.
1202 .BI percentile_list \fR=\fPfloat_list
1203 Overwrite the default list of percentiles for completion
1204 latencies. Each number is a floating number in the range (0,100], and
1205 the maximum length of the list is 20. Use ':' to separate the
1206 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1207 report the values of completion latency below which 99.5% and 99.9% of
1208 the observed latencies fell, respectively.
1209 .SS "Ioengine Parameters List"
1210 Some parameters are only valid when a specific ioengine is in use. These are
1211 used identically to normal parameters, with the caveat that when used on the
1212 command line, the must come after the ioengine that defines them is selected.
1214 .BI (cpu)cpuload \fR=\fPint
1215 Attempt to use the specified percentage of CPU cycles.
1217 .BI (cpu)cpuchunks \fR=\fPint
1218 Split the load into cycles of the given time. In microseconds.
1220 .BI (libaio)userspace_reap
1221 Normally, with the libaio engine in use, fio will use
1222 the io_getevents system call to reap newly returned events.
1223 With this flag turned on, the AIO ring will be read directly
1224 from user-space to reap events. The reaping mode is only
1225 enabled when polling for a minimum of 0 events (eg when
1226 iodepth_batch_complete=0).
1228 .BI (net,netsplice)hostname \fR=\fPstr
1229 The host name or IP address to use for TCP or UDP based IO.
1230 If the job is a TCP listener or UDP reader, the hostname is not
1231 used and must be omitted unless it is a valid UDP multicast address.
1233 .BI (net,netsplice)port \fR=\fPint
1234 The TCP or UDP port to bind to or connect to.
1236 .BI (net,netsplice)interface \fR=\fPstr
1237 The IP address of the network interface used to send or receive UDP multicast
1240 .BI (net,netsplice)ttl \fR=\fPint
1241 Time-to-live value for outgoing UDP multicast packets. Default: 1
1243 .BI (net,netsplice)nodelay \fR=\fPbool
1244 Set TCP_NODELAY on TCP connections.
1246 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1247 The network protocol to use. Accepted values are:
1252 Transmission control protocol
1255 User datagram protocol
1261 When the protocol is TCP or UDP, the port must also be given,
1262 as well as the hostname if the job is a TCP listener or UDP
1263 reader. For unix sockets, the normal filename option should be
1264 used and the port is invalid.
1267 .BI (net,netsplice)listen
1268 For TCP network connections, tell fio to listen for incoming
1269 connections rather than initiating an outgoing connection. The
1270 hostname must be omitted if this option is used.
1272 .BI (net, pingpong) \fR=\fPbool
1273 Normaly a network writer will just continue writing data, and a network reader
1274 will just consume packages. If pingpong=1 is set, a writer will send its normal
1275 payload to the reader, then wait for the reader to send the same payload back.
1276 This allows fio to measure network latencies. The submission and completion
1277 latencies then measure local time spent sending or receiving, and the
1278 completion latency measures how long it took for the other end to receive and
1279 send back. For UDP multicast traffic pingpong=1 should only be set for a single
1280 reader when multiple readers are listening to the same address.
1282 .BI (e4defrag,donorname) \fR=\fPstr
1283 File will be used as a block donor (swap extents between files)
1285 .BI (e4defrag,inplace) \fR=\fPint
1286 Configure donor file block allocation strategy
1289 Preallocate donor's file on init
1292 allocate space immidietly inside defragment event, and free right after event
1296 While running, \fBfio\fR will display the status of the created jobs. For
1300 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1303 The characters in the first set of brackets denote the current status of each
1304 threads. The possible values are:
1310 Setup but not started.
1316 Initialized, waiting.
1319 Running, doing sequential reads.
1322 Running, doing random reads.
1325 Running, doing sequential writes.
1328 Running, doing random writes.
1331 Running, doing mixed sequential reads/writes.
1334 Running, doing mixed random reads/writes.
1337 Running, currently waiting for \fBfsync\fR\|(2).
1340 Running, verifying written data.
1343 Exited, not reaped by main thread.
1346 Exited, thread reaped.
1350 The second set of brackets shows the estimated completion percentage of
1351 the current group. The third set shows the read and write I/O rate,
1352 respectively. Finally, the estimated run time of the job is displayed.
1354 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1355 for each thread, each group of threads, and each disk, in that order.
1357 Per-thread statistics first show the threads client number, group-id, and
1358 error code. The remaining figures are as follows:
1362 Number of megabytes of I/O performed.
1365 Average data rate (bandwidth).
1371 Submission latency minimum, maximum, average and standard deviation. This is
1372 the time it took to submit the I/O.
1375 Completion latency minimum, maximum, average and standard deviation. This
1376 is the time between submission and completion.
1379 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1380 and standard deviation.
1383 CPU usage statistics. Includes user and system time, number of context switches
1384 this thread went through and number of major and minor page faults.
1387 Distribution of I/O depths. Each depth includes everything less than (or equal)
1388 to it, but greater than the previous depth.
1391 Number of read/write requests issued, and number of short read/write requests.
1394 Distribution of I/O completion latencies. The numbers follow the same pattern
1398 The group statistics show:
1403 Number of megabytes I/O performed.
1406 Aggregate bandwidth of threads in the group.
1409 Minimum average bandwidth a thread saw.
1412 Maximum average bandwidth a thread saw.
1415 Shortest runtime of threads in the group.
1418 Longest runtime of threads in the group.
1422 Finally, disk statistics are printed with reads first:
1427 Number of I/Os performed by all groups.
1430 Number of merges in the I/O scheduler.
1433 Number of ticks we kept the disk busy.
1436 Total time spent in the disk queue.
1443 It is also possible to get fio to dump the current output while it is
1444 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1447 If the \fB\-\-minimal\fR option is given, the results will be printed in a
1448 semicolon-delimited format suitable for scripted use - a job description
1449 (if provided) follows on a new line. Note that the first
1450 number in the line is the version number. If the output has to be changed
1451 for some reason, this number will be incremented by 1 to signify that
1452 change. The fields are:
1455 .B terse version, fio version, jobname, groupid, error
1459 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1463 .B min, max, mean, standard deviation
1467 .B min, max, mean, standard deviation
1469 Completion latency percentiles (20 fields):
1471 .B Xth percentile=usec
1475 .B min, max, mean, standard deviation
1479 .B min, max, aggregate percentage of total, mean, standard deviation
1485 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1489 .B min, max, mean, standard deviation
1493 .B min, max, mean, standard deviation
1495 Completion latency percentiles (20 fields):
1497 .B Xth percentile=usec
1501 .B min, max, mean, standard deviation
1505 .B min, max, aggregate percentage of total, mean, standard deviation
1511 .B user, system, context switches, major page faults, minor page faults
1514 IO depth distribution:
1516 .B <=1, 2, 4, 8, 16, 32, >=64
1519 IO latency distribution:
1523 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1527 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1531 Disk utilization (1 for each disk used):
1533 .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
1536 Error Info (dependent on continue_on_error, default off):
1538 .B total # errors, first error code
1541 .B text description (if provided in config - appears on newline)
1544 Normally you would run fio as a stand-alone application on the machine
1545 where the IO workload should be generated. However, it is also possible to
1546 run the frontend and backend of fio separately. This makes it possible to
1547 have a fio server running on the machine(s) where the IO workload should
1548 be running, while controlling it from another machine.
1550 To start the server, you would do:
1552 \fBfio \-\-server=args\fR
1554 on that machine, where args defines what fio listens to. The arguments
1555 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1556 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1557 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1558 listen to (only valid for TCP/IP, not a local socket). Some examples:
1562 Start a fio server, listening on all interfaces on the default port (8765).
1564 2) fio \-\-server=ip:hostname,4444
1566 Start a fio server, listening on IP belonging to hostname and on port 4444.
1568 3) fio \-\-server=ip6:::1,4444
1570 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1572 4) fio \-\-server=,4444
1574 Start a fio server, listening on all interfaces on port 4444.
1576 5) fio \-\-server=1.2.3.4
1578 Start a fio server, listening on IP 1.2.3.4 on the default port.
1580 6) fio \-\-server=sock:/tmp/fio.sock
1582 Start a fio server, listening on the local socket /tmp/fio.sock.
1584 When a server is running, you can connect to it from a client. The client
1587 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1589 where \-\-local-args are arguments that are local to the client where it is
1590 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1591 are sent to the server. The 'server' string follows the same format as it
1592 does on the server side, to allow IP/hostname/socket and port strings.
1593 You can connect to multiple clients as well, to do that you could run:
1595 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1599 was written by Jens Axboe <jens.axboe@oracle.com>,
1600 now Jens Axboe <jaxboe@fusionio.com>.
1602 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1603 on documentation by Jens Axboe.
1604 .SH "REPORTING BUGS"
1605 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1608 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1610 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob