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'.
58 Turn on safety read-only checks, preventing any attempted write.
60 .BI \-\-section \fR=\fPsec
61 Only run section \fIsec\fR from job file. Multiple of these options can be given, adding more sections to run.
63 .BI \-\-alloc\-size \fR=\fPkb
64 Set the internal smalloc pool size to \fIkb\fP kilobytes.
66 .BI \-\-warnings\-fatal
67 All fio parser warnings are fatal, causing fio to exit with an error.
69 .BI \-\-max\-jobs \fR=\fPnr
70 Set the maximum allowed number of jobs (threads/processes) to support.
72 .BI \-\-server \fR=\fPargs
73 Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section.
75 .BI \-\-daemonize \fR=\fPpidfile
76 Background a fio server, writing the pid to the given pid file.
78 .BI \-\-client \fR=\fPhost
79 Instead of running the jobs locally, send and run them on the given host.
81 .BI \-\-idle\-prof \fR=\fPoption
82 Report cpu idleness on a system or percpu basis (\fIoption\fP=system,percpu) or run unit work calibration only (\fIoption\fP=calibrate).
84 Job files are in `ini' format. They consist of one or more
85 job definitions, which begin with a job name in square brackets and
86 extend to the next job name. The job name can be any ASCII string
87 except `global', which has a special meaning. Following the job name is
88 a sequence of zero or more parameters, one per line, that define the
89 behavior of the job. Any line starting with a `;' or `#' character is
90 considered a comment and ignored.
92 If \fIjobfile\fR is specified as `-', the job file will be read from
95 The global section contains default parameters for jobs specified in the
96 job file. A job is only affected by global sections residing above it,
97 and there may be any number of global sections. Specific job definitions
98 may override any parameter set in global sections.
101 Some parameters may take arguments of a specific type. The types used are:
104 String: a sequence of alphanumeric characters.
107 SI integer: a whole number, possibly containing a suffix denoting the base unit
108 of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting
109 kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5)
110 respectively. The suffix is not case sensitive. If prefixed with '0x', the
111 value is assumed to be base 16 (hexadecimal). A suffix may include a trailing 'b',
112 for instance 'kb' is identical to 'k'. You can specify a base 10 value
113 by using 'KiB', 'MiB', 'GiB', etc. This is useful for disk drives where
114 values are often given in base 10 values. Specifying '30GiB' will get you
118 Boolean: a true or false value. `0' denotes false, `1' denotes true.
121 Integer range: a range of integers specified in the format
122 \fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and
123 \fIupper\fR may contain a suffix as described above. If an option allows two
124 sets of ranges, they are separated with a `,' or `/' character. For example:
128 List of floating numbers: A list of floating numbers, separated by
133 May be used to override the job name. On the command line, this parameter
134 has the special purpose of signalling the start of a new job.
136 .BI description \fR=\fPstr
137 Human-readable description of the job. It is printed when the job is run, but
138 otherwise has no special purpose.
140 .BI directory \fR=\fPstr
141 Prefix filenames with this directory. Used to place files in a location other
144 .BI filename \fR=\fPstr
146 normally makes up a file name based on the job name, thread number, and file
147 number. If you want to share files between threads in a job or several jobs,
148 specify a \fIfilename\fR for each of them to override the default.
149 If the I/O engine is file-based, you can specify
150 a number of files by separating the names with a `:' character. `\-' is a
151 reserved name, meaning stdin or stdout, depending on the read/write direction
154 .BI lockfile \fR=\fPstr
155 Fio defaults to not locking any files before it does IO to them. If a file or
156 file descriptor is shared, fio can serialize IO to that file to make the end
157 result consistent. This is usual for emulating real workloads that share files.
163 No locking. This is the default.
166 Only one thread or process may do IO at the time, excluding all others.
169 Read-write locking on the file. Many readers may access the file at the same
170 time, but writes get exclusive access.
173 The option may be post-fixed with a lock batch number. If set, then each
174 thread/process may do that amount of IOs to the file before giving up the lock.
175 Since lock acquisition is expensive, batching the lock/unlocks will speed up IO.
178 .BI opendir \fR=\fPstr
179 Recursively open any files below directory \fIstr\fR.
181 .BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr
182 Type of I/O pattern. Accepted values are:
199 Mixed sequential reads and writes.
202 Mixed random reads and writes.
205 For mixed I/O, the default split is 50/50. For certain types of io the result
206 may still be skewed a bit, since the speed may be different. It is possible to
207 specify a number of IO's to do before getting a new offset, this is done by
208 appending a `:\fI<nr>\fR to the end of the string given. For a random read, it
209 would look like \fBrw=randread:8\fR for passing in an offset modifier with a
210 value of 8. If the postfix is used with a sequential IO pattern, then the value
211 specified will be added to the generated offset for each IO. For instance,
212 using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO
213 into sequential IO with holes. See the \fBrw_sequencer\fR option.
216 .BI rw_sequencer \fR=\fPstr
217 If an offset modifier is given by appending a number to the \fBrw=<str>\fR line,
218 then this option controls how that number modifies the IO offset being
219 generated. Accepted values are:
224 Generate sequential offset
227 Generate the same offset
230 \fBsequential\fR is only useful for random IO, where fio would normally
231 generate a new random offset for every IO. If you append eg 8 to randread, you
232 would get a new random offset for every 8 IO's. The result would be a seek for
233 only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify
234 that. As sequential IO is already sequential, setting \fBsequential\fR for that
235 would not result in any differences. \fBidentical\fR behaves in a similar
236 fashion, except it sends the same offset 8 number of times before generating a
241 .BI kb_base \fR=\fPint
242 The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage
243 manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious
244 reasons. Allow values are 1024 or 1000, with 1024 being the default.
246 .BI unified_rw_reporting \fR=\fPbool
247 Fio normally reports statistics on a per data direction basis, meaning that
248 read, write, and trim are accounted and reported separately. If this option is
249 set, the fio will sum the results and report them as "mixed" instead.
251 .BI randrepeat \fR=\fPbool
252 Seed the random number generator in a predictable way so results are repeatable
253 across runs. Default: true.
255 .BI use_os_rand \fR=\fPbool
256 Fio can either use the random generator supplied by the OS to generator random
257 offsets, or it can use it's own internal generator (based on Tausworthe).
258 Default is to use the internal generator, which is often of better quality and
259 faster. Default: false.
261 .BI fallocate \fR=\fPstr
262 Whether pre-allocation is performed when laying down files. Accepted values
268 Do not pre-allocate space.
271 Pre-allocate via posix_fallocate().
274 Pre-allocate via fallocate() with FALLOC_FL_KEEP_SIZE set.
277 Backward-compatible alias for 'none'.
280 Backward-compatible alias for 'posix'.
283 May not be available on all supported platforms. 'keep' is only
284 available on Linux. If using ZFS on Solaris this must be set to 'none'
285 because ZFS doesn't support it. Default: 'posix'.
288 .BI fadvise_hint \fR=\fPbool
289 Use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns
290 are likely to be issued. Default: true.
293 Total size of I/O for this job. \fBfio\fR will run until this many bytes have
294 been transfered, unless limited by other options (\fBruntime\fR, for instance).
295 Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be
296 divided between the available files for the job. If not set, fio will use the
297 full size of the given files or devices. If the the files do not exist, size
298 must be given. It is also possible to give size as a percentage between 1 and
299 100. If size=20% is given, fio will use 20% of the full size of the given files
302 .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool
303 Sets size to something really large and waits for ENOSPC (no space left on
304 device) as the terminating condition. Only makes sense with sequential write.
305 For a read workload, the mount point will be filled first then IO started on
306 the result. This option doesn't make sense if operating on a raw device node,
307 since the size of that is already known by the file system. Additionally,
308 writing beyond end-of-device will not return ENOSPC there.
310 .BI filesize \fR=\fPirange
311 Individual file sizes. May be a range, in which case \fBfio\fR will select sizes
312 for files at random within the given range, limited to \fBsize\fR in total (if
313 that is given). If \fBfilesize\fR is not specified, each created file is the
316 .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int]
317 Block size for I/O units. Default: 4k. Values for reads and writes can be
318 specified separately in the format \fIread\fR,\fIwrite\fR, either of
319 which may be empty to leave that value at its default.
321 .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange]
322 Specify a range of I/O block sizes. The issued I/O unit will always be a
323 multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies
324 to both reads and writes if only one range is given, but can be specified
325 separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k.
326 Also (see \fBblocksize\fR).
328 .BI bssplit \fR=\fPstr
329 This option allows even finer grained control of the block sizes issued,
330 not just even splits between them. With this option, you can weight various
331 block sizes for exact control of the issued IO for a job that has mixed
332 block sizes. The format of the option is bssplit=blocksize/percentage,
333 optionally adding as many definitions as needed separated by a colon.
334 Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k
335 blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate
336 splits to reads and writes. The format is identical to what the
337 \fBbs\fR option accepts, the read and write parts are separated with a
340 .B blocksize_unaligned\fR,\fP bs_unaligned
341 If set, any size in \fBblocksize_range\fR may be used. This typically won't
342 work with direct I/O, as that normally requires sector alignment.
344 .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int]
345 At what boundary to align random IO offsets. Defaults to the same as 'blocksize'
346 the minimum blocksize given. Minimum alignment is typically 512b
347 for using direct IO, though it usually depends on the hardware block size.
348 This option is mutually exclusive with using a random map for files, so it
349 will turn off that option.
352 Initialise buffers with all zeros. Default: fill buffers with random data.
355 If this option is given, fio will refill the IO buffers on every submit. The
356 default is to only fill it at init time and reuse that data. Only makes sense
357 if zero_buffers isn't specified, naturally. If data verification is enabled,
358 refill_buffers is also automatically enabled.
360 .BI scramble_buffers \fR=\fPbool
361 If \fBrefill_buffers\fR is too costly and the target is using data
362 deduplication, then setting this option will slightly modify the IO buffer
363 contents to defeat normal de-dupe attempts. This is not enough to defeat
364 more clever block compression attempts, but it will stop naive dedupe
365 of blocks. Default: true.
367 .BI buffer_compress_percentage \fR=\fPint
368 If this is set, then fio will attempt to provide IO buffer content (on WRITEs)
369 that compress to the specified level. Fio does this by providing a mix of
370 random data and zeroes. Note that this is per block size unit, for file/disk
371 wide compression level that matches this setting, you'll also want to set
372 \fBrefill_buffers\fR.
374 .BI buffer_compress_chunk \fR=\fPint
375 See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how
376 big the ranges of random data and zeroed data is. Without this set, fio will
377 provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by
378 the remaining zeroed. With this set to some chunk size smaller than the block
379 size, fio can alternate random and zeroed data throughout the IO buffer.
381 .BI nrfiles \fR=\fPint
382 Number of files to use for this job. Default: 1.
384 .BI openfiles \fR=\fPint
385 Number of files to keep open at the same time. Default: \fBnrfiles\fR.
387 .BI file_service_type \fR=\fPstr
388 Defines how files to service are selected. The following types are defined:
393 Choose a file at random
396 Round robin over open files (default).
398 Do each file in the set sequentially.
401 The number of I/Os to issue before switching a new file can be specified by
402 appending `:\fIint\fR' to the service type.
405 .BI ioengine \fR=\fPstr
406 Defines how the job issues I/O. The following types are defined:
411 Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to
412 position the I/O location.
415 Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O.
418 Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by
419 coalescing adjacents IOs into a single submission.
422 Linux native asynchronous I/O. This ioengine defines engine specific options.
425 POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3).
428 Solaris native asynchronous I/O.
431 Windows native asynchronous I/O.
434 File is memory mapped with \fImmap\fR\|(2) and data copied using
438 \fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to
439 transfer data from user-space to the kernel.
442 Use the syslet system calls to make regular read/write asynchronous.
445 SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if
446 the target is an sg character device, we use \fIread\fR\|(2) and
447 \fIwrite\fR\|(2) for asynchronous I/O.
450 Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR
451 itself and for debugging and testing purposes.
454 Transfer over the network. The protocol to be used can be defined with the
455 \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR,
456 \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified.
457 This ioengine defines engine specific options.
460 Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data
461 and send/receive. This ioengine defines engine specific options.
464 Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and
465 \fBcpucycles\fR parameters.
468 The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface
469 approach to asycnronous I/O.
471 See <http://www.xmailserver.org/guasi\-lib.html>.
474 The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ)
475 and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols.
478 Loads an external I/O engine object file. Append the engine filename as
482 IO engine that does regular linux native fallocate callt to simulate data
483 transfer as fio ioengine
485 DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,)
487 DIR_WRITE does fallocate(,mode = 0)
489 DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE)
492 IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity
493 request to DDIR_WRITE event
498 .BI iodepth \fR=\fPint
499 Number of I/O units to keep in flight against the file. Note that increasing
500 iodepth beyond 1 will not affect synchronous ioengines (except for small
501 degress when verify_async is in use). Even async engines my impose OS
502 restrictions causing the desired depth not to be achieved. This may happen on
503 Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is
504 not async on that OS. Keep an eye on the IO depth distribution in the
505 fio output to verify that the achieved depth is as expected. Default: 1.
507 .BI iodepth_batch \fR=\fPint
508 Number of I/Os to submit at once. Default: \fBiodepth\fR.
510 .BI iodepth_batch_complete \fR=\fPint
511 This defines how many pieces of IO to retrieve at once. It defaults to 1 which
512 means that we'll ask for a minimum of 1 IO in the retrieval process from the
513 kernel. The IO retrieval will go on until we hit the limit set by
514 \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for
515 completed events before queuing more IO. This helps reduce IO latency, at the
516 cost of more retrieval system calls.
518 .BI iodepth_low \fR=\fPint
519 Low watermark indicating when to start filling the queue again. Default:
522 .BI direct \fR=\fPbool
523 If true, use non-buffered I/O (usually O_DIRECT). Default: false.
525 .BI buffered \fR=\fPbool
526 If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter.
529 .BI offset \fR=\fPint
530 Offset in the file to start I/O. Data before the offset will not be touched.
532 .BI offset_increment \fR=\fPint
533 If this is provided, then the real offset becomes the
534 offset + offset_increment * thread_number, where the thread number is a counter
535 that starts at 0 and is incremented for each job. This option is useful if
536 there are several jobs which are intended to operate on a file in parallel in
537 disjoint segments, with even spacing between the starting points.
540 How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If
541 0, don't sync. Default: 0.
543 .BI fdatasync \fR=\fPint
544 Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the
545 data parts of the file. Default: 0.
547 .BI sync_file_range \fR=\fPstr:int
548 Use sync_file_range() for every \fRval\fP number of write operations. Fio will
549 track range of writes that have happened since the last sync_file_range() call.
550 \fRstr\fP can currently be one or more of:
554 SYNC_FILE_RANGE_WAIT_BEFORE
557 SYNC_FILE_RANGE_WRITE
560 SYNC_FILE_RANGE_WRITE
564 So if you do sync_file_range=wait_before,write:8, fio would use
565 \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes.
566 Also see the sync_file_range(2) man page. This option is Linux specific.
568 .BI overwrite \fR=\fPbool
569 If writing, setup the file first and do overwrites. Default: false.
571 .BI end_fsync \fR=\fPbool
572 Sync file contents when a write stage has completed. Default: false.
574 .BI fsync_on_close \fR=\fPbool
575 If true, sync file contents on close. This differs from \fBend_fsync\fR in that
576 it will happen on every close, not just at the end of the job. Default: false.
578 .BI rwmixread \fR=\fPint
579 Percentage of a mixed workload that should be reads. Default: 50.
581 .BI rwmixwrite \fR=\fPint
582 Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and
583 \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two
584 overrides the first. This may interfere with a given rate setting, if fio is
585 asked to limit reads or writes to a certain rate. If that is the case, then
586 the distribution may be skewed. Default: 50.
588 .BI random_distribution \fR=\fPstr:float
589 By default, fio will use a completely uniform random distribution when asked
590 to perform random IO. Sometimes it is useful to skew the distribution in
591 specific ways, ensuring that some parts of the data is more hot than others.
592 Fio includes the following distribution models:
596 Uniform random distribution
606 When using a zipf or pareto distribution, an input value is also needed to
607 define the access pattern. For zipf, this is the zipf theta. For pareto,
608 it's the pareto power. Fio includes a test program, genzipf, that can be
609 used visualize what the given input values will yield in terms of hit rates.
610 If you wanted to use zipf with a theta of 1.2, you would use
611 random_distribution=zipf:1.2 as the option. If a non-uniform model is used,
612 fio will disable use of the random map.
615 Normally \fBfio\fR will cover every block of the file when doing random I/O. If
616 this parameter is given, a new offset will be chosen without looking at past
617 I/O history. This parameter is mutually exclusive with \fBverify\fR.
619 .BI softrandommap \fR=\fPbool
620 See \fBnorandommap\fR. If fio runs with the random block map enabled and it
621 fails to allocate the map, if this option is set it will continue without a
622 random block map. As coverage will not be as complete as with random maps, this
623 option is disabled by default.
625 .BI random_generator \fR=\fPstr
626 Fio supports the following engines for generating IO offsets for random IO:
630 Strong 2^88 cycle random number generator
633 Linear feedback shift register generator
637 Tausworthe is a strong random number generator, but it requires tracking on the
638 side if we want to ensure that blocks are only read or written once. LFSR
639 guarantees that we never generate the same offset twice, and it's also less
640 computationally expensive. It's not a true random generator, however, though
641 for IO purposes it's typically good enough. LFSR only works with single block
642 sizes, not with workloads that use multiple block sizes. If used with such a
643 workload, fio may read or write some blocks multiple times.
646 Run job with given nice value. See \fInice\fR\|(2).
649 Set I/O priority value of this job between 0 (highest) and 7 (lowest). See
652 .BI prioclass \fR=\fPint
653 Set I/O priority class. See \fIionice\fR\|(1).
655 .BI thinktime \fR=\fPint
656 Stall job for given number of microseconds between issuing I/Os.
658 .BI thinktime_spin \fR=\fPint
659 Pretend to spend CPU time for given number of microseconds, sleeping the rest
660 of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set.
662 .BI thinktime_blocks \fR=\fPint
663 Number of blocks to issue before waiting \fBthinktime\fR microseconds.
667 Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix
668 rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each,
669 or you can specify read and writes separately. Using \fBrate\fR=1m,500k would
670 limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes
671 can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only
672 limit writes (to 500KB/sec), the latter will only limit reads.
674 .BI ratemin \fR=\fPint
675 Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth.
676 Failing to meet this requirement will cause the job to exit. The same format
677 as \fBrate\fR is used for read vs write separation.
679 .BI rate_iops \fR=\fPint
680 Cap the bandwidth to this number of IOPS. Basically the same as rate, just
681 specified independently of bandwidth. The same format as \fBrate\fR is used for
682 read vs write seperation. If \fBblocksize\fR is a range, the smallest block
683 size is used as the metric.
685 .BI rate_iops_min \fR=\fPint
686 If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR
687 is used for read vs write seperation.
689 .BI ratecycle \fR=\fPint
690 Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of
691 milliseconds. Default: 1000ms.
693 .BI max_latency \fR=\fPint
694 If set, fio will exit the job if it exceeds this maximum latency. It will exit
697 .BI cpumask \fR=\fPint
698 Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job
699 may run on. See \fBsched_setaffinity\fR\|(2).
701 .BI cpus_allowed \fR=\fPstr
702 Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers.
704 .BI numa_cpu_nodes \fR=\fPstr
705 Set this job running on spcified NUMA nodes' CPUs. The arguments allow
706 comma delimited list of cpu numbers, A-B ranges, or 'all'.
708 .BI numa_mem_policy \fR=\fPstr
709 Set this job's memory policy and corresponding NUMA nodes. Format of
713 .B <mode>[:<nodelist>]
716 is one of the following memory policy:
718 .B default, prefer, bind, interleave, local
721 For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is
722 needed to be specified. For \fBprefer\fR, only one node is
723 allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows
724 comma delimited list of numbers, A-B ranges, or 'all'.
726 .BI startdelay \fR=\fPint
727 Delay start of job for the specified number of seconds.
729 .BI runtime \fR=\fPint
730 Terminate processing after the specified number of seconds.
733 If given, run for the specified \fBruntime\fR duration even if the files are
734 completely read or written. The same workload will be repeated as many times
735 as \fBruntime\fR allows.
737 .BI ramp_time \fR=\fPint
738 If set, fio will run the specified workload for this amount of time before
739 logging any performance numbers. Useful for letting performance settle before
740 logging results, thus minimizing the runtime required for stable results. Note
741 that the \fBramp_time\fR is considered lead in time for a job, thus it will
742 increase the total runtime if a special timeout or runtime is specified.
744 .BI invalidate \fR=\fPbool
745 Invalidate buffer-cache for the file prior to starting I/O. Default: true.
748 Use synchronous I/O for buffered writes. For the majority of I/O engines,
749 this means using O_SYNC. Default: false.
751 .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr
752 Allocation method for I/O unit buffer. Allowed values are:
757 Allocate memory with \fImalloc\fR\|(3).
760 Use shared memory buffers allocated through \fIshmget\fR\|(2).
763 Same as \fBshm\fR, but use huge pages as backing.
766 Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename
767 is given after the option in the format `:\fIfile\fR'.
770 Same as \fBmmap\fR, but use huge files as backing.
773 The amount of memory allocated is the maximum allowed \fBblocksize\fR for the
774 job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work,
775 the system must have free huge pages allocated. \fBmmaphuge\fR also needs to
776 have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux,
777 huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR
778 and the documentation for that. Normally you just need to echo an appropriate
779 number, eg echoing 8 will ensure that the OS has 8 huge pages ready for
783 .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint
784 This indiciates the memory alignment of the IO memory buffers. Note that the
785 given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR
786 the alignment of the following buffers are given by the \fBbs\fR used. In
787 other words, if using a \fBbs\fR that is a multiple of the page sized in the
788 system, all buffers will be aligned to this value. If using a \fBbs\fR that
789 is not page aligned, the alignment of subsequent IO memory buffers is the
790 sum of the \fBiomem_align\fR and \fBbs\fR used.
792 .BI hugepage\-size \fR=\fPint
793 Defines the size of a huge page. Must be at least equal to the system setting.
794 Should be a multiple of 1MB. Default: 4MB.
797 Terminate all jobs when one finishes. Default: wait for each job to finish.
799 .BI bwavgtime \fR=\fPint
800 Average bandwidth calculations over the given time in milliseconds. Default:
803 .BI iopsavgtime \fR=\fPint
804 Average IOPS calculations over the given time in milliseconds. Default:
807 .BI create_serialize \fR=\fPbool
808 If true, serialize file creation for the jobs. Default: true.
810 .BI create_fsync \fR=\fPbool
811 \fIfsync\fR\|(2) data file after creation. Default: true.
813 .BI create_on_open \fR=\fPbool
814 If true, the files are not created until they are opened for IO by the job.
816 .BI create_only \fR=\fPbool
817 If true, fio will only run the setup phase of the job. If files need to be
818 laid out or updated on disk, only that will be done. The actual job contents
821 .BI pre_read \fR=\fPbool
822 If this is given, files will be pre-read into memory before starting the given
823 IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is
824 pointless to pre-read and then drop the cache. This will only work for IO
825 engines that are seekable, since they allow you to read the same data
826 multiple times. Thus it will not work on eg network or splice IO.
828 .BI unlink \fR=\fPbool
829 Unlink job files when done. Default: false.
832 Specifies the number of iterations (runs of the same workload) of this job.
835 .BI do_verify \fR=\fPbool
836 Run the verify phase after a write phase. Only valid if \fBverify\fR is set.
839 .BI verify \fR=\fPstr
840 Method of verifying file contents after each iteration of the job. Allowed
845 .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1
846 Store appropriate checksum in the header of each block. crc32c-intel is
847 hardware accelerated SSE4.2 driven, falls back to regular crc32c if
848 not supported by the system.
851 Write extra information about each I/O (timestamp, block number, etc.). The
852 block number is verified. See \fBverify_pattern\fR as well.
855 Pretend to verify. Used for testing internals.
858 This option can be used for repeated burn-in tests of a system to make sure
859 that the written data is also correctly read back. If the data direction given
860 is a read or random read, fio will assume that it should verify a previously
861 written file. If the data direction includes any form of write, the verify will
862 be of the newly written data.
865 .BI verify_sort \fR=\fPbool
866 If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to
867 read them back in a sorted manner. Default: true.
869 .BI verify_offset \fR=\fPint
870 Swap the verification header with data somewhere else in the block before
871 writing. It is swapped back before verifying.
873 .BI verify_interval \fR=\fPint
874 Write the verification header for this number of bytes, which should divide
875 \fBblocksize\fR. Default: \fBblocksize\fR.
877 .BI verify_pattern \fR=\fPstr
878 If set, fio will fill the io buffers with this pattern. Fio defaults to filling
879 with totally random bytes, but sometimes it's interesting to fill with a known
880 pattern for io verification purposes. Depending on the width of the pattern,
881 fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a
882 decimal or a hex number). The verify_pattern if larger than a 32-bit quantity
883 has to be a hex number that starts with either "0x" or "0X". Use with
886 .BI verify_fatal \fR=\fPbool
887 If true, exit the job on the first observed verification failure. Default:
890 .BI verify_dump \fR=\fPbool
891 If set, dump the contents of both the original data block and the data block we
892 read off disk to files. This allows later analysis to inspect just what kind of
893 data corruption occurred. Off by default.
895 .BI verify_async \fR=\fPint
896 Fio will normally verify IO inline from the submitting thread. This option
897 takes an integer describing how many async offload threads to create for IO
898 verification instead, causing fio to offload the duty of verifying IO contents
899 to one or more separate threads. If using this offload option, even sync IO
900 engines can benefit from using an \fBiodepth\fR setting higher than 1, as it
901 allows them to have IO in flight while verifies are running.
903 .BI verify_async_cpus \fR=\fPstr
904 Tell fio to set the given CPU affinity on the async IO verification threads.
905 See \fBcpus_allowed\fP for the format used.
907 .BI verify_backlog \fR=\fPint
908 Fio will normally verify the written contents of a job that utilizes verify
909 once that job has completed. In other words, everything is written then
910 everything is read back and verified. You may want to verify continually
911 instead for a variety of reasons. Fio stores the meta data associated with an
912 IO block in memory, so for large verify workloads, quite a bit of memory would
913 be used up holding this meta data. If this option is enabled, fio will write
914 only N blocks before verifying these blocks.
916 .BI verify_backlog_batch \fR=\fPint
917 Control how many blocks fio will verify if verify_backlog is set. If not set,
918 will default to the value of \fBverify_backlog\fR (meaning the entire queue is
919 read back and verified). If \fBverify_backlog_batch\fR is less than
920 \fBverify_backlog\fR then not all blocks will be verified, if
921 \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks
922 will be verified more than once.
924 .B stonewall "\fR,\fP wait_for_previous"
925 Wait for preceding jobs in the job file to exit before starting this one.
926 \fBstonewall\fR implies \fBnew_group\fR.
929 Start a new reporting group. If not given, all jobs in a file will be part
930 of the same reporting group, unless separated by a stonewall.
932 .BI numjobs \fR=\fPint
933 Number of clones (processes/threads performing the same workload) of this job.
937 If set, display per-group reports instead of per-job when \fBnumjobs\fR is
941 Use threads created with \fBpthread_create\fR\|(3) instead of processes created
942 with \fBfork\fR\|(2).
944 .BI zonesize \fR=\fPint
945 Divide file into zones of the specified size in bytes. See \fBzoneskip\fR.
947 .BI zoneskip \fR=\fPint
948 Skip the specified number of bytes when \fBzonesize\fR bytes of data have been
951 .BI write_iolog \fR=\fPstr
952 Write the issued I/O patterns to the specified file. Specify a separate file
953 for each job, otherwise the iologs will be interspersed and the file may be
956 .BI read_iolog \fR=\fPstr
957 Replay the I/O patterns contained in the specified file generated by
958 \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file.
960 .BI replay_no_stall \fR=\fPint
961 While replaying I/O patterns using \fBread_iolog\fR the default behavior
962 attempts to respect timing information between I/Os. Enabling
963 \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while
964 still respecting ordering.
966 .BI replay_redirect \fR=\fPstr
967 While replaying I/O patterns using \fBread_iolog\fR the default behavior
968 is to replay the IOPS onto the major/minor device that each IOP was recorded
969 from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the
970 single specified device regardless of the device it was recorded from.
972 .BI write_bw_log \fR=\fPstr
973 If given, write a bandwidth log of the jobs in this job file. Can be used to
974 store data of the bandwidth of the jobs in their lifetime. The included
975 fio_generate_plots script uses gnuplot to turn these text files into nice
976 graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this
977 option, the postfix is _bw.log.
979 .BI write_lat_log \fR=\fPstr
980 Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no
981 filename is given with this option, the default filename of "jobname_type.log"
982 is used. Even if the filename is given, fio will still append the type of log.
984 .BI write_iops_log \fR=\fPstr
985 Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this
986 option, the default filename of "jobname_type.log" is used. Even if the
987 filename is given, fio will still append the type of log.
989 .BI log_avg_msec \fR=\fPint
990 By default, fio will log an entry in the iops, latency, or bw log for every
991 IO that completes. When writing to the disk log, that can quickly grow to a
992 very large size. Setting this option makes fio average the each log entry
993 over the specified period of time, reducing the resolution of the log.
996 .BI disable_lat \fR=\fPbool
997 Disable measurements of total latency numbers. Useful only for cutting
998 back the number of calls to gettimeofday, as that does impact performance at
999 really high IOPS rates. Note that to really get rid of a large amount of these
1000 calls, this option must be used with disable_slat and disable_bw as well.
1002 .BI disable_clat \fR=\fPbool
1003 Disable measurements of completion latency numbers. See \fBdisable_lat\fR.
1005 .BI disable_slat \fR=\fPbool
1006 Disable measurements of submission latency numbers. See \fBdisable_lat\fR.
1008 .BI disable_bw_measurement \fR=\fPbool
1009 Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR.
1011 .BI lockmem \fR=\fPint
1012 Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to
1013 simulate a smaller amount of memory.
1015 .BI exec_prerun \fR=\fPstr
1016 Before running the job, execute the specified command with \fBsystem\fR\|(3).
1018 .BI exec_postrun \fR=\fPstr
1019 Same as \fBexec_prerun\fR, but the command is executed after the job completes.
1021 .BI ioscheduler \fR=\fPstr
1022 Attempt to switch the device hosting the file to the specified I/O scheduler.
1024 .BI cpuload \fR=\fPint
1025 If the job is a CPU cycle-eater, attempt to use the specified percentage of
1028 .BI cpuchunks \fR=\fPint
1029 If the job is a CPU cycle-eater, split the load into cycles of the
1030 given time in milliseconds.
1032 .BI disk_util \fR=\fPbool
1033 Generate disk utilization statistics if the platform supports it. Default: true.
1035 .BI clocksource \fR=\fPstr
1036 Use the given clocksource as the base of timing. The supported options are:
1046 Internal CPU clock source
1050 \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast
1051 (and fio is heavy on time calls). Fio will automatically use this clocksource
1052 if it's supported and considered reliable on the system it is running on,
1053 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
1054 means supporting TSC Invariant.
1056 .BI gtod_reduce \fR=\fPbool
1057 Enable all of the gettimeofday() reducing options (disable_clat, disable_slat,
1058 disable_bw) plus reduce precision of the timeout somewhat to really shrink the
1059 gettimeofday() call count. With this option enabled, we only do about 0.4% of
1060 the gtod() calls we would have done if all time keeping was enabled.
1062 .BI gtod_cpu \fR=\fPint
1063 Sometimes it's cheaper to dedicate a single thread of execution to just getting
1064 the current time. Fio (and databases, for instance) are very intensive on
1065 gettimeofday() calls. With this option, you can set one CPU aside for doing
1066 nothing but logging current time to a shared memory location. Then the other
1067 threads/processes that run IO workloads need only copy that segment, instead of
1068 entering the kernel with a gettimeofday() call. The CPU set aside for doing
1069 these time calls will be excluded from other uses. Fio will manually clear it
1070 from the CPU mask of other jobs.
1072 .BI ignore_error \fR=\fPstr
1073 Sometimes you want to ignore some errors during test in that case you can specify
1074 error list for each error type.
1076 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1078 errors for given error type is separated with ':'.
1079 Error may be symbol ('ENOSPC', 'ENOMEM') or an integer.
1081 Example: ignore_error=EAGAIN,ENOSPC:122 .
1083 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE.
1085 .BI error_dump \fR=\fPbool
1086 If set dump every error even if it is non fatal, true by default. If disabled
1087 only fatal error will be dumped
1089 .BI cgroup \fR=\fPstr
1090 Add job to this control group. If it doesn't exist, it will be created.
1091 The system must have a mounted cgroup blkio mount point for this to work. If
1092 your system doesn't have it mounted, you can do so with:
1094 # mount \-t cgroup \-o blkio none /cgroup
1096 .BI cgroup_weight \fR=\fPint
1097 Set the weight of the cgroup to this value. See the documentation that comes
1098 with the kernel, allowed values are in the range of 100..1000.
1100 .BI cgroup_nodelete \fR=\fPbool
1101 Normally fio will delete the cgroups it has created after the job completion.
1102 To override this behavior and to leave cgroups around after the job completion,
1103 set cgroup_nodelete=1. This can be useful if one wants to inspect various
1104 cgroup files after job completion. Default: false
1107 Instead of running as the invoking user, set the user ID to this value before
1108 the thread/process does any work.
1111 Set group ID, see \fBuid\fR.
1113 .BI flow_id \fR=\fPint
1114 The ID of the flow. If not specified, it defaults to being a global flow. See
1118 Weight in token-based flow control. If this value is used, then there is a
1119 \fBflow counter\fR which is used to regulate the proportion of activity between
1120 two or more jobs. fio attempts to keep this flow counter near zero. The
1121 \fBflow\fR parameter stands for how much should be added or subtracted to the
1122 flow counter on each iteration of the main I/O loop. That is, if one job has
1123 \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly
1124 1:8 ratio in how much one runs vs the other.
1126 .BI flow_watermark \fR=\fPint
1127 The maximum value that the absolute value of the flow counter is allowed to
1128 reach before the job must wait for a lower value of the counter.
1130 .BI flow_sleep \fR=\fPint
1131 The period of time, in microseconds, to wait after the flow watermark has been
1132 exceeded before retrying operations
1134 .BI clat_percentiles \fR=\fPbool
1135 Enable the reporting of percentiles of completion latencies.
1137 .BI percentile_list \fR=\fPfloat_list
1138 Overwrite the default list of percentiles for completion
1139 latencies. Each number is a floating number in the range (0,100], and
1140 the maximum length of the list is 20. Use ':' to separate the
1141 numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to
1142 report the values of completion latency below which 99.5% and 99.9% of
1143 the observed latencies fell, respectively.
1144 .SS "Ioengine Parameters List"
1145 Some parameters are only valid when a specific ioengine is in use. These are
1146 used identically to normal parameters, with the caveat that when used on the
1147 command line, the must come after the ioengine that defines them is selected.
1149 .BI (libaio)userspace_reap
1150 Normally, with the libaio engine in use, fio will use
1151 the io_getevents system call to reap newly returned events.
1152 With this flag turned on, the AIO ring will be read directly
1153 from user-space to reap events. The reaping mode is only
1154 enabled when polling for a minimum of 0 events (eg when
1155 iodepth_batch_complete=0).
1157 .BI (net,netsplice)hostname \fR=\fPstr
1158 The host name or IP address to use for TCP or UDP based IO.
1159 If the job is a TCP listener or UDP reader, the hostname is not
1160 used and must be omitted.
1162 .BI (net,netsplice)port \fR=\fPint
1163 The TCP or UDP port to bind to or connect to.
1165 .BI (net,netsplice)nodelay \fR=\fPbool
1166 Set TCP_NODELAY on TCP connections.
1168 .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr
1169 The network protocol to use. Accepted values are:
1174 Transmission control protocol
1177 User datagram protocol
1183 When the protocol is TCP or UDP, the port must also be given,
1184 as well as the hostname if the job is a TCP listener or UDP
1185 reader. For unix sockets, the normal filename option should be
1186 used and the port is invalid.
1189 .BI (net,netsplice)listen
1190 For TCP network connections, tell fio to listen for incoming
1191 connections rather than initiating an outgoing connection. The
1192 hostname must be omitted if this option is used.
1194 .BI (net, pingpong) \fR=\fPbool
1195 Normal a network writer will just continue writing data, and a network reader
1196 will just consume packages. If pingpong=1 is set, a writer will send its normal
1197 payload to the reader, then wait for the reader to send the same payload back.
1198 This allows fio to measure network latencies. The submission and completion
1199 latencies then measure local time spent sending or receiving, and the
1200 completion latency measures how long it took for the other end to receive and
1203 .BI (e4defrag,donorname) \fR=\fPstr
1204 File will be used as a block donor (swap extents between files)
1206 .BI (e4defrag,inplace) \fR=\fPint
1207 Configure donor file block allocation strategy
1210 Preallocate donor's file on init
1213 allocate space immidietly inside defragment event, and free right after event
1217 While running, \fBfio\fR will display the status of the created jobs. For
1221 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1224 The characters in the first set of brackets denote the current status of each
1225 threads. The possible values are:
1231 Setup but not started.
1237 Initialized, waiting.
1240 Running, doing sequential reads.
1243 Running, doing random reads.
1246 Running, doing sequential writes.
1249 Running, doing random writes.
1252 Running, doing mixed sequential reads/writes.
1255 Running, doing mixed random reads/writes.
1258 Running, currently waiting for \fBfsync\fR\|(2).
1261 Running, verifying written data.
1264 Exited, not reaped by main thread.
1267 Exited, thread reaped.
1271 The second set of brackets shows the estimated completion percentage of
1272 the current group. The third set shows the read and write I/O rate,
1273 respectively. Finally, the estimated run time of the job is displayed.
1275 When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data
1276 for each thread, each group of threads, and each disk, in that order.
1278 Per-thread statistics first show the threads client number, group-id, and
1279 error code. The remaining figures are as follows:
1283 Number of megabytes of I/O performed.
1286 Average data rate (bandwidth).
1292 Submission latency minimum, maximum, average and standard deviation. This is
1293 the time it took to submit the I/O.
1296 Completion latency minimum, maximum, average and standard deviation. This
1297 is the time between submission and completion.
1300 Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average
1301 and standard deviation.
1304 CPU usage statistics. Includes user and system time, number of context switches
1305 this thread went through and number of major and minor page faults.
1308 Distribution of I/O depths. Each depth includes everything less than (or equal)
1309 to it, but greater than the previous depth.
1312 Number of read/write requests issued, and number of short read/write requests.
1315 Distribution of I/O completion latencies. The numbers follow the same pattern
1319 The group statistics show:
1324 Number of megabytes I/O performed.
1327 Aggregate bandwidth of threads in the group.
1330 Minimum average bandwidth a thread saw.
1333 Maximum average bandwidth a thread saw.
1336 Shortest runtime of threads in the group.
1339 Longest runtime of threads in the group.
1343 Finally, disk statistics are printed with reads first:
1348 Number of I/Os performed by all groups.
1351 Number of merges in the I/O scheduler.
1354 Number of ticks we kept the disk busy.
1357 Total time spent in the disk queue.
1364 It is also possible to get fio to dump the current output while it is
1365 running, without terminating the job. To do that, send fio the \fBUSR1\fR
1368 If the \fB\-\-minimal\fR option is given, the results will be printed in a
1369 semicolon-delimited format suitable for scripted use - a job description
1370 (if provided) follows on a new line. Note that the first
1371 number in the line is the version number. If the output has to be changed
1372 for some reason, this number will be incremented by 1 to signify that
1373 change. The fields are:
1376 .B terse version, fio version, jobname, groupid, error
1380 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1384 .B min, max, mean, standard deviation
1388 .B min, max, mean, standard deviation
1390 Completion latency percentiles (20 fields):
1392 .B Xth percentile=usec
1396 .B min, max, mean, standard deviation
1400 .B min, max, aggregate percentage of total, mean, standard deviation
1406 .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP
1410 .B min, max, mean, standard deviation
1414 .B min, max, mean, standard deviation
1416 Completion latency percentiles (20 fields):
1418 .B Xth percentile=usec
1422 .B min, max, mean, standard deviation
1426 .B min, max, aggregate percentage of total, mean, standard deviation
1432 .B user, system, context switches, major page faults, minor page faults
1435 IO depth distribution:
1437 .B <=1, 2, 4, 8, 16, 32, >=64
1440 IO latency distribution:
1444 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1448 .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1452 Disk utilization (1 for each disk used):
1454 .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
1457 Error Info (dependent on continue_on_error, default off):
1459 .B total # errors, first error code
1462 .B text description (if provided in config - appears on newline)
1465 Normally you would run fio as a stand-alone application on the machine
1466 where the IO workload should be generated. However, it is also possible to
1467 run the frontend and backend of fio separately. This makes it possible to
1468 have a fio server running on the machine(s) where the IO workload should
1469 be running, while controlling it from another machine.
1471 To start the server, you would do:
1473 \fBfio \-\-server=args\fR
1475 on that machine, where args defines what fio listens to. The arguments
1476 are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4)
1477 for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain
1478 socket. 'hostname' is either a hostname or IP address, and 'port' is the port to
1479 listen to (only valid for TCP/IP, not a local socket). Some examples:
1483 Start a fio server, listening on all interfaces on the default port (8765).
1485 2) fio \-\-server=ip:hostname,4444
1487 Start a fio server, listening on IP belonging to hostname and on port 4444.
1489 3) fio \-\-server=ip6:::1,4444
1491 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
1493 4) fio \-\-server=,4444
1495 Start a fio server, listening on all interfaces on port 4444.
1497 5) fio \-\-server=1.2.3.4
1499 Start a fio server, listening on IP 1.2.3.4 on the default port.
1501 6) fio \-\-server=sock:/tmp/fio.sock
1503 Start a fio server, listening on the local socket /tmp/fio.sock.
1505 When a server is running, you can connect to it from a client. The client
1508 fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)>
1510 where \-\-local-args are arguments that are local to the client where it is
1511 running, 'server' is the connect string, and \-\-remote-args and <job file(s)>
1512 are sent to the server. The 'server' string follows the same format as it
1513 does on the server side, to allow IP/hostname/socket and port strings.
1514 You can connect to multiple clients as well, to do that you could run:
1516 fio \-\-client=server2 \-\-client=server2 <job file(s)>
1520 was written by Jens Axboe <jens.axboe@oracle.com>,
1521 now Jens Axboe <jaxboe@fusionio.com>.
1523 This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based
1524 on documentation by Jens Axboe.
1525 .SH "REPORTING BUGS"
1526 Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>.
1529 For further documentation see \fBHOWTO\fR and \fBREADME\fR.
1531 Sample jobfiles are available in the \fBexamples\fR directory.
projects / fio.git / blob