8 5. Detailed list of parameters
12 9. CPU idleness profiling
14 1.0 Overview and history
15 ------------------------
16 fio was originally written to save me the hassle of writing special test
17 case programs when I wanted to test a specific workload, either for
18 performance reasons or to find/reproduce a bug. The process of writing
19 such a test app can be tiresome, especially if you have to do it often.
20 Hence I needed a tool that would be able to simulate a given io workload
21 without resorting to writing a tailored test case again and again.
23 A test work load is difficult to define, though. There can be any number
24 of processes or threads involved, and they can each be using their own
25 way of generating io. You could have someone dirtying large amounts of
26 memory in an memory mapped file, or maybe several threads issuing
27 reads using asynchronous io. fio needed to be flexible enough to
28 simulate both of these cases, and many more.
32 The first step in getting fio to simulate a desired io workload, is
33 writing a job file describing that specific setup. A job file may contain
34 any number of threads and/or files - the typical contents of the job file
35 is a global section defining shared parameters, and one or more job
36 sections describing the jobs involved. When run, fio parses this file
37 and sets everything up as described. If we break down a job from top to
38 bottom, it contains the following basic parameters:
40 IO type Defines the io pattern issued to the file(s).
41 We may only be reading sequentially from this
42 file(s), or we may be writing randomly. Or even
43 mixing reads and writes, sequentially or randomly.
45 Block size In how large chunks are we issuing io? This may be
46 a single value, or it may describe a range of
49 IO size How much data are we going to be reading/writing.
51 IO engine How do we issue io? We could be memory mapping the
52 file, we could be using regular read/write, we
53 could be using splice, async io, syslet, or even
56 IO depth If the io engine is async, how large a queuing
57 depth do we want to maintain?
59 IO type Should we be doing buffered io, or direct/raw io?
61 Num files How many files are we spreading the workload over.
63 Num threads How many threads or processes should we spread
66 The above are the basic parameters defined for a workload, in addition
67 there's a multitude of parameters that modify other aspects of how this
73 See the README file for command line parameters, there are only a few
76 Running fio is normally the easiest part - you just give it the job file
77 (or job files) as parameters:
81 and it will start doing what the job_file tells it to do. You can give
82 more than one job file on the command line, fio will serialize the running
83 of those files. Internally that is the same as using the 'stonewall'
84 parameter described the the parameter section.
86 If the job file contains only one job, you may as well just give the
87 parameters on the command line. The command line parameters are identical
88 to the job parameters, with a few extra that control global parameters
89 (see README). For example, for the job file parameter iodepth=2, the
90 mirror command line option would be --iodepth 2 or --iodepth=2. You can
91 also use the command line for giving more than one job entry. For each
92 --name option that fio sees, it will start a new job with that name.
93 Command line entries following a --name entry will apply to that job,
94 until there are no more entries or a new --name entry is seen. This is
95 similar to the job file options, where each option applies to the current
96 job until a new [] job entry is seen.
98 fio does not need to run as root, except if the files or devices specified
99 in the job section requires that. Some other options may also be restricted,
100 such as memory locking, io scheduler switching, and decreasing the nice value.
105 As previously described, fio accepts one or more job files describing
106 what it is supposed to do. The job file format is the classic ini file,
107 where the names enclosed in [] brackets define the job name. You are free
108 to use any ascii name you want, except 'global' which has special meaning.
109 A global section sets defaults for the jobs described in that file. A job
110 may override a global section parameter, and a job file may even have
111 several global sections if so desired. A job is only affected by a global
112 section residing above it. If the first character in a line is a ';' or a
113 '#', the entire line is discarded as a comment.
115 So let's look at a really simple job file that defines two processes, each
116 randomly reading from a 128MB file.
118 ; -- start job file --
129 As you can see, the job file sections themselves are empty as all the
130 described parameters are shared. As no filename= option is given, fio
131 makes up a filename for each of the jobs as it sees fit. On the command
132 line, this job would look as follows:
134 $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
137 Let's look at an example that has a number of processes writing randomly
140 ; -- start job file --
152 Here we have no global section, as we only have one job defined anyway.
153 We want to use async io here, with a depth of 4 for each file. We also
154 increased the buffer size used to 32KB and define numjobs to 4 to
155 fork 4 identical jobs. The result is 4 processes each randomly writing
156 to their own 64MB file. Instead of using the above job file, you could
157 have given the parameters on the command line. For this case, you would
160 $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
162 4.1 Environment variables
163 -------------------------
165 fio also supports environment variable expansion in job files. Any
166 substring of the form "${VARNAME}" as part of an option value (in other
167 words, on the right of the `='), will be expanded to the value of the
168 environment variable called VARNAME. If no such environment variable
169 is defined, or VARNAME is the empty string, the empty string will be
172 As an example, let's look at a sample fio invocation and job file:
174 $ SIZE=64m NUMJOBS=4 fio jobfile.fio
176 ; -- start job file --
183 This will expand to the following equivalent job file at runtime:
185 ; -- start job file --
192 fio ships with a few example job files, you can also look there for
195 4.2 Reserved keywords
196 ---------------------
198 Additionally, fio has a set of reserved keywords that will be replaced
199 internally with the appropriate value. Those keywords are:
201 $pagesize The architecture page size of the running system
202 $mb_memory Megabytes of total memory in the system
203 $ncpus Number of online available CPUs
205 These can be used on the command line or in the job file, and will be
206 automatically substituted with the current system values when the job
207 is run. Simple math is also supported on these keywords, so you can
208 perform actions like:
212 and get that properly expanded to 8 times the size of memory in the
216 5.0 Detailed list of parameters
217 -------------------------------
219 This section describes in details each parameter associated with a job.
220 Some parameters take an option of a given type, such as an integer or
221 a string. The following types are used:
223 str String. This is a sequence of alpha characters.
224 time Integer with possible time suffix. In seconds unless otherwise
225 specified, use eg 10m for 10 minutes. Accepts s/m/h for seconds,
227 int SI integer. A whole number value, which may contain a suffix
228 describing the base of the number. Accepted suffixes are k/m/g/t/p,
229 meaning kilo, mega, giga, tera, and peta. The suffix is not case
230 sensitive, and you may also include trailing 'b' (eg 'kb' is the same
231 as 'k'). So if you want to specify 4096, you could either write
232 out '4096' or just give 4k. The suffixes signify base 2 values, so
233 1024 is 1k and 1024k is 1m and so on, unless the suffix is explicitly
234 set to a base 10 value using 'kib', 'mib', 'gib', etc. If that is the
235 case, then 1000 is used as the multiplier. This can be handy for
236 disks, since manufacturers generally use base 10 values when listing
237 the capacity of a drive. If the option accepts an upper and lower
238 range, use a colon ':' or minus '-' to separate such values. May also
239 include a prefix to indicate numbers base. If 0x is used, the number
240 is assumed to be hexadecimal. See irange.
241 bool Boolean. Usually parsed as an integer, however only defined for
242 true and false (1 and 0).
243 irange Integer range with suffix. Allows value range to be given, such
244 as 1024-4096. A colon may also be used as the separator, eg
245 1k:4k. If the option allows two sets of ranges, they can be
246 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
248 float_list A list of floating numbers, separated by a ':' character.
250 With the above in mind, here follows the complete list of fio job
253 name=str ASCII name of the job. This may be used to override the
254 name printed by fio for this job. Otherwise the job
255 name is used. On the command line this parameter has the
256 special purpose of also signaling the start of a new
259 description=str Text description of the job. Doesn't do anything except
260 dump this text description when this job is run. It's
263 directory=str Prefix filenames with this directory. Used to place files
264 in a different location than "./".
266 filename=str Fio normally makes up a filename based on the job name,
267 thread number, and file number. If you want to share
268 files between threads in a job or several jobs, specify
269 a filename for each of them to override the default. If
270 the ioengine used is 'net', the filename is the host, port,
271 and protocol to use in the format of =host,port,protocol.
272 See ioengine=net for more. If the ioengine is file based, you
273 can specify a number of files by separating the names with a
274 ':' colon. So if you wanted a job to open /dev/sda and /dev/sdb
275 as the two working files, you would use
276 filename=/dev/sda:/dev/sdb. On Windows, disk devices are
277 accessed as \\.\PhysicalDrive0 for the first device,
278 \\.\PhysicalDrive1 for the second etc. Note: Windows and
279 FreeBSD prevent write access to areas of the disk containing
280 in-use data (e.g. filesystems).
281 If the wanted filename does need to include a colon, then
282 escape that with a '\' character. For instance, if the filename
283 is "/dev/dsk/foo@3,0:c", then you would use
284 filename="/dev/dsk/foo@3,0\:c". '-' is a reserved name, meaning
285 stdin or stdout. Which of the two depends on the read/write
289 If sharing multiple files between jobs, it is usually necessary
290 to have fio generate the exact names that you want. By default,
291 fio will name a file based on the default file format
292 specification of jobname.jobnumber.filenumber. With this
293 option, that can be customized. Fio will recognize and replace
294 the following keywords in this string:
297 The name of the worker thread or process.
300 The incremental number of the worker thread or
304 The incremental number of the file for that worker
307 To have dependent jobs share a set of files, this option can
308 be set to have fio generate filenames that are shared between
309 the two. For instance, if testfiles.$filenum is specified,
310 file number 4 for any job will be named testfiles.4. The
311 default of $jobname.$jobnum.$filenum will be used if
312 no other format specifier is given.
314 opendir=str Tell fio to recursively add any file it can find in this
315 directory and down the file system tree.
317 lockfile=str Fio defaults to not locking any files before it does
318 IO to them. If a file or file descriptor is shared, fio
319 can serialize IO to that file to make the end result
320 consistent. This is usual for emulating real workloads that
321 share files. The lock modes are:
323 none No locking. The default.
324 exclusive Only one thread/process may do IO,
325 excluding all others.
326 readwrite Read-write locking on the file. Many
327 readers may access the file at the
328 same time, but writes get exclusive
332 rw=str Type of io pattern. Accepted values are:
334 read Sequential reads
335 write Sequential writes
336 randwrite Random writes
337 randread Random reads
338 rw,readwrite Sequential mixed reads and writes
339 randrw Random mixed reads and writes
341 For the mixed io types, the default is to split them 50/50.
342 For certain types of io the result may still be skewed a bit,
343 since the speed may be different. It is possible to specify
344 a number of IO's to do before getting a new offset, this is
345 one by appending a ':<nr>' to the end of the string given.
346 For a random read, it would look like 'rw=randread:8' for
347 passing in an offset modifier with a value of 8. If the
348 suffix is used with a sequential IO pattern, then the value
349 specified will be added to the generated offset for each IO.
350 For instance, using rw=write:4k will skip 4k for every
351 write. It turns sequential IO into sequential IO with holes.
352 See the 'rw_sequencer' option.
354 rw_sequencer=str If an offset modifier is given by appending a number to
355 the rw=<str> line, then this option controls how that
356 number modifies the IO offset being generated. Accepted
359 sequential Generate sequential offset
360 identical Generate the same offset
362 'sequential' is only useful for random IO, where fio would
363 normally generate a new random offset for every IO. If you
364 append eg 8 to randread, you would get a new random offset for
365 every 8 IO's. The result would be a seek for only every 8
366 IO's, instead of for every IO. Use rw=randread:8 to specify
367 that. As sequential IO is already sequential, setting
368 'sequential' for that would not result in any differences.
369 'identical' behaves in a similar fashion, except it sends
370 the same offset 8 number of times before generating a new
373 kb_base=int The base unit for a kilobyte. The defacto base is 2^10, 1024.
374 Storage manufacturers like to use 10^3 or 1000 as a base
375 ten unit instead, for obvious reasons. Allow values are
376 1024 or 1000, with 1024 being the default.
378 unified_rw_reporting=bool Fio normally reports statistics on a per
379 data direction basis, meaning that read, write, and trim are
380 accounted and reported separately. If this option is set,
381 the fio will sum the results and report them as "mixed"
384 randrepeat=bool For random IO workloads, seed the generator in a predictable
385 way so that results are repeatable across repetitions.
387 use_os_rand=bool Fio can either use the random generator supplied by the OS
388 to generator random offsets, or it can use it's own internal
389 generator (based on Tausworthe). Default is to use the
390 internal generator, which is often of better quality and
393 fallocate=str Whether pre-allocation is performed when laying down files.
396 none Do not pre-allocate space
397 posix Pre-allocate via posix_fallocate()
398 keep Pre-allocate via fallocate() with
399 FALLOC_FL_KEEP_SIZE set
400 0 Backward-compatible alias for 'none'
401 1 Backward-compatible alias for 'posix'
403 May not be available on all supported platforms. 'keep' is only
404 available on Linux.If using ZFS on Solaris this must be set to
405 'none' because ZFS doesn't support it. Default: 'posix'.
407 fadvise_hint=bool By default, fio will use fadvise() to advise the kernel
408 on what IO patterns it is likely to issue. Sometimes you
409 want to test specific IO patterns without telling the
410 kernel about it, in which case you can disable this option.
411 If set, fio will use POSIX_FADV_SEQUENTIAL for sequential
412 IO and POSIX_FADV_RANDOM for random IO.
414 size=int The total size of file io for this job. Fio will run until
415 this many bytes has been transferred, unless runtime is
416 limited by other options (such as 'runtime', for instance).
417 Unless specific nrfiles and filesize options are given,
418 fio will divide this size between the available files
419 specified by the job. If not set, fio will use the full
420 size of the given files or devices. If the the files
421 do not exist, size must be given. It is also possible to
422 give size as a percentage between 1 and 100. If size=20%
423 is given, fio will use 20% of the full size of the given
426 filesize=int Individual file sizes. May be a range, in which case fio
427 will select sizes for files at random within the given range
428 and limited to 'size' in total (if that is given). If not
429 given, each created file is the same size.
432 fill_fs=bool Sets size to something really large and waits for ENOSPC (no
433 space left on device) as the terminating condition. Only makes
434 sense with sequential write. For a read workload, the mount
435 point will be filled first then IO started on the result. This
436 option doesn't make sense if operating on a raw device node,
437 since the size of that is already known by the file system.
438 Additionally, writing beyond end-of-device will not return
442 bs=int The block size used for the io units. Defaults to 4k. Values
443 can be given for both read and writes. If a single int is
444 given, it will apply to both. If a second int is specified
445 after a comma, it will apply to writes only. In other words,
446 the format is either bs=read_and_write or bs=read,write,trim.
447 bs=4k,8k will thus use 4k blocks for reads, 8k blocks for
448 writes, and 8k for trims. You can terminate the list with
449 a trailing comma. bs=4k,8k, would use the default value for
450 trims.. If you only wish to set the write size, you
451 can do so by passing an empty read size - bs=,8k will set
452 8k for writes and leave the read default value.
455 ba=int At what boundary to align random IO offsets. Defaults to
456 the same as 'blocksize' the minimum blocksize given.
457 Minimum alignment is typically 512b for using direct IO,
458 though it usually depends on the hardware block size. This
459 option is mutually exclusive with using a random map for
460 files, so it will turn off that option.
462 blocksize_range=irange
463 bsrange=irange Instead of giving a single block size, specify a range
464 and fio will mix the issued io block sizes. The issued
465 io unit will always be a multiple of the minimum value
466 given (also see bs_unaligned). Applies to both reads and
467 writes, however a second range can be given after a comma.
470 bssplit=str Sometimes you want even finer grained control of the
471 block sizes issued, not just an even split between them.
472 This option allows you to weight various block sizes,
473 so that you are able to define a specific amount of
474 block sizes issued. The format for this option is:
476 bssplit=blocksize/percentage:blocksize/percentage
478 for as many block sizes as needed. So if you want to define
479 a workload that has 50% 64k blocks, 10% 4k blocks, and
480 40% 32k blocks, you would write:
482 bssplit=4k/10:64k/50:32k/40
484 Ordering does not matter. If the percentage is left blank,
485 fio will fill in the remaining values evenly. So a bssplit
486 option like this one:
488 bssplit=4k/50:1k/:32k/
490 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages
491 always add up to 100, if bssplit is given a range that adds
492 up to more, it will error out.
494 bssplit also supports giving separate splits to reads and
495 writes. The format is identical to what bs= accepts. You
496 have to separate the read and write parts with a comma. So
497 if you want a workload that has 50% 2k reads and 50% 4k reads,
498 while having 90% 4k writes and 10% 8k writes, you would
501 bssplit=2k/50:4k/50,4k/90,8k/10
504 bs_unaligned If this option is given, any byte size value within bsrange
505 may be used as a block range. This typically wont work with
506 direct IO, as that normally requires sector alignment.
508 bs_is_seq_rand If this option is set, fio will use the normal read,write
509 blocksize settings as sequential,random instead. Any random
510 read or write will use the WRITE blocksize settings, and any
511 sequential read or write will use the READ blocksize setting.
513 zero_buffers If this option is given, fio will init the IO buffers to
514 all zeroes. The default is to fill them with random data.
516 refill_buffers If this option is given, fio will refill the IO buffers
517 on every submit. The default is to only fill it at init
518 time and reuse that data. Only makes sense if zero_buffers
519 isn't specified, naturally. If data verification is enabled,
520 refill_buffers is also automatically enabled.
522 scramble_buffers=bool If refill_buffers is too costly and the target is
523 using data deduplication, then setting this option will
524 slightly modify the IO buffer contents to defeat normal
525 de-dupe attempts. This is not enough to defeat more clever
526 block compression attempts, but it will stop naive dedupe of
527 blocks. Default: true.
529 buffer_compress_percentage=int If this is set, then fio will attempt to
530 provide IO buffer content (on WRITEs) that compress to
531 the specified level. Fio does this by providing a mix of
532 random data and zeroes. Note that this is per block size
533 unit, for file/disk wide compression level that matches
534 this setting, you'll also want to set refill_buffers.
536 buffer_compress_chunk=int See buffer_compress_percentage. This
537 setting allows fio to manage how big the ranges of random
538 data and zeroed data is. Without this set, fio will
539 provide buffer_compress_percentage of blocksize random
540 data, followed by the remaining zeroed. With this set
541 to some chunk size smaller than the block size, fio can
542 alternate random and zeroed data throughout the IO
545 nrfiles=int Number of files to use for this job. Defaults to 1.
547 openfiles=int Number of files to keep open at the same time. Defaults to
548 the same as nrfiles, can be set smaller to limit the number
551 file_service_type=str Defines how fio decides which file from a job to
552 service next. The following types are defined:
554 random Just choose a file at random.
556 roundrobin Round robin over open files. This
559 sequential Finish one file before moving on to
560 the next. Multiple files can still be
561 open depending on 'openfiles'.
563 The string can have a number appended, indicating how
564 often to switch to a new file. So if option random:4 is
565 given, fio will switch to a new random file after 4 ios
568 ioengine=str Defines how the job issues io to the file. The following
571 sync Basic read(2) or write(2) io. lseek(2) is
572 used to position the io location.
574 psync Basic pread(2) or pwrite(2) io.
576 vsync Basic readv(2) or writev(2) IO.
578 psyncv Basic preadv(2) or pwritev(2) IO.
580 libaio Linux native asynchronous io. Note that Linux
581 may only support queued behaviour with
582 non-buffered IO (set direct=1 or buffered=0).
583 This engine defines engine specific options.
585 posixaio glibc posix asynchronous io.
587 solarisaio Solaris native asynchronous io.
589 windowsaio Windows native asynchronous io.
591 mmap File is memory mapped and data copied
592 to/from using memcpy(3).
594 splice splice(2) is used to transfer the data and
595 vmsplice(2) to transfer data from user
598 syslet-rw Use the syslet system calls to make
599 regular read/write async.
601 sg SCSI generic sg v3 io. May either be
602 synchronous using the SG_IO ioctl, or if
603 the target is an sg character device
604 we use read(2) and write(2) for asynchronous
607 null Doesn't transfer any data, just pretends
608 to. This is mainly used to exercise fio
609 itself and for debugging/testing purposes.
611 net Transfer over the network to given host:port.
612 Depending on the protocol used, the hostname,
613 port, listen and filename options are used to
614 specify what sort of connection to make, while
615 the protocol option determines which protocol
617 This engine defines engine specific options.
619 netsplice Like net, but uses splice/vmsplice to
620 map data and send/receive.
621 This engine defines engine specific options.
623 cpuio Doesn't transfer any data, but burns CPU
624 cycles according to the cpuload= and
625 cpucycle= options. Setting cpuload=85
626 will cause that job to do nothing but burn
627 85% of the CPU. In case of SMP machines,
628 use numjobs=<no_of_cpu> to get desired CPU
629 usage, as the cpuload only loads a single
630 CPU at the desired rate.
632 guasi The GUASI IO engine is the Generic Userspace
633 Asyncronous Syscall Interface approach
636 http://www.xmailserver.org/guasi-lib.html
638 for more info on GUASI.
640 rdma The RDMA I/O engine supports both RDMA
641 memory semantics (RDMA_WRITE/RDMA_READ) and
642 channel semantics (Send/Recv) for the
643 InfiniBand, RoCE and iWARP protocols.
645 falloc IO engine that does regular fallocate to
646 simulate data transfer as fio ioengine.
647 DDIR_READ does fallocate(,mode = keep_size,)
648 DDIR_WRITE does fallocate(,mode = 0)
649 DDIR_TRIM does fallocate(,mode = punch_hole)
651 e4defrag IO engine that does regular EXT4_IOC_MOVE_EXT
652 ioctls to simulate defragment activity in
653 request to DDIR_WRITE event
655 external Prefix to specify loading an external
656 IO engine object file. Append the engine
657 filename, eg ioengine=external:/tmp/foo.o
658 to load ioengine foo.o in /tmp.
660 iodepth=int This defines how many io units to keep in flight against
661 the file. The default is 1 for each file defined in this
662 job, can be overridden with a larger value for higher
663 concurrency. Note that increasing iodepth beyond 1 will not
664 affect synchronous ioengines (except for small degress when
665 verify_async is in use). Even async engines may impose OS
666 restrictions causing the desired depth not to be achieved.
667 This may happen on Linux when using libaio and not setting
668 direct=1, since buffered IO is not async on that OS. Keep an
669 eye on the IO depth distribution in the fio output to verify
670 that the achieved depth is as expected. Default: 1.
672 iodepth_batch_submit=int
673 iodepth_batch=int This defines how many pieces of IO to submit at once.
674 It defaults to 1 which means that we submit each IO
675 as soon as it is available, but can be raised to submit
676 bigger batches of IO at the time.
678 iodepth_batch_complete=int This defines how many pieces of IO to retrieve
679 at once. It defaults to 1 which means that we'll ask
680 for a minimum of 1 IO in the retrieval process from
681 the kernel. The IO retrieval will go on until we
682 hit the limit set by iodepth_low. If this variable is
683 set to 0, then fio will always check for completed
684 events before queuing more IO. This helps reduce
685 IO latency, at the cost of more retrieval system calls.
687 iodepth_low=int The low water mark indicating when to start filling
688 the queue again. Defaults to the same as iodepth, meaning
689 that fio will attempt to keep the queue full at all times.
690 If iodepth is set to eg 16 and iodepth_low is set to 4, then
691 after fio has filled the queue of 16 requests, it will let
692 the depth drain down to 4 before starting to fill it again.
694 direct=bool If value is true, use non-buffered io. This is usually
695 O_DIRECT. Note that ZFS on Solaris doesn't support direct io.
696 On Windows the synchronous ioengines don't support direct io.
698 buffered=bool If value is true, use buffered io. This is the opposite
699 of the 'direct' option. Defaults to true.
701 offset=int Start io at the given offset in the file. The data before
702 the given offset will not be touched. This effectively
703 caps the file size at real_size - offset.
705 offset_increment=int If this is provided, then the real offset becomes
706 the offset + offset_increment * thread_number, where the
707 thread number is a counter that starts at 0 and is incremented
708 for each job. This option is useful if there are several jobs
709 which are intended to operate on a file in parallel in disjoint
710 segments, with even spacing between the starting points.
712 fsync=int If writing to a file, issue a sync of the dirty data
713 for every number of blocks given. For example, if you give
714 32 as a parameter, fio will sync the file for every 32
715 writes issued. If fio is using non-buffered io, we may
716 not sync the file. The exception is the sg io engine, which
717 synchronizes the disk cache anyway.
719 fdatasync=int Like fsync= but uses fdatasync() to only sync data and not
721 In FreeBSD and Windows there is no fdatasync(), this falls back to
724 sync_file_range=str:val Use sync_file_range() for every 'val' number of
725 write operations. Fio will track range of writes that
726 have happened since the last sync_file_range() call. 'str'
727 can currently be one or more of:
729 wait_before SYNC_FILE_RANGE_WAIT_BEFORE
730 write SYNC_FILE_RANGE_WRITE
731 wait_after SYNC_FILE_RANGE_WAIT_AFTER
733 So if you do sync_file_range=wait_before,write:8, fio would
734 use SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE for
735 every 8 writes. Also see the sync_file_range(2) man page.
736 This option is Linux specific.
738 overwrite=bool If true, writes to a file will always overwrite existing
739 data. If the file doesn't already exist, it will be
740 created before the write phase begins. If the file exists
741 and is large enough for the specified write phase, nothing
744 end_fsync=bool If true, fsync file contents when a write stage has completed.
746 fsync_on_close=bool If true, fio will fsync() a dirty file on close.
747 This differs from end_fsync in that it will happen on every
748 file close, not just at the end of the job.
750 rwmixread=int How large a percentage of the mix should be reads.
752 rwmixwrite=int How large a percentage of the mix should be writes. If both
753 rwmixread and rwmixwrite is given and the values do not add
754 up to 100%, the latter of the two will be used to override
755 the first. This may interfere with a given rate setting,
756 if fio is asked to limit reads or writes to a certain rate.
757 If that is the case, then the distribution may be skewed.
759 random_distribution=str:float By default, fio will use a completely uniform
760 random distribution when asked to perform random IO. Sometimes
761 it is useful to skew the distribution in specific ways,
762 ensuring that some parts of the data is more hot than others.
763 fio includes the following distribution models:
765 random Uniform random distribution
766 zipf Zipf distribution
767 pareto Pareto distribution
769 When using a zipf or pareto distribution, an input value
770 is also needed to define the access pattern. For zipf, this
771 is the zipf theta. For pareto, it's the pareto power. Fio
772 includes a test program, genzipf, that can be used visualize
773 what the given input values will yield in terms of hit rates.
774 If you wanted to use zipf with a theta of 1.2, you would use
775 random_distribution=zipf:1.2 as the option. If a non-uniform
776 model is used, fio will disable use of the random map.
778 percentage_random=int For a random workload, set how big a percentage should
779 be random. This defaults to 100%, in which case the workload
780 is fully random. It can be set from anywhere from 0 to 100.
781 Setting it to 0 would make the workload fully sequential. Any
782 setting in between will result in a random mix of sequential
783 and random IO, at the given percentages. It is possible to
784 set different values for reads, writes, and trim. To do so,
785 simply use a comma separated list. See blocksize.
787 norandommap Normally fio will cover every block of the file when doing
788 random IO. If this option is given, fio will just get a
789 new random offset without looking at past io history. This
790 means that some blocks may not be read or written, and that
791 some blocks may be read/written more than once. This option
792 is mutually exclusive with verify= if and only if multiple
793 blocksizes (via bsrange=) are used, since fio only tracks
794 complete rewrites of blocks.
796 softrandommap=bool See norandommap. If fio runs with the random block map
797 enabled and it fails to allocate the map, if this option is
798 set it will continue without a random block map. As coverage
799 will not be as complete as with random maps, this option is
802 random_generator=str Fio supports the following engines for generating
803 IO offsets for random IO:
805 tausworthe Strong 2^88 cycle random number generator
806 lfsr Linear feedback shift register generator
808 Tausworthe is a strong random number generator, but it
809 requires tracking on the side if we want to ensure that
810 blocks are only read or written once. LFSR guarantees
811 that we never generate the same offset twice, and it's
812 also less computationally expensive. It's not a true
813 random generator, however, though for IO purposes it's
814 typically good enough. LFSR only works with single
815 block sizes, not with workloads that use multiple block
816 sizes. If used with such a workload, fio may read or write
817 some blocks multiple times.
819 nice=int Run the job with the given nice value. See man nice(2).
821 prio=int Set the io priority value of this job. Linux limits us to
822 a positive value between 0 and 7, with 0 being the highest.
825 prioclass=int Set the io priority class. See man ionice(1).
827 thinktime=int Stall the job x microseconds after an io has completed before
828 issuing the next. May be used to simulate processing being
829 done by an application. See thinktime_blocks and
833 Only valid if thinktime is set - pretend to spend CPU time
834 doing something with the data received, before falling back
835 to sleeping for the rest of the period specified by
839 Only valid if thinktime is set - control how many blocks
840 to issue, before waiting 'thinktime' usecs. If not set,
841 defaults to 1 which will make fio wait 'thinktime' usecs
842 after every block. This effectively makes any queue depth
843 setting redundant, since no more than 1 IO will be queued
844 before we have to complete it and do our thinktime. In
845 other words, this setting effectively caps the queue depth
846 if the latter is larger.
848 rate=int Cap the bandwidth used by this job. The number is in bytes/sec,
849 the normal suffix rules apply. You can use rate=500k to limit
850 reads and writes to 500k each, or you can specify read and
851 writes separately. Using rate=1m,500k would limit reads to
852 1MB/sec and writes to 500KB/sec. Capping only reads or
853 writes can be done with rate=,500k or rate=500k,. The former
854 will only limit writes (to 500KB/sec), the latter will only
857 ratemin=int Tell fio to do whatever it can to maintain at least this
858 bandwidth. Failing to meet this requirement, will cause
859 the job to exit. The same format as rate is used for
860 read vs write separation.
862 rate_iops=int Cap the bandwidth to this number of IOPS. Basically the same
863 as rate, just specified independently of bandwidth. If the
864 job is given a block size range instead of a fixed value,
865 the smallest block size is used as the metric. The same format
866 as rate is used for read vs write seperation.
868 rate_iops_min=int If fio doesn't meet this rate of IO, it will cause
869 the job to exit. The same format as rate is used for read vs
872 max_latency=int If set, fio will exit the job if it exceeds this maximum
873 latency. It will exit with an ETIME error.
875 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
878 cpumask=int Set the CPU affinity of this job. The parameter given is a
879 bitmask of allowed CPU's the job may run on. So if you want
880 the allowed CPUs to be 1 and 5, you would pass the decimal
881 value of (1 << 1 | 1 << 5), or 34. See man
882 sched_setaffinity(2). This may not work on all supported
883 operating systems or kernel versions. This option doesn't
884 work well for a higher CPU count than what you can store in
885 an integer mask, so it can only control cpus 1-32. For
886 boxes with larger CPU counts, use cpus_allowed.
888 cpus_allowed=str Controls the same options as cpumask, but it allows a text
889 setting of the permitted CPUs instead. So to use CPUs 1 and
890 5, you would specify cpus_allowed=1,5. This options also
891 allows a range of CPUs. Say you wanted a binding to CPUs
892 1, 5, and 8-15, you would set cpus_allowed=1,5,8-15.
894 numa_cpu_nodes=str Set this job running on spcified NUMA nodes' CPUs. The
895 arguments allow comma delimited list of cpu numbers,
896 A-B ranges, or 'all'. Note, to enable numa options support,
897 fio must be built on a system with libnuma-dev(el) installed.
899 numa_mem_policy=str Set this job's memory policy and corresponding NUMA
900 nodes. Format of the argements:
902 `mode' is one of the following memory policy:
903 default, prefer, bind, interleave, local
904 For `default' and `local' memory policy, no node is
905 needed to be specified.
906 For `prefer', only one node is allowed.
907 For `bind' and `interleave', it allow comma delimited
908 list of numbers, A-B ranges, or 'all'.
910 startdelay=time Start this job the specified number of seconds after fio
911 has started. Only useful if the job file contains several
912 jobs, and you want to delay starting some jobs to a certain
915 runtime=time Tell fio to terminate processing after the specified number
916 of seconds. It can be quite hard to determine for how long
917 a specified job will run, so this parameter is handy to
918 cap the total runtime to a given time.
920 time_based If set, fio will run for the duration of the runtime
921 specified even if the file(s) are completely read or
922 written. It will simply loop over the same workload
923 as many times as the runtime allows.
925 ramp_time=time If set, fio will run the specified workload for this amount
926 of time before logging any performance numbers. Useful for
927 letting performance settle before logging results, thus
928 minimizing the runtime required for stable results. Note
929 that the ramp_time is considered lead in time for a job,
930 thus it will increase the total runtime if a special timeout
931 or runtime is specified.
933 invalidate=bool Invalidate the buffer/page cache parts for this file prior
934 to starting io. Defaults to true.
936 sync=bool Use sync io for buffered writes. For the majority of the
937 io engines, this means using O_SYNC.
940 mem=str Fio can use various types of memory as the io unit buffer.
941 The allowed values are:
943 malloc Use memory from malloc(3) as the buffers.
945 shm Use shared memory as the buffers. Allocated
948 shmhuge Same as shm, but use huge pages as backing.
950 mmap Use mmap to allocate buffers. May either be
951 anonymous memory, or can be file backed if
952 a filename is given after the option. The
953 format is mem=mmap:/path/to/file.
955 mmaphuge Use a memory mapped huge file as the buffer
956 backing. Append filename after mmaphuge, ala
957 mem=mmaphuge:/hugetlbfs/file
959 The area allocated is a function of the maximum allowed
960 bs size for the job, multiplied by the io depth given. Note
961 that for shmhuge and mmaphuge to work, the system must have
962 free huge pages allocated. This can normally be checked
963 and set by reading/writing /proc/sys/vm/nr_hugepages on a
964 Linux system. Fio assumes a huge page is 4MB in size. So
965 to calculate the number of huge pages you need for a given
966 job file, add up the io depth of all jobs (normally one unless
967 iodepth= is used) and multiply by the maximum bs set. Then
968 divide that number by the huge page size. You can see the
969 size of the huge pages in /proc/meminfo. If no huge pages
970 are allocated by having a non-zero number in nr_hugepages,
971 using mmaphuge or shmhuge will fail. Also see hugepage-size.
973 mmaphuge also needs to have hugetlbfs mounted and the file
974 location should point there. So if it's mounted in /huge,
975 you would use mem=mmaphuge:/huge/somefile.
977 iomem_align=int This indiciates the memory alignment of the IO memory buffers.
978 Note that the given alignment is applied to the first IO unit
979 buffer, if using iodepth the alignment of the following buffers
980 are given by the bs used. In other words, if using a bs that is
981 a multiple of the page sized in the system, all buffers will
982 be aligned to this value. If using a bs that is not page
983 aligned, the alignment of subsequent IO memory buffers is the
984 sum of the iomem_align and bs used.
987 Defines the size of a huge page. Must at least be equal
988 to the system setting, see /proc/meminfo. Defaults to 4MB.
989 Should probably always be a multiple of megabytes, so using
990 hugepage-size=Xm is the preferred way to set this to avoid
991 setting a non-pow-2 bad value.
993 exitall When one job finishes, terminate the rest. The default is
994 to wait for each job to finish, sometimes that is not the
997 bwavgtime=int Average the calculated bandwidth over the given time. Value
998 is specified in milliseconds.
1000 iopsavgtime=int Average the calculated IOPS over the given time. Value
1001 is specified in milliseconds.
1003 create_serialize=bool If true, serialize the file creating for the jobs.
1004 This may be handy to avoid interleaving of data
1005 files, which may greatly depend on the filesystem
1006 used and even the number of processors in the system.
1008 create_fsync=bool fsync the data file after creation. This is the
1011 create_on_open=bool Don't pre-setup the files for IO, just create open()
1012 when it's time to do IO to that file.
1014 create_only=bool If true, fio will only run the setup phase of the job.
1015 If files need to be laid out or updated on disk, only
1016 that will be done. The actual job contents are not
1019 pre_read=bool If this is given, files will be pre-read into memory before
1020 starting the given IO operation. This will also clear
1021 the 'invalidate' flag, since it is pointless to pre-read
1022 and then drop the cache. This will only work for IO engines
1023 that are seekable, since they allow you to read the same data
1024 multiple times. Thus it will not work on eg network or splice
1027 unlink=bool Unlink the job files when done. Not the default, as repeated
1028 runs of that job would then waste time recreating the file
1029 set again and again.
1031 loops=int Run the specified number of iterations of this job. Used
1032 to repeat the same workload a given number of times. Defaults
1035 do_verify=bool Run the verify phase after a write phase. Only makes sense if
1036 verify is set. Defaults to 1.
1038 verify=str If writing to a file, fio can verify the file contents
1039 after each iteration of the job. The allowed values are:
1041 md5 Use an md5 sum of the data area and store
1042 it in the header of each block.
1044 crc64 Use an experimental crc64 sum of the data
1045 area and store it in the header of each
1048 crc32c Use a crc32c sum of the data area and store
1049 it in the header of each block.
1051 crc32c-intel Use hardware assisted crc32c calcuation
1052 provided on SSE4.2 enabled processors. Falls
1053 back to regular software crc32c, if not
1054 supported by the system.
1056 crc32 Use a crc32 sum of the data area and store
1057 it in the header of each block.
1059 crc16 Use a crc16 sum of the data area and store
1060 it in the header of each block.
1062 crc7 Use a crc7 sum of the data area and store
1063 it in the header of each block.
1065 sha512 Use sha512 as the checksum function.
1067 sha256 Use sha256 as the checksum function.
1069 sha1 Use optimized sha1 as the checksum function.
1071 meta Write extra information about each io
1072 (timestamp, block number etc.). The block
1073 number is verified. See also verify_pattern.
1075 null Only pretend to verify. Useful for testing
1076 internals with ioengine=null, not for much
1079 This option can be used for repeated burn-in tests of a
1080 system to make sure that the written data is also
1081 correctly read back. If the data direction given is
1082 a read or random read, fio will assume that it should
1083 verify a previously written file. If the data direction
1084 includes any form of write, the verify will be of the
1087 verifysort=bool If set, fio will sort written verify blocks when it deems
1088 it faster to read them back in a sorted manner. This is
1089 often the case when overwriting an existing file, since
1090 the blocks are already laid out in the file system. You
1091 can ignore this option unless doing huge amounts of really
1092 fast IO where the red-black tree sorting CPU time becomes
1095 verify_offset=int Swap the verification header with data somewhere else
1096 in the block before writing. Its swapped back before
1099 verify_interval=int Write the verification header at a finer granularity
1100 than the blocksize. It will be written for chunks the
1101 size of header_interval. blocksize should divide this
1104 verify_pattern=str If set, fio will fill the io buffers with this
1105 pattern. Fio defaults to filling with totally random
1106 bytes, but sometimes it's interesting to fill with a known
1107 pattern for io verification purposes. Depending on the
1108 width of the pattern, fio will fill 1/2/3/4 bytes of the
1109 buffer at the time(it can be either a decimal or a hex number).
1110 The verify_pattern if larger than a 32-bit quantity has to
1111 be a hex number that starts with either "0x" or "0X". Use
1114 verify_fatal=bool Normally fio will keep checking the entire contents
1115 before quitting on a block verification failure. If this
1116 option is set, fio will exit the job on the first observed
1119 verify_dump=bool If set, dump the contents of both the original data
1120 block and the data block we read off disk to files. This
1121 allows later analysis to inspect just what kind of data
1122 corruption occurred. Off by default.
1124 verify_async=int Fio will normally verify IO inline from the submitting
1125 thread. This option takes an integer describing how many
1126 async offload threads to create for IO verification instead,
1127 causing fio to offload the duty of verifying IO contents
1128 to one or more separate threads. If using this offload
1129 option, even sync IO engines can benefit from using an
1130 iodepth setting higher than 1, as it allows them to have
1131 IO in flight while verifies are running.
1133 verify_async_cpus=str Tell fio to set the given CPU affinity on the
1134 async IO verification threads. See cpus_allowed for the
1137 verify_backlog=int Fio will normally verify the written contents of a
1138 job that utilizes verify once that job has completed. In
1139 other words, everything is written then everything is read
1140 back and verified. You may want to verify continually
1141 instead for a variety of reasons. Fio stores the meta data
1142 associated with an IO block in memory, so for large
1143 verify workloads, quite a bit of memory would be used up
1144 holding this meta data. If this option is enabled, fio
1145 will write only N blocks before verifying these blocks.
1147 will verify the previously written blocks before continuing
1150 verify_backlog_batch=int Control how many blocks fio will verify
1151 if verify_backlog is set. If not set, will default to
1152 the value of verify_backlog (meaning the entire queue
1153 is read back and verified). If verify_backlog_batch is
1154 less than verify_backlog then not all blocks will be verified,
1155 if verify_backlog_batch is larger than verify_backlog, some
1156 blocks will be verified more than once.
1159 wait_for_previous Wait for preceeding jobs in the job file to exit, before
1160 starting this one. Can be used to insert serialization
1161 points in the job file. A stone wall also implies starting
1162 a new reporting group.
1164 new_group Start a new reporting group. See: group_reporting.
1166 numjobs=int Create the specified number of clones of this job. May be
1167 used to setup a larger number of threads/processes doing
1168 the same thing. Each thread is reported separately; to see
1169 statistics for all clones as a whole, use group_reporting in
1170 conjunction with new_group.
1172 group_reporting It may sometimes be interesting to display statistics for
1173 groups of jobs as a whole instead of for each individual job.
1174 This is especially true if 'numjobs' is used; looking at
1175 individual thread/process output quickly becomes unwieldy.
1176 To see the final report per-group instead of per-job, use
1177 'group_reporting'. Jobs in a file will be part of the same
1178 reporting group, unless if separated by a stonewall, or by
1181 thread fio defaults to forking jobs, however if this option is
1182 given, fio will use pthread_create(3) to create threads
1185 zonesize=int Divide a file into zones of the specified size. See zoneskip.
1187 zoneskip=int Skip the specified number of bytes when zonesize data has
1188 been read. The two zone options can be used to only do
1189 io on zones of a file.
1191 write_iolog=str Write the issued io patterns to the specified file. See
1192 read_iolog. Specify a separate file for each job, otherwise
1193 the iologs will be interspersed and the file may be corrupt.
1195 read_iolog=str Open an iolog with the specified file name and replay the
1196 io patterns it contains. This can be used to store a
1197 workload and replay it sometime later. The iolog given
1198 may also be a blktrace binary file, which allows fio
1199 to replay a workload captured by blktrace. See blktrace
1200 for how to capture such logging data. For blktrace replay,
1201 the file needs to be turned into a blkparse binary data
1202 file first (blkparse <device> -o /dev/null -d file_for_fio.bin).
1204 replay_no_stall=int When replaying I/O with read_iolog the default behavior
1205 is to attempt to respect the time stamps within the log and
1206 replay them with the appropriate delay between IOPS. By
1207 setting this variable fio will not respect the timestamps and
1208 attempt to replay them as fast as possible while still
1209 respecting ordering. The result is the same I/O pattern to a
1210 given device, but different timings.
1212 replay_redirect=str While replaying I/O patterns using read_iolog the
1213 default behavior is to replay the IOPS onto the major/minor
1214 device that each IOP was recorded from. This is sometimes
1215 undesireable because on a different machine those major/minor
1216 numbers can map to a different device. Changing hardware on
1217 the same system can also result in a different major/minor
1218 mapping. Replay_redirect causes all IOPS to be replayed onto
1219 the single specified device regardless of the device it was
1220 recorded from. i.e. replay_redirect=/dev/sdc would cause all
1221 IO in the blktrace to be replayed onto /dev/sdc. This means
1222 multiple devices will be replayed onto a single, if the trace
1223 contains multiple devices. If you want multiple devices to be
1224 replayed concurrently to multiple redirected devices you must
1225 blkparse your trace into separate traces and replay them with
1226 independent fio invocations. Unfortuantely this also breaks
1227 the strict time ordering between multiple device accesses.
1229 write_bw_log=str If given, write a bandwidth log of the jobs in this job
1230 file. Can be used to store data of the bandwidth of the
1231 jobs in their lifetime. The included fio_generate_plots
1232 script uses gnuplot to turn these text files into nice
1233 graphs. See write_lat_log for behaviour of given
1234 filename. For this option, the suffix is _bw.log.
1236 write_lat_log=str Same as write_bw_log, except that this option stores io
1237 submission, completion, and total latencies instead. If no
1238 filename is given with this option, the default filename of
1239 "jobname_type.log" is used. Even if the filename is given,
1240 fio will still append the type of log. So if one specifies
1244 The actual log names will be foo_slat.log, foo_clat.log,
1245 and foo_lat.log. This helps fio_generate_plot fine the logs
1248 write_bw_log=str If given, write an IOPS log of the jobs in this job
1249 file. See write_bw_log.
1251 write_iops_log=str Same as write_bw_log, but writes IOPS. If no filename is
1252 given with this option, the default filename of
1253 "jobname_type.log" is used. Even if the filename is given,
1254 fio will still append the type of log.
1256 log_avg_msec=int By default, fio will log an entry in the iops, latency,
1257 or bw log for every IO that completes. When writing to the
1258 disk log, that can quickly grow to a very large size. Setting
1259 this option makes fio average the each log entry over the
1260 specified period of time, reducing the resolution of the log.
1263 lockmem=int Pin down the specified amount of memory with mlock(2). Can
1264 potentially be used instead of removing memory or booting
1265 with less memory to simulate a smaller amount of memory.
1266 The amount specified is per worker.
1268 exec_prerun=str Before running this job, issue the command specified
1269 through system(3). Output is redirected in a file called
1272 exec_postrun=str After the job completes, issue the command specified
1273 though system(3). Output is redirected in a file called
1274 jobname.postrun.txt.
1276 ioscheduler=str Attempt to switch the device hosting the file to the specified
1277 io scheduler before running.
1279 disk_util=bool Generate disk utilization statistics, if the platform
1280 supports it. Defaults to on.
1282 disable_lat=bool Disable measurements of total latency numbers. Useful
1283 only for cutting back the number of calls to gettimeofday,
1284 as that does impact performance at really high IOPS rates.
1285 Note that to really get rid of a large amount of these
1286 calls, this option must be used with disable_slat and
1289 disable_clat=bool Disable measurements of completion latency numbers. See
1292 disable_slat=bool Disable measurements of submission latency numbers. See
1295 disable_bw=bool Disable measurements of throughput/bandwidth numbers. See
1298 clat_percentiles=bool Enable the reporting of percentiles of
1299 completion latencies.
1301 percentile_list=float_list Overwrite the default list of percentiles
1302 for completion latencies. Each number is a floating
1303 number in the range (0,100], and the maximum length of
1304 the list is 20. Use ':' to separate the numbers, and
1305 list the numbers in ascending order. For example,
1306 --percentile_list=99.5:99.9 will cause fio to report
1307 the values of completion latency below which 99.5% and
1308 99.9% of the observed latencies fell, respectively.
1310 clocksource=str Use the given clocksource as the base of timing. The
1311 supported options are:
1313 gettimeofday gettimeofday(2)
1315 clock_gettime clock_gettime(2)
1317 cpu Internal CPU clock source
1319 cpu is the preferred clocksource if it is reliable, as it
1320 is very fast (and fio is heavy on time calls). Fio will
1321 automatically use this clocksource if it's supported and
1322 considered reliable on the system it is running on, unless
1323 another clocksource is specifically set. For x86/x86-64 CPUs,
1324 this means supporting TSC Invariant.
1326 gtod_reduce=bool Enable all of the gettimeofday() reducing options
1327 (disable_clat, disable_slat, disable_bw) plus reduce
1328 precision of the timeout somewhat to really shrink
1329 the gettimeofday() call count. With this option enabled,
1330 we only do about 0.4% of the gtod() calls we would have
1331 done if all time keeping was enabled.
1333 gtod_cpu=int Sometimes it's cheaper to dedicate a single thread of
1334 execution to just getting the current time. Fio (and
1335 databases, for instance) are very intensive on gettimeofday()
1336 calls. With this option, you can set one CPU aside for
1337 doing nothing but logging current time to a shared memory
1338 location. Then the other threads/processes that run IO
1339 workloads need only copy that segment, instead of entering
1340 the kernel with a gettimeofday() call. The CPU set aside
1341 for doing these time calls will be excluded from other
1342 uses. Fio will manually clear it from the CPU mask of other
1345 continue_on_error=str Normally fio will exit the job on the first observed
1346 failure. If this option is set, fio will continue the job when
1347 there is a 'non-fatal error' (EIO or EILSEQ) until the runtime
1348 is exceeded or the I/O size specified is completed. If this
1349 option is used, there are two more stats that are appended,
1350 the total error count and the first error. The error field
1351 given in the stats is the first error that was hit during the
1354 The allowed values are:
1356 none Exit on any IO or verify errors.
1358 read Continue on read errors, exit on all others.
1360 write Continue on write errors, exit on all others.
1362 io Continue on any IO error, exit on all others.
1364 verify Continue on verify errors, exit on all others.
1366 all Continue on all errors.
1368 0 Backward-compatible alias for 'none'.
1370 1 Backward-compatible alias for 'all'.
1372 ignore_error=str Sometimes you want to ignore some errors during test
1373 in that case you can specify error list for each error type.
1374 ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST
1375 errors for given error type is separated with ':'. Error
1376 may be symbol ('ENOSPC', 'ENOMEM') or integer.
1378 ignore_error=EAGAIN,ENOSPC:122
1379 This option will ignore EAGAIN from READ, and ENOSPC and
1380 122(EDQUOT) from WRITE.
1382 error_dump=bool If set dump every error even if it is non fatal, true
1383 by default. If disabled only fatal error will be dumped
1385 cgroup=str Add job to this control group. If it doesn't exist, it will
1386 be created. The system must have a mounted cgroup blkio
1387 mount point for this to work. If your system doesn't have it
1388 mounted, you can do so with:
1390 # mount -t cgroup -o blkio none /cgroup
1392 cgroup_weight=int Set the weight of the cgroup to this value. See
1393 the documentation that comes with the kernel, allowed values
1394 are in the range of 100..1000.
1396 cgroup_nodelete=bool Normally fio will delete the cgroups it has created after
1397 the job completion. To override this behavior and to leave
1398 cgroups around after the job completion, set cgroup_nodelete=1.
1399 This can be useful if one wants to inspect various cgroup
1400 files after job completion. Default: false
1402 uid=int Instead of running as the invoking user, set the user ID to
1403 this value before the thread/process does any work.
1405 gid=int Set group ID, see uid.
1407 flow_id=int The ID of the flow. If not specified, it defaults to being a
1408 global flow. See flow.
1410 flow=int Weight in token-based flow control. If this value is used, then
1411 there is a 'flow counter' which is used to regulate the
1412 proportion of activity between two or more jobs. fio attempts
1413 to keep this flow counter near zero. The 'flow' parameter
1414 stands for how much should be added or subtracted to the flow
1415 counter on each iteration of the main I/O loop. That is, if
1416 one job has flow=8 and another job has flow=-1, then there
1417 will be a roughly 1:8 ratio in how much one runs vs the other.
1419 flow_watermark=int The maximum value that the absolute value of the flow
1420 counter is allowed to reach before the job must wait for a
1421 lower value of the counter.
1423 flow_sleep=int The period of time, in microseconds, to wait after the flow
1424 watermark has been exceeded before retrying operations
1426 In addition, there are some parameters which are only valid when a specific
1427 ioengine is in use. These are used identically to normal parameters, with the
1428 caveat that when used on the command line, they must come after the ioengine
1429 that defines them is selected.
1431 [libaio] userspace_reap Normally, with the libaio engine in use, fio will use
1432 the io_getevents system call to reap newly returned events.
1433 With this flag turned on, the AIO ring will be read directly
1434 from user-space to reap events. The reaping mode is only
1435 enabled when polling for a minimum of 0 events (eg when
1436 iodepth_batch_complete=0).
1438 [cpu] cpuload=int Attempt to use the specified percentage of CPU cycles.
1440 [cpu] cpuchunks=int Split the load into cycles of the given time. In
1443 [netsplice] hostname=str
1444 [net] hostname=str The host name or IP address to use for TCP or UDP based IO.
1445 If the job is a TCP listener or UDP reader, the hostname is not
1446 used and must be omitted unless it is a valid UDP multicast
1449 [netsplice] port=int
1450 [net] port=int The TCP or UDP port to bind to or connect to.
1452 [netsplice] interface=str
1453 [net] interface=str The IP address of the network interface used to send or
1454 receive UDP multicast
1457 [net] ttl=int Time-to-live value for outgoing UDP multicast packets.
1460 [netsplice] nodelay=bool
1461 [net] nodelay=bool Set TCP_NODELAY on TCP connections.
1463 [netsplice] protocol=str
1464 [netsplice] proto=str
1466 [net] proto=str The network protocol to use. Accepted values are:
1468 tcp Transmission control protocol
1469 udp User datagram protocol
1470 unix UNIX domain socket
1472 When the protocol is TCP or UDP, the port must also be given,
1473 as well as the hostname if the job is a TCP listener or UDP
1474 reader. For unix sockets, the normal filename option should be
1475 used and the port is invalid.
1477 [net] listen For TCP network connections, tell fio to listen for incoming
1478 connections rather than initiating an outgoing connection. The
1479 hostname must be omitted if this option is used.
1480 [net] pingpong Normaly a network writer will just continue writing data, and
1481 a network reader will just consume packages. If pingpong=1
1482 is set, a writer will send its normal payload to the reader,
1483 then wait for the reader to send the same payload back. This
1484 allows fio to measure network latencies. The submission
1485 and completion latencies then measure local time spent
1486 sending or receiving, and the completion latency measures
1487 how long it took for the other end to receive and send back.
1488 For UDP multicast traffic pingpong=1 should only be set for a
1489 single reader when multiple readers are listening to the same
1492 [e4defrag] donorname=str
1493 File will be used as a block donor(swap extents between files)
1494 [e4defrag] inplace=int
1495 Configure donor file blocks allocation strategy
1496 0(default): Preallocate donor's file on init
1497 1 : allocate space immidietly inside defragment event,
1498 and free right after event
1502 6.0 Interpreting the output
1503 ---------------------------
1505 fio spits out a lot of output. While running, fio will display the
1506 status of the jobs created. An example of that would be:
1508 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1510 The characters inside the square brackets denote the current status of
1511 each thread. The possible values (in typical life cycle order) are:
1515 P Thread setup, but not started.
1517 I Thread initialized, waiting or generating necessary data.
1518 p Thread running pre-reading file(s).
1519 R Running, doing sequential reads.
1520 r Running, doing random reads.
1521 W Running, doing sequential writes.
1522 w Running, doing random writes.
1523 M Running, doing mixed sequential reads/writes.
1524 m Running, doing mixed random reads/writes.
1525 F Running, currently waiting for fsync()
1526 V Running, doing verification of written data.
1527 E Thread exited, not reaped by main thread yet.
1529 X Thread reaped, exited with an error.
1530 K Thread reaped, exited due to signal.
1532 The other values are fairly self explanatory - number of threads
1533 currently running and doing io, rate of io since last check (read speed
1534 listed first, then write speed), and the estimated completion percentage
1535 and time for the running group. It's impossible to estimate runtime of
1536 the following groups (if any). Note that the string is displayed in order,
1537 so it's possible to tell which of the jobs are currently doing what. The
1538 first character is the first job defined in the job file, and so forth.
1540 When fio is done (or interrupted by ctrl-c), it will show the data for
1541 each thread, group of threads, and disks in that order. For each data
1542 direction, the output looks like:
1544 Client1 (g=0): err= 0:
1545 write: io= 32MB, bw= 666KB/s, iops=89 , runt= 50320msec
1546 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
1547 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
1548 bw (KB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
1549 cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17
1550 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
1551 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
1552 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
1553 issued r/w: total=0/32768, short=0/0
1554 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
1555 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
1557 The client number is printed, along with the group id and error of that
1558 thread. Below is the io statistics, here for writes. In the order listed,
1561 io= Number of megabytes io performed
1562 bw= Average bandwidth rate
1563 iops= Average IOs performed per second
1564 runt= The runtime of that thread
1565 slat= Submission latency (avg being the average, stdev being the
1566 standard deviation). This is the time it took to submit
1567 the io. For sync io, the slat is really the completion
1568 latency, since queue/complete is one operation there. This
1569 value can be in milliseconds or microseconds, fio will choose
1570 the most appropriate base and print that. In the example
1571 above, milliseconds is the best scale. Note: in --minimal mode
1572 latencies are always expressed in microseconds.
1573 clat= Completion latency. Same names as slat, this denotes the
1574 time from submission to completion of the io pieces. For
1575 sync io, clat will usually be equal (or very close) to 0,
1576 as the time from submit to complete is basically just
1577 CPU time (io has already been done, see slat explanation).
1578 bw= Bandwidth. Same names as the xlat stats, but also includes
1579 an approximate percentage of total aggregate bandwidth
1580 this thread received in this group. This last value is
1581 only really useful if the threads in this group are on the
1582 same disk, since they are then competing for disk access.
1583 cpu= CPU usage. User and system time, along with the number
1584 of context switches this thread went through, usage of
1585 system and user time, and finally the number of major
1586 and minor page faults.
1587 IO depths= The distribution of io depths over the job life time. The
1588 numbers are divided into powers of 2, so for example the
1589 16= entries includes depths up to that value but higher
1590 than the previous entry. In other words, it covers the
1591 range from 16 to 31.
1592 IO submit= How many pieces of IO were submitting in a single submit
1593 call. Each entry denotes that amount and below, until
1594 the previous entry - eg, 8=100% mean that we submitted
1595 anywhere in between 5-8 ios per submit call.
1596 IO complete= Like the above submit number, but for completions instead.
1597 IO issued= The number of read/write requests issued, and how many
1599 IO latencies= The distribution of IO completion latencies. This is the
1600 time from when IO leaves fio and when it gets completed.
1601 The numbers follow the same pattern as the IO depths,
1602 meaning that 2=1.6% means that 1.6% of the IO completed
1603 within 2 msecs, 20=12.8% means that 12.8% of the IO
1604 took more than 10 msecs, but less than (or equal to) 20 msecs.
1606 After each client has been listed, the group statistics are printed. They
1607 will look like this:
1609 Run status group 0 (all jobs):
1610 READ: io=64MB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
1611 WRITE: io=64MB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
1613 For each data direction, it prints:
1615 io= Number of megabytes io performed.
1616 aggrb= Aggregate bandwidth of threads in this group.
1617 minb= The minimum average bandwidth a thread saw.
1618 maxb= The maximum average bandwidth a thread saw.
1619 mint= The smallest runtime of the threads in that group.
1620 maxt= The longest runtime of the threads in that group.
1622 And finally, the disk statistics are printed. They will look like this:
1624 Disk stats (read/write):
1625 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
1627 Each value is printed for both reads and writes, with reads first. The
1630 ios= Number of ios performed by all groups.
1631 merge= Number of merges io the io scheduler.
1632 ticks= Number of ticks we kept the disk busy.
1633 io_queue= Total time spent in the disk queue.
1634 util= The disk utilization. A value of 100% means we kept the disk
1635 busy constantly, 50% would be a disk idling half of the time.
1637 It is also possible to get fio to dump the current output while it is
1638 running, without terminating the job. To do that, send fio the USR1 signal.
1639 You can also get regularly timed dumps by using the --status-interval
1640 parameter, or by creating a file in /tmp named fio-dump-status. If fio
1641 sees this file, it will unlink it and dump the current output status.
1647 For scripted usage where you typically want to generate tables or graphs
1648 of the results, fio can output the results in a semicolon separated format.
1649 The format is one long line of values, such as:
1651 2;card0;0;0;7139336;121836;60004;1;10109;27.932460;116.933948;220;126861;3495.446807;1085.368601;226;126864;3523.635629;1089.012448;24063;99944;50.275485%;59818.274627;5540.657370;7155060;122104;60004;1;8338;29.086342;117.839068;388;128077;5032.488518;1234.785715;391;128085;5061.839412;1236.909129;23436;100928;50.287926%;59964.832030;5644.844189;14.595833%;19.394167%;123706;0;7313;0.1%;0.1%;0.1%;0.1%;0.1%;0.1%;100.0%;0.00%;0.00%;0.00%;0.00%;0.00%;0.00%;0.01%;0.02%;0.05%;0.16%;6.04%;40.40%;52.68%;0.64%;0.01%;0.00%;0.01%;0.00%;0.00%;0.00%;0.00%;0.00%
1652 A description of this job goes here.
1654 The job description (if provided) follows on a second line.
1656 To enable terse output, use the --minimal command line option. The first
1657 value is the version of the terse output format. If the output has to
1658 be changed for some reason, this number will be incremented by 1 to
1659 signify that change.
1661 Split up, the format is as follows:
1663 terse version, fio version, jobname, groupid, error
1665 Total IO (KB), bandwidth (KB/sec), IOPS, runtime (msec)
1666 Submission latency: min, max, mean, deviation (usec)
1667 Completion latency: min, max, mean, deviation (usec)
1668 Completion latency percentiles: 20 fields (see below)
1669 Total latency: min, max, mean, deviation (usec)
1670 Bw (KB/s): min, max, aggregate percentage of total, mean, deviation
1672 Total IO (KB), bandwidth (KB/sec), IOPS, runtime (msec)
1673 Submission latency: min, max, mean, deviation (usec)
1674 Completion latency: min, max, mean, deviation (usec)
1675 Completion latency percentiles: 20 fields (see below)
1676 Total latency: min, max, mean, deviation (usec)
1677 Bw (KB/s): min, max, aggregate percentage of total, mean, deviation
1678 CPU usage: user, system, context switches, major faults, minor faults
1679 IO depths: <=1, 2, 4, 8, 16, 32, >=64
1680 IO latencies microseconds: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1681 IO latencies milliseconds: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1682 Disk utilization: Disk name, Read ios, write ios,
1683 Read merges, write merges,
1684 Read ticks, write ticks,
1685 Time spent in queue, disk utilization percentage
1686 Additional Info (dependant on continue_on_error, default off): total # errors, first error code
1688 Additional Info (dependant on description being set): Text description
1690 Completion latency percentiles can be a grouping of up to 20 sets, so
1691 for the terse output fio writes all of them. Each field will look like this:
1695 which is the Xth percentile, and the usec latency associated with it.
1697 For disk utilization, all disks used by fio are shown. So for each disk
1698 there will be a disk utilization section.
1701 8.0 Trace file format
1702 ---------------------
1703 There are two trace file format that you can encounter. The older (v1) format
1704 is unsupported since version 1.20-rc3 (March 2008). It will still be described
1705 below in case that you get an old trace and want to understand it.
1707 In any case the trace is a simple text file with a single action per line.
1710 8.1 Trace file format v1
1711 ------------------------
1712 Each line represents a single io action in the following format:
1716 where rw=0/1 for read/write, and the offset and length entries being in bytes.
1718 This format is not supported in Fio versions => 1.20-rc3.
1721 8.2 Trace file format v2
1722 ------------------------
1723 The second version of the trace file format was added in Fio version 1.17.
1724 It allows to access more then one file per trace and has a bigger set of
1725 possible file actions.
1727 The first line of the trace file has to be:
1731 Following this can be lines in two different formats, which are described below.
1733 The file management format:
1737 The filename is given as an absolute path. The action can be one of these:
1739 add Add the given filename to the trace
1740 open Open the file with the given filename. The filename has to have
1741 been added with the add action before.
1742 close Close the file with the given filename. The file has to have been
1746 The file io action format:
1748 filename action offset length
1750 The filename is given as an absolute path, and has to have been added and opened
1751 before it can be used with this format. The offset and length are given in
1752 bytes. The action can be one of these:
1754 wait Wait for 'offset' microseconds. Everything below 100 is discarded.
1755 read Read 'length' bytes beginning from 'offset'
1756 write Write 'length' bytes beginning from 'offset'
1757 sync fsync() the file
1758 datasync fdatasync() the file
1759 trim trim the given file from the given 'offset' for 'length' bytes
1762 9.0 CPU idleness profiling
1763 --------------------------
1764 In some cases, we want to understand CPU overhead in a test. For example,
1765 we test patches for the specific goodness of whether they reduce CPU usage.
1766 fio implements a balloon approach to create a thread per CPU that runs at
1767 idle priority, meaning that it only runs when nobody else needs the cpu.
1768 By measuring the amount of work completed by the thread, idleness of each
1769 CPU can be derived accordingly.
1771 An unit work is defined as touching a full page of unsigned characters. Mean
1772 and standard deviation of time to complete an unit work is reported in "unit
1773 work" section. Options can be chosen to report detailed percpu idleness or
1774 overall system idleness by aggregating percpu stats.
projects / fio.git / blob