8 5. Detailed list of parameters
13 1.0 Overview and history
14 ------------------------
15 fio was originally written to save me the hassle of writing special test
16 case programs when I wanted to test a specific workload, either for
17 performance reasons or to find/reproduce a bug. The process of writing
18 such a test app can be tiresome, especially if you have to do it often.
19 Hence I needed a tool that would be able to simulate a given io workload
20 without resorting to writing a tailored test case again and again.
22 A test work load is difficult to define, though. There can be any number
23 of processes or threads involved, and they can each be using their own
24 way of generating io. You could have someone dirtying large amounts of
25 memory in an memory mapped file, or maybe several threads issuing
26 reads using asynchronous io. fio needed to be flexible enough to
27 simulate both of these cases, and many more.
31 The first step in getting fio to simulate a desired io workload, is
32 writing a job file describing that specific setup. A job file may contain
33 any number of threads and/or files - the typical contents of the job file
34 is a global section defining shared parameters, and one or more job
35 sections describing the jobs involved. When run, fio parses this file
36 and sets everything up as described. If we break down a job from top to
37 bottom, it contains the following basic parameters:
39 IO type Defines the io pattern issued to the file(s).
40 We may only be reading sequentially from this
41 file(s), or we may be writing randomly. Or even
42 mixing reads and writes, sequentially or randomly.
44 Block size In how large chunks are we issuing io? This may be
45 a single value, or it may describe a range of
48 IO size How much data are we going to be reading/writing.
50 IO engine How do we issue io? We could be memory mapping the
51 file, we could be using regular read/write, we
52 could be using splice, async io, syslet, or even
55 IO depth If the io engine is async, how large a queuing
56 depth do we want to maintain?
58 IO type Should we be doing buffered io, or direct/raw io?
60 Num files How many files are we spreading the workload over.
62 Num threads How many threads or processes should we spread
65 The above are the basic parameters defined for a workload, in addition
66 there's a multitude of parameters that modify other aspects of how this
72 See the README file for command line parameters, there are only a few
75 Running fio is normally the easiest part - you just give it the job file
76 (or job files) as parameters:
80 and it will start doing what the job_file tells it to do. You can give
81 more than one job file on the command line, fio will serialize the running
82 of those files. Internally that is the same as using the 'stonewall'
83 parameter described the the parameter section.
85 If the job file contains only one job, you may as well just give the
86 parameters on the command line. The command line parameters are identical
87 to the job parameters, with a few extra that control global parameters
88 (see README). For example, for the job file parameter iodepth=2, the
89 mirror command line option would be --iodepth 2 or --iodepth=2. You can
90 also use the command line for giving more than one job entry. For each
91 --name option that fio sees, it will start a new job with that name.
92 Command line entries following a --name entry will apply to that job,
93 until there are no more entries or a new --name entry is seen. This is
94 similar to the job file options, where each option applies to the current
95 job until a new [] job entry is seen.
97 fio does not need to run as root, except if the files or devices specified
98 in the job section requires that. Some other options may also be restricted,
99 such as memory locking, io scheduler switching, and decreasing the nice value.
104 As previously described, fio accepts one or more job files describing
105 what it is supposed to do. The job file format is the classic ini file,
106 where the names enclosed in [] brackets define the job name. You are free
107 to use any ascii name you want, except 'global' which has special meaning.
108 A global section sets defaults for the jobs described in that file. A job
109 may override a global section parameter, and a job file may even have
110 several global sections if so desired. A job is only affected by a global
111 section residing above it. If the first character in a line is a ';' or a
112 '#', the entire line is discarded as a comment.
114 So let's look at a really simple job file that defines two processes, each
115 randomly reading from a 128MB file.
117 ; -- start job file --
128 As you can see, the job file sections themselves are empty as all the
129 described parameters are shared. As no filename= option is given, fio
130 makes up a filename for each of the jobs as it sees fit. On the command
131 line, this job would look as follows:
133 $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
136 Let's look at an example that has a number of processes writing randomly
139 ; -- start job file --
151 Here we have no global section, as we only have one job defined anyway.
152 We want to use async io here, with a depth of 4 for each file. We also
153 increased the buffer size used to 32KB and define numjobs to 4 to
154 fork 4 identical jobs. The result is 4 processes each randomly writing
155 to their own 64MB file. Instead of using the above job file, you could
156 have given the parameters on the command line. For this case, you would
159 $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
161 4.1 Environment variables
162 -------------------------
164 fio also supports environment variable expansion in job files. Any
165 substring of the form "${VARNAME}" as part of an option value (in other
166 words, on the right of the `='), will be expanded to the value of the
167 environment variable called VARNAME. If no such environment variable
168 is defined, or VARNAME is the empty string, the empty string will be
171 As an example, let's look at a sample fio invocation and job file:
173 $ SIZE=64m NUMJOBS=4 fio jobfile.fio
175 ; -- start job file --
182 This will expand to the following equivalent job file at runtime:
184 ; -- start job file --
191 fio ships with a few example job files, you can also look there for
194 4.2 Reserved keywords
195 ---------------------
197 Additionally, fio has a set of reserved keywords that will be replaced
198 internally with the appropriate value. Those keywords are:
200 $pagesize The architecture page size of the running system
201 $mb_memory Megabytes of total memory in the system
202 $ncpus Number of online available CPUs
204 These can be used on the command line or in the job file, and will be
205 automatically substituted with the current system values when the job
206 is run. Simple math is also supported on these keywords, so you can
207 perform actions like:
211 and get that properly expanded to 8 times the size of memory in the
215 5.0 Detailed list of parameters
216 -------------------------------
218 This section describes in details each parameter associated with a job.
219 Some parameters take an option of a given type, such as an integer or
220 a string. The following types are used:
222 str String. This is a sequence of alpha characters.
223 time Integer with possible time suffix. In seconds unless otherwise
224 specified, use eg 10m for 10 minutes. Accepts s/m/h for seconds,
226 int SI integer. A whole number value, which may contain a suffix
227 describing the base of the number. Accepted suffixes are k/m/g/t/p,
228 meaning kilo, mega, giga, tera, and peta. The suffix is not case
229 sensitive, and you may also include trailing 'b' (eg 'kb' is the same
230 as 'k'). So if you want to specify 4096, you could either write
231 out '4096' or just give 4k. The suffixes signify base 2 values, so
232 1024 is 1k and 1024k is 1m and so on, unless the suffix is explicitly
233 set to a base 10 value using 'kib', 'mib', 'gib', etc. If that is the
234 case, then 1000 is used as the multiplier. This can be handy for
235 disks, since manufacturers generally use base 10 values when listing
236 the capacity of a drive. If the option accepts an upper and lower
237 range, use a colon ':' or minus '-' to separate such values. May also
238 include a prefix to indicate numbers base. If 0x is used, the number
239 is assumed to be hexadecimal. See irange.
240 bool Boolean. Usually parsed as an integer, however only defined for
241 true and false (1 and 0).
242 irange Integer range with suffix. Allows value range to be given, such
243 as 1024-4096. A colon may also be used as the separator, eg
244 1k:4k. If the option allows two sets of ranges, they can be
245 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
248 With the above in mind, here follows the complete list of fio job
251 name=str ASCII name of the job. This may be used to override the
252 name printed by fio for this job. Otherwise the job
253 name is used. On the command line this parameter has the
254 special purpose of also signaling the start of a new
257 description=str Text description of the job. Doesn't do anything except
258 dump this text description when this job is run. It's
261 directory=str Prefix filenames with this directory. Used to place files
262 in a different location than "./".
264 filename=str Fio normally makes up a filename based on the job name,
265 thread number, and file number. If you want to share
266 files between threads in a job or several jobs, specify
267 a filename for each of them to override the default. If
268 the ioengine used is 'net', the filename is the host, port,
269 and protocol to use in the format of =host/port/protocol.
270 See ioengine=net for more. If the ioengine is file based, you
271 can specify a number of files by separating the names with a
272 ':' colon. So if you wanted a job to open /dev/sda and /dev/sdb
273 as the two working files, you would use
274 filename=/dev/sda:/dev/sdb. On Windows, disk devices are accessed
275 as \\.\PhysicalDrive0 for the first device, \\.\PhysicalDrive1
276 for the second etc. If the wanted filename does need to
277 include a colon, then escape that with a '\' character.
278 For instance, if the filename is "/dev/dsk/foo@3,0:c",
279 then you would use filename="/dev/dsk/foo@3,0\:c".
280 '-' is a reserved name, meaning stdin or stdout. Which of the
281 two depends on the read/write direction set.
283 opendir=str Tell fio to recursively add any file it can find in this
284 directory and down the file system tree.
286 lockfile=str Fio defaults to not locking any files before it does
287 IO to them. If a file or file descriptor is shared, fio
288 can serialize IO to that file to make the end result
289 consistent. This is usual for emulating real workloads that
290 share files. The lock modes are:
292 none No locking. The default.
293 exclusive Only one thread/process may do IO,
294 excluding all others.
295 readwrite Read-write locking on the file. Many
296 readers may access the file at the
297 same time, but writes get exclusive
300 The option may be post-fixed with a lock batch number. If
301 set, then each thread/process may do that amount of IOs to
302 the file before giving up the lock. Since lock acquisition is
303 expensive, batching the lock/unlocks will speed up IO.
306 rw=str Type of io pattern. Accepted values are:
308 read Sequential reads
309 write Sequential writes
310 randwrite Random writes
311 randread Random reads
312 rw Sequential mixed reads and writes
313 randrw Random mixed reads and writes
315 For the mixed io types, the default is to split them 50/50.
316 For certain types of io the result may still be skewed a bit,
317 since the speed may be different. It is possible to specify
318 a number of IO's to do before getting a new offset, this is
319 one by appending a ':<nr>' to the end of the string given.
320 For a random read, it would look like 'rw=randread:8' for
321 passing in an offset modifier with a value of 8. See the
322 'rw_sequencer' option.
324 rw_sequencer=str If an offset modifier is given by appending a number to
325 the rw=<str> line, then this option controls how that
326 number modifies the IO offset being generated. Accepted
329 sequential Generate sequential offset
330 identical Generate the same offset
332 'sequential' is only useful for random IO, where fio would
333 normally generate a new random offset for every IO. If you
334 append eg 8 to randread, you would get a new random offset for
335 every 8 IO's. The result would be a seek for only every 8
336 IO's, instead of for every IO. Use rw=randread:8 to specify
337 that. As sequential IO is already sequential, setting
338 'sequential' for that would not result in any differences.
339 'identical' behaves in a similar fashion, except it sends
340 the same offset 8 number of times before generating a new
343 kb_base=int The base unit for a kilobyte. The defacto base is 2^10, 1024.
344 Storage manufacturers like to use 10^3 or 1000 as a base
345 ten unit instead, for obvious reasons. Allow values are
346 1024 or 1000, with 1024 being the default.
348 randrepeat=bool For random IO workloads, seed the generator in a predictable
349 way so that results are repeatable across repetitions.
351 use_os_rand=bool Fio can either use the random generator supplied by the OS
352 to generator random offsets, or it can use it's own internal
353 generator (based on Tausworthe). Default is to use the
354 internal generator, which is often of better quality and
357 fallocate=str Whether pre-allocation is performed when laying down files.
360 none Do not pre-allocate space
361 posix Pre-allocate via posix_fallocate()
362 keep Pre-allocate via fallocate() with
363 FALLOC_FL_KEEP_SIZE set
364 0 Backward-compatible alias for 'none'
365 1 Backward-compatible alias for 'posix'
367 May not be available on all supported platforms. 'keep' is only
368 available on Linux.If using ZFS on Solaris this must be set to
369 'none' because ZFS doesn't support it. Default: 'posix'.
371 fadvise_hint=bool By default, fio will use fadvise() to advise the kernel
372 on what IO patterns it is likely to issue. Sometimes you
373 want to test specific IO patterns without telling the
374 kernel about it, in which case you can disable this option.
375 If set, fio will use POSIX_FADV_SEQUENTIAL for sequential
376 IO and POSIX_FADV_RANDOM for random IO.
378 size=int The total size of file io for this job. Fio will run until
379 this many bytes has been transferred, unless runtime is
380 limited by other options (such as 'runtime', for instance).
381 Unless specific nrfiles and filesize options are given,
382 fio will divide this size between the available files
383 specified by the job. If not set, fio will use the full
384 size of the given files or devices. If the the files
385 do not exist, size must be given. It is also possible to
386 give size as a percentage between 1 and 100. If size=20%
387 is given, fio will use 20% of the full size of the given
390 filesize=int Individual file sizes. May be a range, in which case fio
391 will select sizes for files at random within the given range
392 and limited to 'size' in total (if that is given). If not
393 given, each created file is the same size.
396 fill_fs=bool Sets size to something really large and waits for ENOSPC (no
397 space left on device) as the terminating condition. Only makes
398 sense with sequential write. For a read workload, the mount
399 point will be filled first then IO started on the result. This
400 option doesn't make sense if operating on a raw device node,
401 since the size of that is already known by the file system.
402 Additionally, writing beyond end-of-device will not return
406 bs=int The block size used for the io units. Defaults to 4k. Values
407 can be given for both read and writes. If a single int is
408 given, it will apply to both. If a second int is specified
409 after a comma, it will apply to writes only. In other words,
410 the format is either bs=read_and_write or bs=read,write.
411 bs=4k,8k will thus use 4k blocks for reads, and 8k blocks
412 for writes. If you only wish to set the write size, you
413 can do so by passing an empty read size - bs=,8k will set
414 8k for writes and leave the read default value.
417 ba=int At what boundary to align random IO offsets. Defaults to
418 the same as 'blocksize' the minimum blocksize given.
419 Minimum alignment is typically 512b for using direct IO,
420 though it usually depends on the hardware block size. This
421 option is mutually exclusive with using a random map for
422 files, so it will turn off that option.
424 blocksize_range=irange
425 bsrange=irange Instead of giving a single block size, specify a range
426 and fio will mix the issued io block sizes. The issued
427 io unit will always be a multiple of the minimum value
428 given (also see bs_unaligned). Applies to both reads and
429 writes, however a second range can be given after a comma.
432 bssplit=str Sometimes you want even finer grained control of the
433 block sizes issued, not just an even split between them.
434 This option allows you to weight various block sizes,
435 so that you are able to define a specific amount of
436 block sizes issued. The format for this option is:
438 bssplit=blocksize/percentage:blocksize/percentage
440 for as many block sizes as needed. So if you want to define
441 a workload that has 50% 64k blocks, 10% 4k blocks, and
442 40% 32k blocks, you would write:
444 bssplit=4k/10:64k/50:32k/40
446 Ordering does not matter. If the percentage is left blank,
447 fio will fill in the remaining values evenly. So a bssplit
448 option like this one:
450 bssplit=4k/50:1k/:32k/
452 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages
453 always add up to 100, if bssplit is given a range that adds
454 up to more, it will error out.
456 bssplit also supports giving separate splits to reads and
457 writes. The format is identical to what bs= accepts. You
458 have to separate the read and write parts with a comma. So
459 if you want a workload that has 50% 2k reads and 50% 4k reads,
460 while having 90% 4k writes and 10% 8k writes, you would
463 bssplit=2k/50:4k/50,4k/90,8k/10
466 bs_unaligned If this option is given, any byte size value within bsrange
467 may be used as a block range. This typically wont work with
468 direct IO, as that normally requires sector alignment.
470 zero_buffers If this option is given, fio will init the IO buffers to
471 all zeroes. The default is to fill them with random data.
473 refill_buffers If this option is given, fio will refill the IO buffers
474 on every submit. The default is to only fill it at init
475 time and reuse that data. Only makes sense if zero_buffers
476 isn't specified, naturally. If data verification is enabled,
477 refill_buffers is also automatically enabled.
479 nrfiles=int Number of files to use for this job. Defaults to 1.
481 openfiles=int Number of files to keep open at the same time. Defaults to
482 the same as nrfiles, can be set smaller to limit the number
485 file_service_type=str Defines how fio decides which file from a job to
486 service next. The following types are defined:
488 random Just choose a file at random.
490 roundrobin Round robin over open files. This
493 sequential Finish one file before moving on to
494 the next. Multiple files can still be
495 open depending on 'openfiles'.
497 The string can have a number appended, indicating how
498 often to switch to a new file. So if option random:4 is
499 given, fio will switch to a new random file after 4 ios
502 ioengine=str Defines how the job issues io to the file. The following
505 sync Basic read(2) or write(2) io. lseek(2) is
506 used to position the io location.
508 psync Basic pread(2) or pwrite(2) io.
510 vsync Basic readv(2) or writev(2) IO.
512 libaio Linux native asynchronous io. Note that Linux
513 may only support queued behaviour with
514 non-buffered IO (set direct=1 or buffered=0).
516 posixaio glibc posix asynchronous io.
518 solarisaio Solaris native asynchronous io.
520 windowsaio Windows native asynchronous io.
522 mmap File is memory mapped and data copied
523 to/from using memcpy(3).
525 splice splice(2) is used to transfer the data and
526 vmsplice(2) to transfer data from user
529 syslet-rw Use the syslet system calls to make
530 regular read/write async.
532 sg SCSI generic sg v3 io. May either be
533 synchronous using the SG_IO ioctl, or if
534 the target is an sg character device
535 we use read(2) and write(2) for asynchronous
538 null Doesn't transfer any data, just pretends
539 to. This is mainly used to exercise fio
540 itself and for debugging/testing purposes.
542 net Transfer over the network to given host:port.
543 'filename' must be set appropriately to
544 filename=host/port/protocol regardless of send
545 or receive, if the latter only the port
546 argument is used. 'host' may be an IP address
547 or hostname, port is the port number to be used,
548 and protocol may be 'udp' or 'tcp'. If no
549 protocol is given, TCP is used.
551 netsplice Like net, but uses splice/vmsplice to
552 map data and send/receive.
554 cpuio Doesn't transfer any data, but burns CPU
555 cycles according to the cpuload= and
556 cpucycle= options. Setting cpuload=85
557 will cause that job to do nothing but burn
558 85% of the CPU. In case of SMP machines,
559 use numjobs=<no_of_cpu> to get desired CPU
560 usage, as the cpuload only loads a single
561 CPU at the desired rate.
563 guasi The GUASI IO engine is the Generic Userspace
564 Asyncronous Syscall Interface approach
567 http://www.xmailserver.org/guasi-lib.html
569 for more info on GUASI.
571 rdma The RDMA I/O engine supports both RDMA
572 memory semantic(RDMA_WRITE/RDMA_READ) and
573 channel semantic(Send/Recv) in InfiniBand, RoCE
574 and iWarp environment.
576 external Prefix to specify loading an external
577 IO engine object file. Append the engine
578 filename, eg ioengine=external:/tmp/foo.o
579 to load ioengine foo.o in /tmp.
581 iodepth=int This defines how many io units to keep in flight against
582 the file. The default is 1 for each file defined in this
583 job, can be overridden with a larger value for higher
584 concurrency. Note that increasing iodepth beyond 1 will not
585 affect synchronous ioengines (except for small degress when
586 verify_async is in use). Even async engines may impose OS
587 restrictions causing the desired depth not to be achieved.
588 This may happen on Linux when using libaio and not setting
589 direct=1, since buffered IO is not async on that OS. Keep an
590 eye on the IO depth distribution in the fio output to verify
591 that the achieved depth is as expected. Default: 1.
593 iodepth_batch_submit=int
594 iodepth_batch=int This defines how many pieces of IO to submit at once.
595 It defaults to 1 which means that we submit each IO
596 as soon as it is available, but can be raised to submit
597 bigger batches of IO at the time.
599 iodepth_batch_complete=int This defines how many pieces of IO to retrieve
600 at once. It defaults to 1 which means that we'll ask
601 for a minimum of 1 IO in the retrieval process from
602 the kernel. The IO retrieval will go on until we
603 hit the limit set by iodepth_low. If this variable is
604 set to 0, then fio will always check for completed
605 events before queuing more IO. This helps reduce
606 IO latency, at the cost of more retrieval system calls.
608 iodepth_low=int The low water mark indicating when to start filling
609 the queue again. Defaults to the same as iodepth, meaning
610 that fio will attempt to keep the queue full at all times.
611 If iodepth is set to eg 16 and iodepth_low is set to 4, then
612 after fio has filled the queue of 16 requests, it will let
613 the depth drain down to 4 before starting to fill it again.
615 direct=bool If value is true, use non-buffered io. This is usually
616 O_DIRECT. Note that ZFS on Solaris doesn't support direct io.
618 buffered=bool If value is true, use buffered io. This is the opposite
619 of the 'direct' option. Defaults to true.
621 offset=int Start io at the given offset in the file. The data before
622 the given offset will not be touched. This effectively
623 caps the file size at real_size - offset.
625 fsync=int If writing to a file, issue a sync of the dirty data
626 for every number of blocks given. For example, if you give
627 32 as a parameter, fio will sync the file for every 32
628 writes issued. If fio is using non-buffered io, we may
629 not sync the file. The exception is the sg io engine, which
630 synchronizes the disk cache anyway.
632 fdatasync=int Like fsync= but uses fdatasync() to only sync data and not
634 In FreeBSD there is no fdatasync(), this falls back to
637 sync_file_range=str:val Use sync_file_range() for every 'val' number of
638 write operations. Fio will track range of writes that
639 have happened since the last sync_file_range() call. 'str'
640 can currently be one or more of:
642 wait_before SYNC_FILE_RANGE_WAIT_BEFORE
643 write SYNC_FILE_RANGE_WRITE
644 wait_after SYNC_FILE_RANGE_WAIT_AFTER
646 So if you do sync_file_range=wait_before,write:8, fio would
647 use SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE for
648 every 8 writes. Also see the sync_file_range(2) man page.
649 This option is Linux specific.
651 overwrite=bool If true, writes to a file will always overwrite existing
652 data. If the file doesn't already exist, it will be
653 created before the write phase begins. If the file exists
654 and is large enough for the specified write phase, nothing
657 end_fsync=bool If true, fsync file contents when the job exits.
659 fsync_on_close=bool If true, fio will fsync() a dirty file on close.
660 This differs from end_fsync in that it will happen on every
661 file close, not just at the end of the job.
663 rwmixread=int How large a percentage of the mix should be reads.
665 rwmixwrite=int How large a percentage of the mix should be writes. If both
666 rwmixread and rwmixwrite is given and the values do not add
667 up to 100%, the latter of the two will be used to override
668 the first. This may interfere with a given rate setting,
669 if fio is asked to limit reads or writes to a certain rate.
670 If that is the case, then the distribution may be skewed.
672 norandommap Normally fio will cover every block of the file when doing
673 random IO. If this option is given, fio will just get a
674 new random offset without looking at past io history. This
675 means that some blocks may not be read or written, and that
676 some blocks may be read/written more than once. This option
677 is mutually exclusive with verify= if and only if multiple
678 blocksizes (via bsrange=) are used, since fio only tracks
679 complete rewrites of blocks.
681 softrandommap See norandommap. If fio runs with the random block map enabled
682 and it fails to allocate the map, if this option is set it
683 will continue without a random block map. As coverage will
684 not be as complete as with random maps, this option is
687 nice=int Run the job with the given nice value. See man nice(2).
689 prio=int Set the io priority value of this job. Linux limits us to
690 a positive value between 0 and 7, with 0 being the highest.
693 prioclass=int Set the io priority class. See man ionice(1).
695 thinktime=int Stall the job x microseconds after an io has completed before
696 issuing the next. May be used to simulate processing being
697 done by an application. See thinktime_blocks and
701 Only valid if thinktime is set - pretend to spend CPU time
702 doing something with the data received, before falling back
703 to sleeping for the rest of the period specified by
707 Only valid if thinktime is set - control how many blocks
708 to issue, before waiting 'thinktime' usecs. If not set,
709 defaults to 1 which will make fio wait 'thinktime' usecs
712 rate=int Cap the bandwidth used by this job. The number is in bytes/sec,
713 the normal suffix rules apply. You can use rate=500k to limit
714 reads and writes to 500k each, or you can specify read and
715 writes separately. Using rate=1m,500k would limit reads to
716 1MB/sec and writes to 500KB/sec. Capping only reads or
717 writes can be done with rate=,500k or rate=500k,. The former
718 will only limit writes (to 500KB/sec), the latter will only
721 ratemin=int Tell fio to do whatever it can to maintain at least this
722 bandwidth. Failing to meet this requirement, will cause
723 the job to exit. The same format as rate is used for
724 read vs write separation.
726 rate_iops=int Cap the bandwidth to this number of IOPS. Basically the same
727 as rate, just specified independently of bandwidth. If the
728 job is given a block size range instead of a fixed value,
729 the smallest block size is used as the metric. The same format
730 as rate is used for read vs write seperation.
732 rate_iops_min=int If fio doesn't meet this rate of IO, it will cause
733 the job to exit. The same format as rate is used for read vs
736 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
739 cpumask=int Set the CPU affinity of this job. The parameter given is a
740 bitmask of allowed CPU's the job may run on. So if you want
741 the allowed CPUs to be 1 and 5, you would pass the decimal
742 value of (1 << 1 | 1 << 5), or 34. See man
743 sched_setaffinity(2). This may not work on all supported
744 operating systems or kernel versions. This option doesn't
745 work well for a higher CPU count than what you can store in
746 an integer mask, so it can only control cpus 1-32. For
747 boxes with larger CPU counts, use cpus_allowed.
749 cpus_allowed=str Controls the same options as cpumask, but it allows a text
750 setting of the permitted CPUs instead. So to use CPUs 1 and
751 5, you would specify cpus_allowed=1,5. This options also
752 allows a range of CPUs. Say you wanted a binding to CPUs
753 1, 5, and 8-15, you would set cpus_allowed=1,5,8-15.
755 startdelay=time Start this job the specified number of seconds after fio
756 has started. Only useful if the job file contains several
757 jobs, and you want to delay starting some jobs to a certain
760 runtime=time Tell fio to terminate processing after the specified number
761 of seconds. It can be quite hard to determine for how long
762 a specified job will run, so this parameter is handy to
763 cap the total runtime to a given time.
765 time_based If set, fio will run for the duration of the runtime
766 specified even if the file(s) are completely read or
767 written. It will simply loop over the same workload
768 as many times as the runtime allows.
770 ramp_time=time If set, fio will run the specified workload for this amount
771 of time before logging any performance numbers. Useful for
772 letting performance settle before logging results, thus
773 minimizing the runtime required for stable results. Note
774 that the ramp_time is considered lead in time for a job,
775 thus it will increase the total runtime if a special timeout
776 or runtime is specified.
778 invalidate=bool Invalidate the buffer/page cache parts for this file prior
779 to starting io. Defaults to true.
781 sync=bool Use sync io for buffered writes. For the majority of the
782 io engines, this means using O_SYNC.
785 mem=str Fio can use various types of memory as the io unit buffer.
786 The allowed values are:
788 malloc Use memory from malloc(3) as the buffers.
790 shm Use shared memory as the buffers. Allocated
793 shmhuge Same as shm, but use huge pages as backing.
795 mmap Use mmap to allocate buffers. May either be
796 anonymous memory, or can be file backed if
797 a filename is given after the option. The
798 format is mem=mmap:/path/to/file.
800 mmaphuge Use a memory mapped huge file as the buffer
801 backing. Append filename after mmaphuge, ala
802 mem=mmaphuge:/hugetlbfs/file
804 The area allocated is a function of the maximum allowed
805 bs size for the job, multiplied by the io depth given. Note
806 that for shmhuge and mmaphuge to work, the system must have
807 free huge pages allocated. This can normally be checked
808 and set by reading/writing /proc/sys/vm/nr_hugepages on a
809 Linux system. Fio assumes a huge page is 4MB in size. So
810 to calculate the number of huge pages you need for a given
811 job file, add up the io depth of all jobs (normally one unless
812 iodepth= is used) and multiply by the maximum bs set. Then
813 divide that number by the huge page size. You can see the
814 size of the huge pages in /proc/meminfo. If no huge pages
815 are allocated by having a non-zero number in nr_hugepages,
816 using mmaphuge or shmhuge will fail. Also see hugepage-size.
818 mmaphuge also needs to have hugetlbfs mounted and the file
819 location should point there. So if it's mounted in /huge,
820 you would use mem=mmaphuge:/huge/somefile.
822 iomem_align=int This indiciates the memory alignment of the IO memory buffers.
823 Note that the given alignment is applied to the first IO unit
824 buffer, if using iodepth the alignment of the following buffers
825 are given by the bs used. In other words, if using a bs that is
826 a multiple of the page sized in the system, all buffers will
827 be aligned to this value. If using a bs that is not page
828 aligned, the alignment of subsequent IO memory buffers is the
829 sum of the iomem_align and bs used.
832 Defines the size of a huge page. Must at least be equal
833 to the system setting, see /proc/meminfo. Defaults to 4MB.
834 Should probably always be a multiple of megabytes, so using
835 hugepage-size=Xm is the preferred way to set this to avoid
836 setting a non-pow-2 bad value.
838 exitall When one job finishes, terminate the rest. The default is
839 to wait for each job to finish, sometimes that is not the
842 bwavgtime=int Average the calculated bandwidth over the given time. Value
843 is specified in milliseconds.
845 create_serialize=bool If true, serialize the file creating for the jobs.
846 This may be handy to avoid interleaving of data
847 files, which may greatly depend on the filesystem
848 used and even the number of processors in the system.
850 create_fsync=bool fsync the data file after creation. This is the
853 create_on_open=bool Don't pre-setup the files for IO, just create open()
854 when it's time to do IO to that file.
856 pre_read=bool If this is given, files will be pre-read into memory before
857 starting the given IO operation. This will also clear
858 the 'invalidate' flag, since it is pointless to pre-read
859 and then drop the cache. This will only work for IO engines
860 that are seekable, since they allow you to read the same data
861 multiple times. Thus it will not work on eg network or splice
864 unlink=bool Unlink the job files when done. Not the default, as repeated
865 runs of that job would then waste time recreating the file
868 loops=int Run the specified number of iterations of this job. Used
869 to repeat the same workload a given number of times. Defaults
872 do_verify=bool Run the verify phase after a write phase. Only makes sense if
873 verify is set. Defaults to 1.
875 verify=str If writing to a file, fio can verify the file contents
876 after each iteration of the job. The allowed values are:
878 md5 Use an md5 sum of the data area and store
879 it in the header of each block.
881 crc64 Use an experimental crc64 sum of the data
882 area and store it in the header of each
885 crc32c Use a crc32c sum of the data area and store
886 it in the header of each block.
888 crc32c-intel Use hardware assisted crc32c calcuation
889 provided on SSE4.2 enabled processors. Falls
890 back to regular software crc32c, if not
891 supported by the system.
893 crc32 Use a crc32 sum of the data area and store
894 it in the header of each block.
896 crc16 Use a crc16 sum of the data area and store
897 it in the header of each block.
899 crc7 Use a crc7 sum of the data area and store
900 it in the header of each block.
902 sha512 Use sha512 as the checksum function.
904 sha256 Use sha256 as the checksum function.
906 sha1 Use optimized sha1 as the checksum function.
908 meta Write extra information about each io
909 (timestamp, block number etc.). The block
910 number is verified. See also verify_pattern.
912 null Only pretend to verify. Useful for testing
913 internals with ioengine=null, not for much
916 This option can be used for repeated burn-in tests of a
917 system to make sure that the written data is also
918 correctly read back. If the data direction given is
919 a read or random read, fio will assume that it should
920 verify a previously written file. If the data direction
921 includes any form of write, the verify will be of the
924 verifysort=bool If set, fio will sort written verify blocks when it deems
925 it faster to read them back in a sorted manner. This is
926 often the case when overwriting an existing file, since
927 the blocks are already laid out in the file system. You
928 can ignore this option unless doing huge amounts of really
929 fast IO where the red-black tree sorting CPU time becomes
932 verify_offset=int Swap the verification header with data somewhere else
933 in the block before writing. Its swapped back before
936 verify_interval=int Write the verification header at a finer granularity
937 than the blocksize. It will be written for chunks the
938 size of header_interval. blocksize should divide this
941 verify_pattern=str If set, fio will fill the io buffers with this
942 pattern. Fio defaults to filling with totally random
943 bytes, but sometimes it's interesting to fill with a known
944 pattern for io verification purposes. Depending on the
945 width of the pattern, fio will fill 1/2/3/4 bytes of the
946 buffer at the time(it can be either a decimal or a hex number).
947 The verify_pattern if larger than a 32-bit quantity has to
948 be a hex number that starts with either "0x" or "0X". Use
951 verify_fatal=bool Normally fio will keep checking the entire contents
952 before quitting on a block verification failure. If this
953 option is set, fio will exit the job on the first observed
956 verify_dump=bool If set, dump the contents of both the original data
957 block and the data block we read off disk to files. This
958 allows later analysis to inspect just what kind of data
959 corruption occurred. On by default.
961 verify_async=int Fio will normally verify IO inline from the submitting
962 thread. This option takes an integer describing how many
963 async offload threads to create for IO verification instead,
964 causing fio to offload the duty of verifying IO contents
965 to one or more separate threads. If using this offload
966 option, even sync IO engines can benefit from using an
967 iodepth setting higher than 1, as it allows them to have
968 IO in flight while verifies are running.
970 verify_async_cpus=str Tell fio to set the given CPU affinity on the
971 async IO verification threads. See cpus_allowed for the
974 verify_backlog=int Fio will normally verify the written contents of a
975 job that utilizes verify once that job has completed. In
976 other words, everything is written then everything is read
977 back and verified. You may want to verify continually
978 instead for a variety of reasons. Fio stores the meta data
979 associated with an IO block in memory, so for large
980 verify workloads, quite a bit of memory would be used up
981 holding this meta data. If this option is enabled, fio
982 will write only N blocks before verifying these blocks.
984 will verify the previously written blocks before continuing
987 verify_backlog_batch=int Control how many blocks fio will verify
988 if verify_backlog is set. If not set, will default to
989 the value of verify_backlog (meaning the entire queue
990 is read back and verified). If verify_backlog_batch is
991 less than verify_backlog then not all blocks will be verified,
992 if verify_backlog_batch is larger than verify_backlog, some
993 blocks will be verified more than once.
996 wait_for_previous Wait for preceeding jobs in the job file to exit, before
997 starting this one. Can be used to insert serialization
998 points in the job file. A stone wall also implies starting
999 a new reporting group.
1001 new_group Start a new reporting group. If this option isn't given,
1002 jobs in a file will be part of the same reporting group
1003 unless separated by a stone wall (or if it's a group
1004 by itself, with the numjobs option).
1006 numjobs=int Create the specified number of clones of this job. May be
1007 used to setup a larger number of threads/processes doing
1008 the same thing. We regard that grouping of jobs as a
1011 group_reporting If 'numjobs' is set, it may be interesting to display
1012 statistics for the group as a whole instead of for each
1013 individual job. This is especially true of 'numjobs' is
1014 large, looking at individual thread/process output quickly
1015 becomes unwieldy. If 'group_reporting' is specified, fio
1016 will show the final report per-group instead of per-job.
1018 thread fio defaults to forking jobs, however if this option is
1019 given, fio will use pthread_create(3) to create threads
1022 zonesize=int Divide a file into zones of the specified size. See zoneskip.
1024 zoneskip=int Skip the specified number of bytes when zonesize data has
1025 been read. The two zone options can be used to only do
1026 io on zones of a file.
1028 write_iolog=str Write the issued io patterns to the specified file. See
1029 read_iolog. Specify a separate file for each job, otherwise
1030 the iologs will be interspersed and the file may be corrupt.
1032 read_iolog=str Open an iolog with the specified file name and replay the
1033 io patterns it contains. This can be used to store a
1034 workload and replay it sometime later. The iolog given
1035 may also be a blktrace binary file, which allows fio
1036 to replay a workload captured by blktrace. See blktrace
1037 for how to capture such logging data. For blktrace replay,
1038 the file needs to be turned into a blkparse binary data
1039 file first (blkparse <device> -o /dev/null -d file_for_fio.bin).
1041 replay_no_stall=int When replaying I/O with read_iolog the default behavior
1042 is to attempt to respect the time stamps within the log and
1043 replay them with the appropriate delay between IOPS. By
1044 setting this variable fio will not respect the timestamps and
1045 attempt to replay them as fast as possible while still
1046 respecting ordering. The result is the same I/O pattern to a
1047 given device, but different timings.
1049 replay_redirect=str While replaying I/O patterns using read_iolog the
1050 default behavior is to replay the IOPS onto the major/minor
1051 device that each IOP was recorded from. This is sometimes
1052 undesireable because on a different machine those major/minor
1053 numbers can map to a different device. Changing hardware on
1054 the same system can also result in a different major/minor
1055 mapping. Replay_redirect causes all IOPS to be replayed onto
1056 the single specified device regardless of the device it was
1057 recorded from. i.e. replay_redirect=/dev/sdc would cause all
1058 IO in the blktrace to be replayed onto /dev/sdc. This means
1059 multiple devices will be replayed onto a single, if the trace
1060 contains multiple devices. If you want multiple devices to be
1061 replayed concurrently to multiple redirected devices you must
1062 blkparse your trace into separate traces and replay them with
1063 independent fio invocations. Unfortuantely this also breaks
1064 the strict time ordering between multiple device accesses.
1066 write_bw_log=str If given, write a bandwidth log of the jobs in this job
1067 file. Can be used to store data of the bandwidth of the
1068 jobs in their lifetime. The included fio_generate_plots
1069 script uses gnuplot to turn these text files into nice
1070 graphs. See write_log_log for behaviour of given
1071 filename. For this option, the postfix is _bw.log.
1073 write_lat_log=str Same as write_bw_log, except that this option stores io
1074 submission, completion, and total latencies instead. If no
1075 filename is given with this option, the default filename of
1076 "jobname_type.log" is used. Even if the filename is given,
1077 fio will still append the type of log. So if one specifies
1081 The actual log names will be foo_slat.log, foo_slat.log,
1082 and foo_lat.log. This helps fio_generate_plot fine the logs
1085 lockmem=int Pin down the specified amount of memory with mlock(2). Can
1086 potentially be used instead of removing memory or booting
1087 with less memory to simulate a smaller amount of memory.
1089 exec_prerun=str Before running this job, issue the command specified
1092 exec_postrun=str After the job completes, issue the command specified
1095 ioscheduler=str Attempt to switch the device hosting the file to the specified
1096 io scheduler before running.
1098 cpuload=int If the job is a CPU cycle eater, attempt to use the specified
1099 percentage of CPU cycles.
1101 cpuchunks=int If the job is a CPU cycle eater, split the load into
1102 cycles of the given time. In microseconds.
1104 disk_util=bool Generate disk utilization statistics, if the platform
1105 supports it. Defaults to on.
1107 disable_lat=bool Disable measurements of total latency numbers. Useful
1108 only for cutting back the number of calls to gettimeofday,
1109 as that does impact performance at really high IOPS rates.
1110 Note that to really get rid of a large amount of these
1111 calls, this option must be used with disable_slat and
1114 disable_clat=bool Disable measurements of completion latency numbers. See
1117 disable_slat=bool Disable measurements of submission latency numbers. See
1120 disable_bw=bool Disable measurements of throughput/bandwidth numbers. See
1123 gtod_reduce=bool Enable all of the gettimeofday() reducing options
1124 (disable_clat, disable_slat, disable_bw) plus reduce
1125 precision of the timeout somewhat to really shrink
1126 the gettimeofday() call count. With this option enabled,
1127 we only do about 0.4% of the gtod() calls we would have
1128 done if all time keeping was enabled.
1130 gtod_cpu=int Sometimes it's cheaper to dedicate a single thread of
1131 execution to just getting the current time. Fio (and
1132 databases, for instance) are very intensive on gettimeofday()
1133 calls. With this option, you can set one CPU aside for
1134 doing nothing but logging current time to a shared memory
1135 location. Then the other threads/processes that run IO
1136 workloads need only copy that segment, instead of entering
1137 the kernel with a gettimeofday() call. The CPU set aside
1138 for doing these time calls will be excluded from other
1139 uses. Fio will manually clear it from the CPU mask of other
1142 continue_on_error=bool Normally fio will exit the job on the first observed
1143 failure. If this option is set, fio will continue the job when
1144 there is a 'non-fatal error' (EIO or EILSEQ) until the runtime
1145 is exceeded or the I/O size specified is completed. If this
1146 option is used, there are two more stats that are appended,
1147 the total error count and the first error. The error field
1148 given in the stats is the first error that was hit during the
1151 cgroup=str Add job to this control group. If it doesn't exist, it will
1152 be created. The system must have a mounted cgroup blkio
1153 mount point for this to work. If your system doesn't have it
1154 mounted, you can do so with:
1156 # mount -t cgroup -o blkio none /cgroup
1158 cgroup_weight=int Set the weight of the cgroup to this value. See
1159 the documentation that comes with the kernel, allowed values
1160 are in the range of 100..1000.
1162 cgroup_nodelete=bool Normally fio will delete the cgroups it has created after
1163 the job completion. To override this behavior and to leave
1164 cgroups around after the job completion, set cgroup_nodelete=1.
1165 This can be useful if one wants to inspect various cgroup
1166 files after job completion. Default: false
1168 uid=int Instead of running as the invoking user, set the user ID to
1169 this value before the thread/process does any work.
1171 gid=int Set group ID, see uid.
1173 6.0 Interpreting the output
1174 ---------------------------
1176 fio spits out a lot of output. While running, fio will display the
1177 status of the jobs created. An example of that would be:
1179 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
1181 The characters inside the square brackets denote the current status of
1182 each thread. The possible values (in typical life cycle order) are:
1186 P Thread setup, but not started.
1188 I Thread initialized, waiting.
1189 p Thread running pre-reading file(s).
1190 R Running, doing sequential reads.
1191 r Running, doing random reads.
1192 W Running, doing sequential writes.
1193 w Running, doing random writes.
1194 M Running, doing mixed sequential reads/writes.
1195 m Running, doing mixed random reads/writes.
1196 F Running, currently waiting for fsync()
1197 V Running, doing verification of written data.
1198 E Thread exited, not reaped by main thread yet.
1201 The other values are fairly self explanatory - number of threads
1202 currently running and doing io, rate of io since last check (read speed
1203 listed first, then write speed), and the estimated completion percentage
1204 and time for the running group. It's impossible to estimate runtime of
1205 the following groups (if any).
1207 When fio is done (or interrupted by ctrl-c), it will show the data for
1208 each thread, group of threads, and disks in that order. For each data
1209 direction, the output looks like:
1211 Client1 (g=0): err= 0:
1212 write: io= 32MB, bw= 666KB/s, iops=89 , runt= 50320msec
1213 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
1214 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
1215 bw (KB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
1216 cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17
1217 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
1218 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
1219 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
1220 issued r/w: total=0/32768, short=0/0
1221 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
1222 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
1224 The client number is printed, along with the group id and error of that
1225 thread. Below is the io statistics, here for writes. In the order listed,
1228 io= Number of megabytes io performed
1229 bw= Average bandwidth rate
1230 iops= Average IOs performed per second
1231 runt= The runtime of that thread
1232 slat= Submission latency (avg being the average, stdev being the
1233 standard deviation). This is the time it took to submit
1234 the io. For sync io, the slat is really the completion
1235 latency, since queue/complete is one operation there. This
1236 value can be in milliseconds or microseconds, fio will choose
1237 the most appropriate base and print that. In the example
1238 above, milliseconds is the best scale.
1239 clat= Completion latency. Same names as slat, this denotes the
1240 time from submission to completion of the io pieces. For
1241 sync io, clat will usually be equal (or very close) to 0,
1242 as the time from submit to complete is basically just
1243 CPU time (io has already been done, see slat explanation).
1244 bw= Bandwidth. Same names as the xlat stats, but also includes
1245 an approximate percentage of total aggregate bandwidth
1246 this thread received in this group. This last value is
1247 only really useful if the threads in this group are on the
1248 same disk, since they are then competing for disk access.
1249 cpu= CPU usage. User and system time, along with the number
1250 of context switches this thread went through, usage of
1251 system and user time, and finally the number of major
1252 and minor page faults.
1253 IO depths= The distribution of io depths over the job life time. The
1254 numbers are divided into powers of 2, so for example the
1255 16= entries includes depths up to that value but higher
1256 than the previous entry. In other words, it covers the
1257 range from 16 to 31.
1258 IO submit= How many pieces of IO were submitting in a single submit
1259 call. Each entry denotes that amount and below, until
1260 the previous entry - eg, 8=100% mean that we submitted
1261 anywhere in between 5-8 ios per submit call.
1262 IO complete= Like the above submit number, but for completions instead.
1263 IO issued= The number of read/write requests issued, and how many
1265 IO latencies= The distribution of IO completion latencies. This is the
1266 time from when IO leaves fio and when it gets completed.
1267 The numbers follow the same pattern as the IO depths,
1268 meaning that 2=1.6% means that 1.6% of the IO completed
1269 within 2 msecs, 20=12.8% means that 12.8% of the IO
1270 took more than 10 msecs, but less than (or equal to) 20 msecs.
1272 After each client has been listed, the group statistics are printed. They
1273 will look like this:
1275 Run status group 0 (all jobs):
1276 READ: io=64MB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
1277 WRITE: io=64MB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
1279 For each data direction, it prints:
1281 io= Number of megabytes io performed.
1282 aggrb= Aggregate bandwidth of threads in this group.
1283 minb= The minimum average bandwidth a thread saw.
1284 maxb= The maximum average bandwidth a thread saw.
1285 mint= The smallest runtime of the threads in that group.
1286 maxt= The longest runtime of the threads in that group.
1288 And finally, the disk statistics are printed. They will look like this:
1290 Disk stats (read/write):
1291 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
1293 Each value is printed for both reads and writes, with reads first. The
1296 ios= Number of ios performed by all groups.
1297 merge= Number of merges io the io scheduler.
1298 ticks= Number of ticks we kept the disk busy.
1299 io_queue= Total time spent in the disk queue.
1300 util= The disk utilization. A value of 100% means we kept the disk
1301 busy constantly, 50% would be a disk idling half of the time.
1307 For scripted usage where you typically want to generate tables or graphs
1308 of the results, fio can output the results in a semicolon separated format.
1309 The format is one long line of values, such as:
1311 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%
1312 A description of this job goes here.
1314 The job description (if provided) follows on a second line.
1316 To enable terse output, use the --minimal command line option. The first
1317 value is the version of the terse output format. If the output has to
1318 be changed for some reason, this number will be incremented by 1 to
1319 signify that change.
1321 Split up, the format is as follows:
1323 version, jobname, groupid, error
1325 KB IO, bandwidth (KB/sec), runtime (msec)
1326 Submission latency: min, max, mean, deviation
1327 Completion latency: min, max, mean, deviation
1328 Total latency: min, max, mean, deviation
1329 Bw: min, max, aggregate percentage of total, mean, deviation
1331 KB IO, bandwidth (KB/sec), runtime (msec)
1332 Submission latency: min, max, mean, deviation
1333 Completion latency: min, max, mean, deviation
1334 Total latency: min, max, mean, deviation
1335 Bw: min, max, aggregate percentage of total, mean, deviation
1336 CPU usage: user, system, context switches, major faults, minor faults
1337 IO depths: <=1, 2, 4, 8, 16, 32, >=64
1338 IO latencies microseconds: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
1339 IO latencies milliseconds: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
1340 Additional Info (dependant on continue_on_error, default off): total # errors, first error code
1342 Additional Info (dependant on description being set): Text description
1345 8.0 Trace file format
1346 ---------------------
1347 There are two trace file format that you can encounter. The older (v1) format
1348 is unsupported since version 1.20-rc3 (March 2008). It will still be described
1349 below in case that you get an old trace and want to understand it.
1351 In any case the trace is a simple text file with a single action per line.
1354 8.1 Trace file format v1
1355 ------------------------
1356 Each line represents a single io action in the following format:
1360 where rw=0/1 for read/write, and the offset and length entries being in bytes.
1362 This format is not supported in Fio versions => 1.20-rc3.
1365 8.2 Trace file format v2
1366 ------------------------
1367 The second version of the trace file format was added in Fio version 1.17.
1368 It allows to access more then one file per trace and has a bigger set of
1369 possible file actions.
1371 The first line of the trace file has to be:
1375 Following this can be lines in two different formats, which are described below.
1377 The file management format:
1381 The filename is given as an absolute path. The action can be one of these:
1383 add Add the given filename to the trace
1384 open Open the file with the given filename. The filename has to have
1385 been added with the add action before.
1386 close Close the file with the given filename. The file has to have been
1390 The file io action format:
1392 filename action offset length
1394 The filename is given as an absolute path, and has to have been added and opened
1395 before it can be used with this format. The offset and length are given in
1396 bytes. The action can be one of these:
1398 wait Wait for 'offset' microseconds. Everything below 100 is discarded.
1399 read Read 'length' bytes beginning from 'offset'
1400 write Write 'length' bytes beginning from 'offset'
1401 sync fsync() the file
1402 datasync fdatasync() the file
1403 trim trim the given file from the given 'offset' for 'length' bytes