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 128MiB 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 32KiB and define numjobs to 4 to
154 fork 4 identical jobs. The result is 4 processes each randomly writing
155 to their own 64MiB 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 fio also supports environment variable expansion in job files. Any
162 substring of the form "${VARNAME}" as part of an option value (in other
163 words, on the right of the `='), will be expanded to the value of the
164 environment variable called VARNAME. If no such environment variable
165 is defined, or VARNAME is the empty string, the empty string will be
168 As an example, let's look at a sample fio invocation and job file:
170 $ SIZE=64m NUMJOBS=4 fio jobfile.fio
172 ; -- start job file --
179 This will expand to the following equivalent job file at runtime:
181 ; -- start job file --
188 fio ships with a few example job files, you can also look there for
192 5.0 Detailed list of parameters
193 -------------------------------
195 This section describes in details each parameter associated with a job.
196 Some parameters take an option of a given type, such as an integer or
197 a string. The following types are used:
199 str String. This is a sequence of alpha characters.
200 int Integer. A whole number value, can be negative. If prefixed with
201 0x, the integer is assumed to be of base 16 (hexadecimal).
202 time Integer with possible time postfix. In seconds unless otherwise
203 specified, use eg 10m for 10 minutes. Accepts s/m/h for seconds,
205 siint SI integer. A whole number value, which may contain a postfix
206 describing the base of the number. Accepted postfixes are k/m/g,
207 meaning kilo, mega, and giga. So if you want to specify 4096,
208 you could either write out '4096' or just give 4k. The postfixes
209 signify base 2 values, so 1024 is 1k and 1024k is 1m and so on.
210 If the option accepts an upper and lower range, use a colon ':'
211 or minus '-' to separate such values. See irange.
212 bool Boolean. Usually parsed as an integer, however only defined for
213 true and false (1 and 0).
214 irange Integer range with postfix. Allows value range to be given, such
215 as 1024-4096. A colon may also be used as the separator, eg
216 1k:4k. If the option allows two sets of ranges, they can be
217 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
220 With the above in mind, here follows the complete list of fio job
223 name=str ASCII name of the job. This may be used to override the
224 name printed by fio for this job. Otherwise the job
225 name is used. On the command line this parameter has the
226 special purpose of also signaling the start of a new
229 description=str Text description of the job. Doesn't do anything except
230 dump this text description when this job is run. It's
233 directory=str Prefix filenames with this directory. Used to places files
234 in a different location than "./".
236 filename=str Fio normally makes up a filename based on the job name,
237 thread number, and file number. If you want to share
238 files between threads in a job or several jobs, specify
239 a filename for each of them to override the default. If
240 the ioengine used is 'net', the filename is the host, port,
241 and protocol to use in the format of =host/port/protocol.
242 See ioengine=net for more. If the ioengine is file based, you
243 can specify a number of files by separating the names with a
244 ':' colon. So if you wanted a job to open /dev/sda and /dev/sdb
245 as the two working files, you would use
246 filename=/dev/sda:/dev/sdb. '-' is a reserved name, meaning
247 stdin or stdout. Which of the two depends on the read/write
250 opendir=str Tell fio to recursively add any file it can find in this
251 directory and down the file system tree.
253 lockfile=str Fio defaults to not doing any locking files before it does
254 IO to them. If a file or file descriptor is shared, fio
255 can serialize IO to that file to make the end result
256 consistent. This is usual for emulating real workloads that
257 share files. The lock modes are:
259 none No locking. The default.
260 exclusive Only one thread/process may do IO,
261 excluding all others.
262 readwrite Read-write locking on the file. Many
263 readers may access the file at the
264 same time, but writes get exclusive
267 The option may be post-fixed with a lock batch number. If
268 set, then each thread/process may do that amount of IOs to
269 the file before giving up the lock. Since lock acquisition is
270 expensive, batching the lock/unlocks will speed up IO.
273 rw=str Type of io pattern. Accepted values are:
275 read Sequential reads
276 write Sequential writes
277 randwrite Random writes
278 randread Random reads
279 rw Sequential mixed reads and writes
280 randrw Random mixed reads and writes
282 For the mixed io types, the default is to split them 50/50.
283 For certain types of io the result may still be skewed a bit,
284 since the speed may be different. It is possible to specify
285 a number of IO's to do before getting a new offset - this
286 is only useful for random IO, where fio would normally
287 generate a new random offset for every IO. If you append
288 eg 8 to randread, you would get a new random offset for
289 every 8 IO's. The result would be a seek for only every 8
290 IO's, instead of for every IO. Use rw=randread:8 to specify
293 randrepeat=bool For random IO workloads, seed the generator in a predictable
294 way so that results are repeatable across repetitions.
296 fadvise_hint=bool By default, fio will use fadvise() to advise the kernel
297 on what IO patterns it is likely to issue. Sometimes you
298 want to test specific IO patterns without telling the
299 kernel about it, in which case you can disable this option.
300 If set, fio will use POSIX_FADV_SEQUENTIAL for sequential
301 IO and POSIX_FADV_RANDOM for random IO.
303 size=siint The total size of file io for this job. Fio will run until
304 this many bytes has been transferred, unless runtime is
305 limited by other options (such as 'runtime', for instance).
306 Unless specific nr_files and filesize options are given,
307 fio will divide this size between the available files
308 specified by the job.
310 filesize=siint Individual file sizes. May be a range, in which case fio
311 will select sizes for files at random within the given range
312 and limited to 'size' in total (if that is given). If not
313 given, each created file is the same size.
315 fill_device=bool Sets size to something really large and waits for ENOSPC (no
316 space left on device) as the terminating condition. Only makes
317 sense with sequential write.
320 bs=siint The block size used for the io units. Defaults to 4k. Values
321 can be given for both read and writes. If a single siint is
322 given, it will apply to both. If a second siint is specified
323 after a comma, it will apply to writes only. In other words,
324 the format is either bs=read_and_write or bs=read,write.
325 bs=4k,8k will thus use 4k blocks for reads, and 8k blocks
326 for writes. If you only wish to set the write size, you
327 can do so by passing an empty read size - bs=,8k will set
328 8k for writes and leave the read default value.
330 blocksize_range=irange
331 bsrange=irange Instead of giving a single block size, specify a range
332 and fio will mix the issued io block sizes. The issued
333 io unit will always be a multiple of the minimum value
334 given (also see bs_unaligned). Applies to both reads and
335 writes, however a second range can be given after a comma.
338 bssplit=str Sometimes you want even finer grained control of the
339 block sizes issued, not just an even split between them.
340 This option allows you to weight various block sizes,
341 so that you are able to define a specific amount of
342 block sizes issued. The format for this option is:
344 bssplit=blocksize/percentage:blocksize/percentage
346 for as many block sizes as needed. So if you want to define
347 a workload that has 50% 64k blocks, 10% 4k blocks, and
348 40% 32k blocks, you would write:
350 bssplit=4k/10:64k/50:32k/40
352 Ordering does not matter. If the percentage is left blank,
353 fio will fill in the remaining values evenly. So a bssplit
354 option like this one:
356 bssplit=4k/50:1k/:32k/
358 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages
359 always add up to 100, if bssplit is given a range that adds
360 up to more, it will error out.
363 bs_unaligned If this option is given, any byte size value within bsrange
364 may be used as a block range. This typically wont work with
365 direct IO, as that normally requires sector alignment.
367 zero_buffers If this option is given, fio will init the IO buffers to
368 all zeroes. The default is to fill them with random data.
370 refill_buffers If this option is given, fio will refill the IO buffers
371 on every submit. The default is to only fill it at init
372 time and reuse that data. Only makes sense if zero_buffers
373 isn't specified, naturally. If data verification is enabled,
374 refill_buffers is also automatically enabled.
376 nrfiles=int Number of files to use for this job. Defaults to 1.
378 openfiles=int Number of files to keep open at the same time. Defaults to
379 the same as nrfiles, can be set smaller to limit the number
382 file_service_type=str Defines how fio decides which file from a job to
383 service next. The following types are defined:
385 random Just choose a file at random.
387 roundrobin Round robin over open files. This
390 The string can have a number appended, indicating how
391 often to switch to a new file. So if option random:4 is
392 given, fio will switch to a new random file after 4 ios
395 ioengine=str Defines how the job issues io to the file. The following
398 sync Basic read(2) or write(2) io. lseek(2) is
399 used to position the io location.
401 psync Basic pread(2) or pwrite(2) io.
403 vsync Basic readv(2) or writev(2) IO.
405 libaio Linux native asynchronous io.
407 posixaio glibc posix asynchronous io.
409 solarisaio Solaris native asynchronous io.
411 mmap File is memory mapped and data copied
412 to/from using memcpy(3).
414 splice splice(2) is used to transfer the data and
415 vmsplice(2) to transfer data from user
418 syslet-rw Use the syslet system calls to make
419 regular read/write async.
421 sg SCSI generic sg v3 io. May either be
422 synchronous using the SG_IO ioctl, or if
423 the target is an sg character device
424 we use read(2) and write(2) for asynchronous
427 null Doesn't transfer any data, just pretends
428 to. This is mainly used to exercise fio
429 itself and for debugging/testing purposes.
431 net Transfer over the network to given host:port.
432 'filename' must be set appropriately to
433 filename=host/port/protocol regardless of send
434 or receive, if the latter only the port
435 argument is used. 'host' may be an IP address
436 or hostname, port is the port number to be used,
437 and protocol may be 'udp' or 'tcp'. If no
438 protocol is given, TCP is used.
440 netsplice Like net, but uses splice/vmsplice to
441 map data and send/receive.
443 cpuio Doesn't transfer any data, but burns CPU
444 cycles according to the cpuload= and
445 cpucycle= options. Setting cpuload=85
446 will cause that job to do nothing but burn
447 85% of the CPU. In case of SMP machines,
448 use numjobs=<no_of_cpu> to get desired CPU
449 usage, as the cpuload only loads a single
450 CPU at the desired rate.
452 guasi The GUASI IO engine is the Generic Userspace
453 Asyncronous Syscall Interface approach
456 http://www.xmailserver.org/guasi-lib.html
458 for more info on GUASI.
460 external Prefix to specify loading an external
461 IO engine object file. Append the engine
462 filename, eg ioengine=external:/tmp/foo.o
463 to load ioengine foo.o in /tmp.
465 iodepth=int This defines how many io units to keep in flight against
466 the file. The default is 1 for each file defined in this
467 job, can be overridden with a larger value for higher
470 iodepth_batch_submit=int
471 iodepth_batch=int This defines how many pieces of IO to submit at once.
472 It defaults to 1 which means that we submit each IO
473 as soon as it is available, but can be raised to submit
474 bigger batches of IO at the time.
476 iodepth_batch_complete=int This defines how many pieces of IO to retrieve
477 at once. It defaults to 1 which means that we'll ask
478 for a minimum of 1 IO in the retrieval process from
479 the kernel. The IO retrieval will go on until we
480 hit the limit set by iodepth_low. If this variable is
481 set to 0, then fio will always check for completed
482 events before queuing more IO. This helps reduce
483 IO latency, at the cost of more retrieval system calls.
485 iodepth_low=int The low water mark indicating when to start filling
486 the queue again. Defaults to the same as iodepth, meaning
487 that fio will attempt to keep the queue full at all times.
488 If iodepth is set to eg 16 and iodepth_low is set to 4, then
489 after fio has filled the queue of 16 requests, it will let
490 the depth drain down to 4 before starting to fill it again.
492 direct=bool If value is true, use non-buffered io. This is usually
495 buffered=bool If value is true, use buffered io. This is the opposite
496 of the 'direct' option. Defaults to true.
498 offset=siint Start io at the given offset in the file. The data before
499 the given offset will not be touched. This effectively
500 caps the file size at real_size - offset.
502 fsync=int If writing to a file, issue a sync of the dirty data
503 for every number of blocks given. For example, if you give
504 32 as a parameter, fio will sync the file for every 32
505 writes issued. If fio is using non-buffered io, we may
506 not sync the file. The exception is the sg io engine, which
507 synchronizes the disk cache anyway.
509 overwrite=bool If true, writes to a file will always overwrite existing
510 data. If the file doesn't already exist, it will be
511 created before the write phase begins. If the file exists
512 and is large enough for the specified write phase, nothing
515 end_fsync=bool If true, fsync file contents when the job exits.
517 fsync_on_close=bool If true, fio will fsync() a dirty file on close.
518 This differs from end_fsync in that it will happen on every
519 file close, not just at the end of the job.
521 rwmixread=int How large a percentage of the mix should be reads.
523 rwmixwrite=int How large a percentage of the mix should be writes. If both
524 rwmixread and rwmixwrite is given and the values do not add
525 up to 100%, the latter of the two will be used to override
528 norandommap Normally fio will cover every block of the file when doing
529 random IO. If this option is given, fio will just get a
530 new random offset without looking at past io history. This
531 means that some blocks may not be read or written, and that
532 some blocks may be read/written more than once. This option
533 is mutually exclusive with verify= for that reason, since
534 fio doesn't track potential block rewrites which may alter
535 the calculated checksum for that block.
537 softrandommap See norandommap. If fio runs with the random block map enabled
538 and it fails to allocate the map, if this option is set it
539 will continue without a random block map. As coverage will
540 not be as complete as with random maps, this option is
543 nice=int Run the job with the given nice value. See man nice(2).
545 prio=int Set the io priority value of this job. Linux limits us to
546 a positive value between 0 and 7, with 0 being the highest.
549 prioclass=int Set the io priority class. See man ionice(1).
551 thinktime=int Stall the job x microseconds after an io has completed before
552 issuing the next. May be used to simulate processing being
553 done by an application. See thinktime_blocks and
557 Only valid if thinktime is set - pretend to spend CPU time
558 doing something with the data received, before falling back
559 to sleeping for the rest of the period specified by
563 Only valid if thinktime is set - control how many blocks
564 to issue, before waiting 'thinktime' usecs. If not set,
565 defaults to 1 which will make fio wait 'thinktime' usecs
568 rate=int Cap the bandwidth used by this job to this number of KiB/sec.
570 ratemin=int Tell fio to do whatever it can to maintain at least this
571 bandwidth. Failing to meet this requirement, will cause
574 rate_iops=int Cap the bandwidth to this number of IOPS. Basically the same
575 as rate, just specified independently of bandwidth. If the
576 job is given a block size range instead of a fixed value,
577 the smallest block size is used as the metric.
579 rate_iops_min=int If fio doesn't meet this rate of IO, it will cause
582 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
585 cpumask=int Set the CPU affinity of this job. The parameter given is a
586 bitmask of allowed CPU's the job may run on. So if you want
587 the allowed CPUs to be 1 and 5, you would pass the decimal
588 value of (1 << 1 | 1 << 5), or 34. See man
589 sched_setaffinity(2). This may not work on all supported
590 operating systems or kernel versions. This option doesn't
591 work well for a higher CPU count than what you can store in
592 an integer mask, so it can only control cpus 1-32. For
593 boxes with larger CPU counts, use cpus_allowed.
595 cpus_allowed=str Controls the same options as cpumask, but it allows a text
596 setting of the permitted CPUs instead. So to use CPUs 1 and
597 5, you would specify cpus_allowed=1,5. This options also
598 allows a range of CPUs. Say you wanted a binding to CPUs
599 1, 5, and 8-15, you would set cpus_allowed=1,5,8-15.
601 startdelay=time Start this job the specified number of seconds after fio
602 has started. Only useful if the job file contains several
603 jobs, and you want to delay starting some jobs to a certain
606 runtime=time Tell fio to terminate processing after the specified number
607 of seconds. It can be quite hard to determine for how long
608 a specified job will run, so this parameter is handy to
609 cap the total runtime to a given time.
611 time_based If set, fio will run for the duration of the runtime
612 specified even if the file(s) are completely read or
613 written. It will simply loop over the same workload
614 as many times as the runtime allows.
616 ramp_time=time If set, fio will run the specified workload for this amount
617 of time before logging any performance numbers. Useful for
618 letting performance settle before logging results, thus
619 minimizing the runtime required for stable results. Note
620 that the ramp_time is considered lead in time for a job,
621 thus it will increase the total runtime if a special timeout
622 or runtime is specified.
624 invalidate=bool Invalidate the buffer/page cache parts for this file prior
625 to starting io. Defaults to true.
627 sync=bool Use sync io for buffered writes. For the majority of the
628 io engines, this means using O_SYNC.
631 mem=str Fio can use various types of memory as the io unit buffer.
632 The allowed values are:
634 malloc Use memory from malloc(3) as the buffers.
636 shm Use shared memory as the buffers. Allocated
639 shmhuge Same as shm, but use huge pages as backing.
641 mmap Use mmap to allocate buffers. May either be
642 anonymous memory, or can be file backed if
643 a filename is given after the option. The
644 format is mem=mmap:/path/to/file.
646 mmaphuge Use a memory mapped huge file as the buffer
647 backing. Append filename after mmaphuge, ala
648 mem=mmaphuge:/hugetlbfs/file
650 The area allocated is a function of the maximum allowed
651 bs size for the job, multiplied by the io depth given. Note
652 that for shmhuge and mmaphuge to work, the system must have
653 free huge pages allocated. This can normally be checked
654 and set by reading/writing /proc/sys/vm/nr_hugepages on a
655 Linux system. Fio assumes a huge page is 4MiB in size. So
656 to calculate the number of huge pages you need for a given
657 job file, add up the io depth of all jobs (normally one unless
658 iodepth= is used) and multiply by the maximum bs set. Then
659 divide that number by the huge page size. You can see the
660 size of the huge pages in /proc/meminfo. If no huge pages
661 are allocated by having a non-zero number in nr_hugepages,
662 using mmaphuge or shmhuge will fail. Also see hugepage-size.
664 mmaphuge also needs to have hugetlbfs mounted and the file
665 location should point there. So if it's mounted in /huge,
666 you would use mem=mmaphuge:/huge/somefile.
669 Defines the size of a huge page. Must at least be equal
670 to the system setting, see /proc/meminfo. Defaults to 4MiB.
671 Should probably always be a multiple of megabytes, so using
672 hugepage-size=Xm is the preferred way to set this to avoid
673 setting a non-pow-2 bad value.
675 exitall When one job finishes, terminate the rest. The default is
676 to wait for each job to finish, sometimes that is not the
679 bwavgtime=int Average the calculated bandwidth over the given time. Value
680 is specified in milliseconds.
682 create_serialize=bool If true, serialize the file creating for the jobs.
683 This may be handy to avoid interleaving of data
684 files, which may greatly depend on the filesystem
685 used and even the number of processors in the system.
687 create_fsync=bool fsync the data file after creation. This is the
690 unlink=bool Unlink the job files when done. Not the default, as repeated
691 runs of that job would then waste time recreating the file
694 loops=int Run the specified number of iterations of this job. Used
695 to repeat the same workload a given number of times. Defaults
698 do_verify=bool Run the verify phase after a write phase. Only makes sense if
699 verify is set. Defaults to 1.
701 verify=str If writing to a file, fio can verify the file contents
702 after each iteration of the job. The allowed values are:
704 md5 Use an md5 sum of the data area and store
705 it in the header of each block.
707 crc64 Use an experimental crc64 sum of the data
708 area and store it in the header of each
711 crc32c Use a crc32c sum of the data area and store
712 it in the header of each block.
714 crc32c-intel Use hardware assisted crc32c calcuation
715 provided on SSE4.2 enabled processors.
717 crc32 Use a crc32 sum of the data area and store
718 it in the header of each block.
720 crc16 Use a crc16 sum of the data area and store
721 it in the header of each block.
723 crc7 Use a crc7 sum of the data area and store
724 it in the header of each block.
726 sha512 Use sha512 as the checksum function.
728 sha256 Use sha256 as the checksum function.
730 meta Write extra information about each io
731 (timestamp, block number etc.). The block
734 null Only pretend to verify. Useful for testing
735 internals with ioengine=null, not for much
738 This option can be used for repeated burn-in tests of a
739 system to make sure that the written data is also
742 verifysort=bool If set, fio will sort written verify blocks when it deems
743 it faster to read them back in a sorted manner. This is
744 often the case when overwriting an existing file, since
745 the blocks are already laid out in the file system. You
746 can ignore this option unless doing huge amounts of really
747 fast IO where the red-black tree sorting CPU time becomes
750 verify_offset=siint Swap the verification header with data somewhere else
751 in the block before writing. Its swapped back before
754 verify_interval=siint Write the verification header at a finer granularity
755 than the blocksize. It will be written for chunks the
756 size of header_interval. blocksize should divide this
759 verify_pattern=int If set, fio will fill the io buffers with this
760 pattern. Fio defaults to filling with totally random
761 bytes, but sometimes it's interesting to fill with a known
762 pattern for io verification purposes. Depending on the
763 width of the pattern, fio will fill 1/2/3/4 bytes of the
764 buffer at the time. The verify_pattern cannot be larger than
767 verify_fatal=bool Normally fio will keep checking the entire contents
768 before quitting on a block verification failure. If this
769 option is set, fio will exit the job on the first observed
772 stonewall Wait for preceeding jobs in the job file to exit, before
773 starting this one. Can be used to insert serialization
774 points in the job file. A stone wall also implies starting
775 a new reporting group.
777 new_group Start a new reporting group. If this option isn't given,
778 jobs in a file will be part of the same reporting group
779 unless separated by a stone wall (or if it's a group
780 by itself, with the numjobs option).
782 numjobs=int Create the specified number of clones of this job. May be
783 used to setup a larger number of threads/processes doing
784 the same thing. We regard that grouping of jobs as a
787 group_reporting If 'numjobs' is set, it may be interesting to display
788 statistics for the group as a whole instead of for each
789 individual job. This is especially true of 'numjobs' is
790 large, looking at individual thread/process output quickly
791 becomes unwieldy. If 'group_reporting' is specified, fio
792 will show the final report per-group instead of per-job.
794 thread fio defaults to forking jobs, however if this option is
795 given, fio will use pthread_create(3) to create threads
798 zonesize=siint Divide a file into zones of the specified size. See zoneskip.
800 zoneskip=siint Skip the specified number of bytes when zonesize data has
801 been read. The two zone options can be used to only do
802 io on zones of a file.
804 write_iolog=str Write the issued io patterns to the specified file. See
807 read_iolog=str Open an iolog with the specified file name and replay the
808 io patterns it contains. This can be used to store a
809 workload and replay it sometime later. The iolog given
810 may also be a blktrace binary file, which allows fio
811 to replay a workload captured by blktrace. See blktrace
812 for how to capture such logging data. For blktrace replay,
813 the file needs to be turned into a blkparse binary data
814 file first (blktrace <device> -d file_for_fio.bin).
816 write_bw_log=str If given, write a bandwidth log of the jobs in this job
817 file. Can be used to store data of the bandwidth of the
818 jobs in their lifetime. The included fio_generate_plots
819 script uses gnuplot to turn these text files into nice
820 graphs. See write_log_log for behaviour of given
821 filename. For this option, the postfix is _bw.log.
823 write_lat_log=str Same as write_bw_log, except that this option stores io
824 completion latencies instead. If no filename is given
825 with this option, the default filename of "jobname_type.log"
826 is used. Even if the filename is given, fio will still
827 append the type of log. So if one specifies
831 The actual log names will be foo_clat.log and foo_slat.log.
832 This helps fio_generate_plot fine the logs automatically.
834 lockmem=siint Pin down the specified amount of memory with mlock(2). Can
835 potentially be used instead of removing memory or booting
836 with less memory to simulate a smaller amount of memory.
838 exec_prerun=str Before running this job, issue the command specified
841 exec_postrun=str After the job completes, issue the command specified
844 ioscheduler=str Attempt to switch the device hosting the file to the specified
845 io scheduler before running.
847 cpuload=int If the job is a CPU cycle eater, attempt to use the specified
848 percentage of CPU cycles.
850 cpuchunks=int If the job is a CPU cycle eater, split the load into
851 cycles of the given time. In milliseconds.
853 disk_util=bool Generate disk utilization statistics, if the platform
854 supports it. Defaults to on.
856 disable_clat=bool Disable measurements of completion latency numbers. Useful
857 only for cutting back the number of calls to gettimeofday,
858 as that does impact performance at really high IOPS rates.
859 Note that to really get rid of a large amount of these
860 calls, this option must be used with disable_slat and
863 disable_slat=bool Disable measurements of submission latency numbers. See
866 disable_bw=bool Disable measurements of throughput/bandwidth numbers. See
869 gtod_reduce=bool Enable all of the gettimeofday() reducing options
870 (disable_clat, disable_slat, disable_bw) plus reduce
871 precision of the timeout somewhat to really shrink
872 the gettimeofday() call count. With this option enabled,
873 we only do about 0.4% of the gtod() calls we would have
874 done if all time keeping was enabled.
876 gtod_cpu=int Sometimes it's cheaper to dedicate a single thread of
877 execution to just getting the current time. Fio (and
878 databases, for instance) are very intensive on gettimeofday()
879 calls. With this option, you can set one CPU aside for
880 doing nothing but logging current time to a shared memory
881 location. Then the other threads/processes that run IO
882 workloads need only copy that segment, instead of entering
883 the kernel with a gettimeofday() call. The CPU set aside
884 for doing these time calls will be excluded from other
885 uses. Fio will manually clear it from the CPU mask of other
889 6.0 Interpreting the output
890 ---------------------------
892 fio spits out a lot of output. While running, fio will display the
893 status of the jobs created. An example of that would be:
895 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
897 The characters inside the square brackets denote the current status of
898 each thread. The possible values (in typical life cycle order) are:
902 P Thread setup, but not started.
904 I Thread initialized, waiting.
905 R Running, doing sequential reads.
906 r Running, doing random reads.
907 W Running, doing sequential writes.
908 w Running, doing random writes.
909 M Running, doing mixed sequential reads/writes.
910 m Running, doing mixed random reads/writes.
911 F Running, currently waiting for fsync()
912 V Running, doing verification of written data.
913 E Thread exited, not reaped by main thread yet.
916 The other values are fairly self explanatory - number of threads
917 currently running and doing io, rate of io since last check (read speed
918 listed first, then write speed), and the estimated completion percentage
919 and time for the running group. It's impossible to estimate runtime of
920 the following groups (if any).
922 When fio is done (or interrupted by ctrl-c), it will show the data for
923 each thread, group of threads, and disks in that order. For each data
924 direction, the output looks like:
926 Client1 (g=0): err= 0:
927 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
928 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
929 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
930 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
931 cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17
932 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
933 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
934 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
935 issued r/w: total=0/32768, short=0/0
936 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
937 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
939 The client number is printed, along with the group id and error of that
940 thread. Below is the io statistics, here for writes. In the order listed,
943 io= Number of megabytes io performed
944 bw= Average bandwidth rate
945 runt= The runtime of that thread
946 slat= Submission latency (avg being the average, stdev being the
947 standard deviation). This is the time it took to submit
948 the io. For sync io, the slat is really the completion
949 latency, since queue/complete is one operation there. This
950 value can be in milliseconds or microseconds, fio will choose
951 the most appropriate base and print that. In the example
952 above, milliseconds is the best scale.
953 clat= Completion latency. Same names as slat, this denotes the
954 time from submission to completion of the io pieces. For
955 sync io, clat will usually be equal (or very close) to 0,
956 as the time from submit to complete is basically just
957 CPU time (io has already been done, see slat explanation).
958 bw= Bandwidth. Same names as the xlat stats, but also includes
959 an approximate percentage of total aggregate bandwidth
960 this thread received in this group. This last value is
961 only really useful if the threads in this group are on the
962 same disk, since they are then competing for disk access.
963 cpu= CPU usage. User and system time, along with the number
964 of context switches this thread went through, usage of
965 system and user time, and finally the number of major
966 and minor page faults.
967 IO depths= The distribution of io depths over the job life time. The
968 numbers are divided into powers of 2, so for example the
969 16= entries includes depths up to that value but higher
970 than the previous entry. In other words, it covers the
972 IO submit= How many pieces of IO were submitting in a single submit
973 call. Each entry denotes that amount and below, until
974 the previous entry - eg, 8=100% mean that we submitted
975 anywhere in between 5-8 ios per submit call.
976 IO complete= Like the above submit number, but for completions instead.
977 IO issued= The number of read/write requests issued, and how many
979 IO latencies= The distribution of IO completion latencies. This is the
980 time from when IO leaves fio and when it gets completed.
981 The numbers follow the same pattern as the IO depths,
982 meaning that 2=1.6% means that 1.6% of the IO completed
983 within 2 msecs, 20=12.8% means that 12.8% of the IO
984 took more than 10 msecs, but less than (or equal to) 20 msecs.
986 After each client has been listed, the group statistics are printed. They
989 Run status group 0 (all jobs):
990 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
991 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
993 For each data direction, it prints:
995 io= Number of megabytes io performed.
996 aggrb= Aggregate bandwidth of threads in this group.
997 minb= The minimum average bandwidth a thread saw.
998 maxb= The maximum average bandwidth a thread saw.
999 mint= The smallest runtime of the threads in that group.
1000 maxt= The longest runtime of the threads in that group.
1002 And finally, the disk statistics are printed. They will look like this:
1004 Disk stats (read/write):
1005 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
1007 Each value is printed for both reads and writes, with reads first. The
1010 ios= Number of ios performed by all groups.
1011 merge= Number of merges io the io scheduler.
1012 ticks= Number of ticks we kept the disk busy.
1013 io_queue= Total time spent in the disk queue.
1014 util= The disk utilization. A value of 100% means we kept the disk
1015 busy constantly, 50% would be a disk idling half of the time.
1021 For scripted usage where you typically want to generate tables or graphs
1022 of the results, fio can output the results in a semicolon separated format.
1023 The format is one long line of values, such as:
1025 client1;0;0;1906777;1090804;1790;0;0;0.000000;0.000000;0;0;0.000000;0.000000;929380;1152890;25.510151%;1078276.333333;128948.113404;0;0;0;0;0;0.000000;0.000000;0;0;0.000000;0.000000;0;0;0.000000%;0.000000;0.000000;100.000000%;0.000000%;324;100.0%;0.0%;0.0%;0.0%;0.0%;0.0%;0.0%;100.0%;0.0%;0.0%;0.0%;0.0%;0.0%
1026 ;0.0%;0.0%;0.0%;0.0%;0.0%
1028 To enable terse output, use the --minimal command line option.
1030 Split up, the format is as follows:
1032 jobname, groupid, error
1034 KiB IO, bandwidth (KiB/sec), runtime (msec)
1035 Submission latency: min, max, mean, deviation
1036 Completion latency: min, max, mean, deviation
1037 Bw: min, max, aggregate percentage of total, mean, deviation
1039 KiB IO, bandwidth (KiB/sec), runtime (msec)
1040 Submission latency: min, max, mean, deviation
1041 Completion latency: min, max, mean, deviation
1042 Bw: min, max, aggregate percentage of total, mean, deviation
1043 CPU usage: user, system, context switches, major faults, minor faults
1044 IO depths: <=1, 2, 4, 8, 16, 32, >=64
1045 IO latencies: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000