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 lets look at a really simple job file that define to threads, 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 Lets look at an example that have 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 ships with a few example job files, you can also look there for
165 5.0 Detailed list of parameters
166 -------------------------------
168 This section describes in details each parameter associated with a job.
169 Some parameters take an option of a given type, such as an integer or
170 a string. The following types are used:
172 str String. This is a sequence of alpha characters.
173 int Integer. A whole number value, may be negative.
174 siint SI integer. A whole number value, which may contain a postfix
175 describing the base of the number. Accepted postfixes are k/m/g,
176 meaning kilo, mega, and giga. So if you want to specify 4096,
177 you could either write out '4096' or just give 4k. The postfixes
178 signify base 2 values, so 1024 is 1k and 1024k is 1m and so on.
179 If the option accepts an upper and lower range, use a colon ':'
180 or minus '-' to seperate such values. See irange.
181 bool Boolean. Usually parsed as an integer, however only defined for
182 true and false (1 and 0).
183 irange Integer range with postfix. Allows value range to be given, such
184 as 1024-4096. A colon may also be used as the seperator, eg
185 1k:4k. If the option allows two sets of ranges, they can be
186 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
189 With the above in mind, here follows the complete list of fio job
192 name=str ASCII name of the job. This may be used to override the
193 name printed by fio for this job. Otherwise the job
194 name is used. On the command line this parameter has the
195 special purpose of also signaling the start of a new
198 description=str Text description of the job. Doesn't do anything except
199 dump this text description when this job is run. It's
202 directory=str Prefix filenames with this directory. Used to places files
203 in a different location than "./".
205 filename=str Fio normally makes up a filename based on the job name,
206 thread number, and file number. If you want to share
207 files between threads in a job or several jobs, specify
208 a filename for each of them to override the default. If
209 the ioengine used is 'net', the filename is the host and
210 port to connect to in the format of =host/port. If the
211 ioengine is file based, you can specify a number of files
212 by seperating the names with a ':' colon. So if you wanted
213 a job to open /dev/sda and /dev/sdb as the two working files,
214 you would use filename=/dev/sda:/dev/sdb
216 opendir=str Tell fio to recursively add any file it can find in this
217 directory and down the file system tree.
220 rw=str Type of io pattern. Accepted values are:
222 read Sequential reads
223 write Sequential writes
224 randwrite Random writes
225 randread Random reads
226 rw Sequential mixed reads and writes
227 randrw Random mixed reads and writes
229 For the mixed io types, the default is to split them 50/50.
230 For certain types of io the result may still be skewed a bit,
231 since the speed may be different. It is possible to specify
232 a number of IO's to do before getting a new offset - this
233 is only useful for random IO, where fio would normally
234 generate a new random offset for every IO. If you append
235 eg 8 to randread, you would get a new random offset for
236 every 8 IO's. The result would be a seek for only every 8
237 IO's, instead of for every IO. Use rw=randread:8 to specify
240 randrepeat=bool For random IO workloads, seed the generator in a predictable
241 way so that results are repeatable across repetitions.
243 size=siint The total size of file io for this job. This may describe
244 the size of the single file the job uses, or it may be
245 divided between the number of files in the job. If the
246 file already exists, the file size will be adjusted to this
247 size if larger than the current file size. If this parameter
248 is not given and the file exists, the file size will be used.
250 filesize=siint Individual file sizes. May be a range, in which case fio
251 will select sizes for files at random within the given range
252 and limited to 'size' in total (if that is given). If not
253 given, each created file is the same size.
256 bs=siint The block size used for the io units. Defaults to 4k. Values
257 can be given for both read and writes. If a single siint is
258 given, it will apply to both. If a second siint is specified
259 after a comma, it will apply to writes only. In other words,
260 the format is either bs=read_and_write or bs=read,write.
261 bs=4k,8k will thus use 4k blocks for reads, and 8k blocks
262 for writes. If you only wish to set the write size, you
263 can do so by passing an empty read size - bs=,8k will set
264 8k for writes and leave the read default value.
266 blocksize_range=irange
267 bsrange=irange Instead of giving a single block size, specify a range
268 and fio will mix the issued io block sizes. The issued
269 io unit will always be a multiple of the minimum value
270 given (also see bs_unaligned). Applies to both reads and
271 writes, however a second range can be given after a comma.
275 bs_unaligned If this option is given, any byte size value within bsrange
276 may be used as a block range. This typically wont work with
277 direct IO, as that normally requires sector alignment.
279 nrfiles=int Number of files to use for this job. Defaults to 1.
281 openfiles=int Number of files to keep open at the same time. Defaults to
282 the same as nrfiles, can be set smaller to limit the number
285 file_service_type=str Defines how fio decides which file from a job to
286 service next. The following types are defined:
288 random Just choose a file at random.
290 roundrobin Round robin over open files. This
293 The string can have a number appended, indicating how
294 often to switch to a new file. So if option random:4 is
295 given, fio will switch to a new random file after 4 ios
298 ioengine=str Defines how the job issues io to the file. The following
301 sync Basic read(2) or write(2) io. lseek(2) is
302 used to position the io location.
304 libaio Linux native asynchronous io.
306 posixaio glibc posix asynchronous io.
308 mmap File is memory mapped and data copied
309 to/from using memcpy(3).
311 splice splice(2) is used to transfer the data and
312 vmsplice(2) to transfer data from user
315 syslet-rw Use the syslet system calls to make
316 regular read/write async.
318 sg SCSI generic sg v3 io. May either be
319 synchronous using the SG_IO ioctl, or if
320 the target is an sg character device
321 we use read(2) and write(2) for asynchronous
324 null Doesn't transfer any data, just pretends
325 to. This is mainly used to exercise fio
326 itself and for debugging/testing purposes.
328 net Transfer over the network to given host:port.
329 'filename' must be set appropriately to
330 filename=host/port regardless of send
331 or receive, if the latter only the port
334 cpu Doesn't transfer any data, but burns CPU
335 cycles according to the cpuload= and
336 cpucycle= options. Setting cpuload=85
337 will cause that job to do nothing but burn
340 guasi The GUASI IO engine is the Generic Userspace
341 Asyncronous Syscall Interface approach
344 http://www.xmailserver.org/guasi-lib.html
346 for more info on GUASI.
348 external Prefix to specify loading an external
349 IO engine object file. Append the engine
350 filename, eg ioengine=external:/tmp/foo.o
351 to load ioengine foo.o in /tmp.
353 iodepth=int This defines how many io units to keep in flight against
354 the file. The default is 1 for each file defined in this
355 job, can be overridden with a larger value for higher
358 iodepth_batch=int This defines how many pieces of IO to submit at once.
359 It defaults to the same as iodepth, but can be set lower
362 iodepth_low=int The low water mark indicating when to start filling
363 the queue again. Defaults to the same as iodepth, meaning
364 that fio will attempt to keep the queue full at all times.
365 If iodepth is set to eg 16 and iodepth_low is set to 4, then
366 after fio has filled the queue of 16 requests, it will let
367 the depth drain down to 4 before starting to fill it again.
369 direct=bool If value is true, use non-buffered io. This is usually
372 buffered=bool If value is true, use buffered io. This is the opposite
373 of the 'direct' option. Defaults to true.
375 offset=siint Start io at the given offset in the file. The data before
376 the given offset will not be touched. This effectively
377 caps the file size at real_size - offset.
379 fsync=int If writing to a file, issue a sync of the dirty data
380 for every number of blocks given. For example, if you give
381 32 as a parameter, fio will sync the file for every 32
382 writes issued. If fio is using non-buffered io, we may
383 not sync the file. The exception is the sg io engine, which
384 synchronizes the disk cache anyway.
386 overwrite=bool If writing to a file, setup the file first and do overwrites.
388 end_fsync=bool If true, fsync file contents when the job exits.
390 fsync_on_close=bool If true, fio will fsync() a dirty file on close.
391 This differs from end_fsync in that it will happen on every
392 file close, not just at the end of the job.
394 rwmixcycle=int Value in milliseconds describing how often to switch between
395 reads and writes for a mixed workload. The default is
398 rwmixread=int How large a percentage of the mix should be reads.
400 rwmixwrite=int How large a percentage of the mix should be writes. If both
401 rwmixread and rwmixwrite is given and the values do not add
402 up to 100%, the latter of the two will be used to override
405 norandommap Normally fio will cover every block of the file when doing
406 random IO. If this option is given, fio will just get a
407 new random offset without looking at past io history. This
408 means that some blocks may not be read or written, and that
409 some blocks may be read/written more than once. This option
410 is mutually exclusive with verify= for that reason.
412 nice=int Run the job with the given nice value. See man nice(2).
414 prio=int Set the io priority value of this job. Linux limits us to
415 a positive value between 0 and 7, with 0 being the highest.
418 prioclass=int Set the io priority class. See man ionice(1).
420 thinktime=int Stall the job x microseconds after an io has completed before
421 issuing the next. May be used to simulate processing being
422 done by an application. See thinktime_blocks and
426 Only valid if thinktime is set - pretend to spend CPU time
427 doing something with the data received, before falling back
428 to sleeping for the rest of the period specified by
432 Only valid if thinktime is set - control how many blocks
433 to issue, before waiting 'thinktime' usecs. If not set,
434 defaults to 1 which will make fio wait 'thinktime' usecs
437 rate=int Cap the bandwidth used by this job to this number of KiB/sec.
439 ratemin=int Tell fio to do whatever it can to maintain at least this
440 bandwidth. Failing to meet this requirement, will cause
443 rate_iops=int Cap the bandwidth to this number of IOPS. Basically the same
444 as rate, just specified independently of bandwidth. If the
445 job is given a block size range instead of a fixed value,
446 the smallest block size is used as the metric.
448 rate_iops_min=int If fio doesn't meet this rate of IO, it will cause
451 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
454 cpumask=int Set the CPU affinity of this job. The parameter given is a
455 bitmask of allowed CPU's the job may run on. See man
456 sched_setaffinity(2).
458 startdelay=int Start this job the specified number of seconds after fio
459 has started. Only useful if the job file contains several
460 jobs, and you want to delay starting some jobs to a certain
463 runtime=int Tell fio to terminate processing after the specified number
464 of seconds. It can be quite hard to determine for how long
465 a specified job will run, so this parameter is handy to
466 cap the total runtime to a given time.
468 invalidate=bool Invalidate the buffer/page cache parts for this file prior
469 to starting io. Defaults to true.
471 sync=bool Use sync io for buffered writes. For the majority of the
472 io engines, this means using O_SYNC.
475 mem=str Fio can use various types of memory as the io unit buffer.
476 The allowed values are:
478 malloc Use memory from malloc(3) as the buffers.
480 shm Use shared memory as the buffers. Allocated
483 shmhuge Same as shm, but use huge pages as backing.
485 mmap Use mmap to allocate buffers. May either be
486 anonymous memory, or can be file backed if
487 a filename is given after the option. The
488 format is mem=mmap:/path/to/file.
490 mmaphuge Use a memory mapped huge file as the buffer
491 backing. Append filename after mmaphuge, ala
492 mem=mmaphuge:/hugetlbfs/file
494 The area allocated is a function of the maximum allowed
495 bs size for the job, multiplied by the io depth given. Note
496 that for shmhuge and mmaphuge to work, the system must have
497 free huge pages allocated. This can normally be checked
498 and set by reading/writing /proc/sys/vm/nr_hugepages on a
499 Linux system. Fio assumes a huge page is 4MiB in size. So
500 to calculate the number of huge pages you need for a given
501 job file, add up the io depth of all jobs (normally one unless
502 iodepth= is used) and multiply by the maximum bs set. Then
503 divide that number by the huge page size. You can see the
504 size of the huge pages in /proc/meminfo. If no huge pages
505 are allocated by having a non-zero number in nr_hugepages,
506 using mmaphuge or shmhuge will fail. Also see hugepage-size.
508 mmaphuge also needs to have hugetlbfs mounted and the file
509 location should point there. So if it's mounted in /huge,
510 you would use mem=mmaphuge:/huge/somefile.
513 Defines the size of a huge page. Must at least be equal
514 to the system setting, see /proc/meminfo. Defaults to 4MiB.
515 Should probably always be a multiple of megabytes, so using
516 hugepage-size=Xm is the preferred way to set this to avoid
517 setting a non-pow-2 bad value.
519 exitall When one job finishes, terminate the rest. The default is
520 to wait for each job to finish, sometimes that is not the
523 bwavgtime=int Average the calculated bandwidth over the given time. Value
524 is specified in milliseconds.
526 create_serialize=bool If true, serialize the file creating for the jobs.
527 This may be handy to avoid interleaving of data
528 files, which may greatly depend on the filesystem
529 used and even the number of processors in the system.
531 create_fsync=bool fsync the data file after creation. This is the
534 unlink=bool Unlink the job files when done. Not the default, as repeated
535 runs of that job would then waste time recreating the fileset
538 loops=int Run the specified number of iterations of this job. Used
539 to repeat the same workload a given number of times. Defaults
542 verify=str If writing to a file, fio can verify the file contents
543 after each iteration of the job. The allowed values are:
545 md5 Use an md5 sum of the data area and store
546 it in the header of each block.
548 crc32 Use a crc32 sum of the data area and store
549 it in the header of each block.
551 This option can be used for repeated burn-in tests of a
552 system to make sure that the written data is also
555 stonewall Wait for preceeding jobs in the job file to exit, before
556 starting this one. Can be used to insert serialization
557 points in the job file. A stone wall also implies starting
558 a new reporting group.
560 new_group Start a new reporting group. If this option isn't given,
561 jobs in a file will be part of the same reporting group
562 unless seperated by a stone wall (or if it's a group
563 by itself, with the numjobs option).
565 numjobs=int Create the specified number of clones of this job. May be
566 used to setup a larger number of threads/processes doing
567 the same thing. We regard that grouping of jobs as a
570 group_reporting If 'numjobs' is set, it may be interesting to display
571 statistics for the group as a whole instead of for each
572 individual job. This is especially true of 'numjobs' is
573 large, looking at individual thread/process output quickly
574 becomes unwieldy. If 'group_reporting' is specified, fio
575 will show the final report per-group instead of per-job.
577 thread fio defaults to forking jobs, however if this option is
578 given, fio will use pthread_create(3) to create threads
581 zonesize=siint Divide a file into zones of the specified size. See zoneskip.
583 zoneskip=siint Skip the specified number of bytes when zonesize data has
584 been read. The two zone options can be used to only do
585 io on zones of a file.
587 write_iolog=str Write the issued io patterns to the specified file. See
590 read_iolog=str Open an iolog with the specified file name and replay the
591 io patterns it contains. This can be used to store a
592 workload and replay it sometime later.
594 write_bw_log If given, write a bandwidth log of the jobs in this job
595 file. Can be used to store data of the bandwidth of the
596 jobs in their lifetime. The included fio_generate_plots
597 script uses gnuplot to turn these text files into nice
600 write_lat_log Same as write_bw_log, except that this option stores io
601 completion latencies instead.
603 lockmem=siint Pin down the specified amount of memory with mlock(2). Can
604 potentially be used instead of removing memory or booting
605 with less memory to simulate a smaller amount of memory.
607 exec_prerun=str Before running this job, issue the command specified
610 exec_postrun=str After the job completes, issue the command specified
613 ioscheduler=str Attempt to switch the device hosting the file to the specified
614 io scheduler before running.
616 cpuload=int If the job is a CPU cycle eater, attempt to use the specified
617 percentage of CPU cycles.
619 cpuchunks=int If the job is a CPU cycle eater, split the load into
620 cycles of the given time. In milliseconds.
623 6.0 Interpreting the output
624 ---------------------------
626 fio spits out a lot of output. While running, fio will display the
627 status of the jobs created. An example of that would be:
629 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
631 The characters inside the square brackets denote the current status of
632 each thread. The possible values (in typical life cycle order) are:
636 P Thread setup, but not started.
638 I Thread initialized, waiting.
639 R Running, doing sequential reads.
640 r Running, doing random reads.
641 W Running, doing sequential writes.
642 w Running, doing random writes.
643 M Running, doing mixed sequential reads/writes.
644 m Running, doing mixed random reads/writes.
645 F Running, currently waiting for fsync()
646 V Running, doing verification of written data.
647 E Thread exited, not reaped by main thread yet.
650 The other values are fairly self explanatory - number of threads
651 currently running and doing io, rate of io since last check, and the estimated
652 completion percentage and time for the running group. It's impossible to
653 estimate runtime of the following groups (if any).
655 When fio is done (or interrupted by ctrl-c), it will show the data for
656 each thread, group of threads, and disks in that order. For each data
657 direction, the output looks like:
659 Client1 (g=0): err= 0:
660 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
661 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
662 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
663 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
664 cpu : usr=1.49%, sys=0.25%, ctx=7969
665 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
666 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
667 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
669 The client number is printed, along with the group id and error of that
670 thread. Below is the io statistics, here for writes. In the order listed,
673 io= Number of megabytes io performed
674 bw= Average bandwidth rate
675 runt= The runtime of that thread
676 slat= Submission latency (avg being the average, stdev being the
677 standard deviation). This is the time it took to submit
678 the io. For sync io, the slat is really the completion
679 latency, since queue/complete is one operation there.
680 clat= Completion latency. Same names as slat, this denotes the
681 time from submission to completion of the io pieces. For
682 sync io, clat will usually be equal (or very close) to 0,
683 as the time from submit to complete is basically just
684 CPU time (io has already been done, see slat explanation).
685 bw= Bandwidth. Same names as the xlat stats, but also includes
686 an approximate percentage of total aggregate bandwidth
687 this thread received in this group. This last value is
688 only really useful if the threads in this group are on the
689 same disk, since they are then competing for disk access.
690 cpu= CPU usage. User and system time, along with the number
691 of context switches this thread went through.
692 IO depths= The distribution of io depths over the job life time. The
693 numbers are divided into powers of 2, so for example the
694 16= entries includes depths up to that value but higher
695 than the previous entry. In other words, it covers the
697 IO latencies= The distribution of IO completion latencies. This is the
698 time from when IO leaves fio and when it gets completed.
699 The numbers follow the same pattern as the IO depths,
700 meaning that 2=1.6% means that 1.6% of the IO completed
701 within 2 msecs, 20=12.8% means that 12.8% of the IO
702 took more than 10 msecs, but less than (or equal to) 20 msecs.
704 After each client has been listed, the group statistics are printed. They
707 Run status group 0 (all jobs):
708 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
709 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
711 For each data direction, it prints:
713 io= Number of megabytes io performed.
714 aggrb= Aggregate bandwidth of threads in this group.
715 minb= The minimum average bandwidth a thread saw.
716 maxb= The maximum average bandwidth a thread saw.
717 mint= The smallest runtime of the threads in that group.
718 maxt= The longest runtime of the threads in that group.
720 And finally, the disk statistics are printed. They will look like this:
722 Disk stats (read/write):
723 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
725 Each value is printed for both reads and writes, with reads first. The
728 ios= Number of ios performed by all groups.
729 merge= Number of merges io the io scheduler.
730 ticks= Number of ticks we kept the disk busy.
731 io_queue= Total time spent in the disk queue.
732 util= The disk utilization. A value of 100% means we kept the disk
733 busy constantly, 50% would be a disk idling half of the time.
739 For scripted usage where you typically want to generate tables or graphs
740 of the results, fio can output the results in a semicolon separated format.
741 The format is one long line of values, such as:
743 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%
744 ;0.0%;0.0%;0.0%;0.0%;0.0%
746 Split up, the format is as follows:
748 jobname, groupid, error
750 KiB IO, bandwidth (KiB/sec), runtime (msec)
751 Submission latency: min, max, mean, deviation
752 Completion latency: min, max, mean, deviation
753 Bw: min, max, aggregate percentage of total, mean, deviation
755 KiB IO, bandwidth (KiB/sec), runtime (msec)
756 Submission latency: min, max, mean, deviation
757 Completion latency: min, max, mean, deviation
758 Bw: min, max, aggregate percentage of total, mean, deviation
759 CPU usage: user, system, context switches
760 IO depths: <=1, 2, 4, 8, 16, 32, >=64
761 IO latencies: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000