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 bool Boolean. Usually parsed as an integer, however only defined for
180 true and false (1 and 0).
181 irange Integer range with postfix. Allows value range to be given, such
182 as 1024-4096. A colon may also be used as the seperator, eg
183 1k:4k. If the option allows two sets of ranges, they can be
184 specified with a ',' or '/' delimiter: 1k-4k/8k-32k. Also see
187 With the above in mind, here follows the complete list of fio job
190 name=str ASCII name of the job. This may be used to override the
191 name printed by fio for this job. Otherwise the job
192 name is used. On the command line this parameter has the
193 special purpose of also signaling the start of a new
196 description=str Text description of the job. Doesn't do anything except
197 dump this text description when this job is run. It's
200 directory=str Prefix filenames with this directory. Used to places files
201 in a different location than "./".
203 filename=str Fio normally makes up a filename based on the job name,
204 thread number, and file number. If you want to share
205 files between threads in a job or several jobs, specify
206 a filename for each of them to override the default. If
207 the ioengine used is 'net', the filename is the host and
208 port to connect to in the format of =host:port.
210 rw=str Type of io pattern. Accepted values are:
212 read Sequential reads
213 write Sequential writes
214 randwrite Random writes
215 randread Random reads
216 rw Sequential mixed reads and writes
217 randrw Random mixed reads and writes
219 For the mixed io types, the default is to split them 50/50.
220 For certain types of io the result may still be skewed a bit,
221 since the speed may be different.
223 randrepeat=bool For random IO workloads, seed the generator in a predictable
224 way so that results are repeatable across repetitions.
226 size=siint The total size of file io for this job. This may describe
227 the size of the single file the job uses, or it may be
228 divided between the number of files in the job. If the
229 file already exists, the file size will be adjusted to this
230 size if larger than the current file size. If this parameter
231 is not given and the file exists, the file size will be used.
233 bs=siint The block size used for the io units. Defaults to 4k. Values
234 can be given for both read and writes. If a single siint is
235 given, it will apply to both. If a second siint is specified
236 after a comma, it will apply to writes only. In other words,
237 the format is either bs=read_and_write or bs=read,write.
238 bs=4k,8k will thus use 4k blocks for reads, and 8k blocks
239 for writes. If you only wish to set the write size, you
240 can do so by passing an empty read size - bs=,8k will set
241 8k for writes and leave the read default value.
243 bsrange=irange Instead of giving a single block size, specify a range
244 and fio will mix the issued io block sizes. The issued
245 io unit will always be a multiple of the minimum value
246 given (also see bs_unaligned). Applies to both reads and
247 writes, however a second range can be given after a comma.
250 bs_unaligned If this option is given, any byte size value within bsrange
251 may be used as a block range. This typically wont work with
252 direct IO, as that normally requires sector alignment.
254 nrfiles=int Number of files to use for this job. Defaults to 1.
256 openfiles=int Number of files to keep open at the same time. Defaults to
257 the same as nrfiles, can be set smaller to limit the number
260 file_service_type=str Defines how fio decides which file from a job to
261 service next. The following types are defined:
263 random Just choose a file at random.
265 roundrobin Round robin over open files. This
268 ioengine=str Defines how the job issues io to the file. The following
271 sync Basic read(2) or write(2) io. lseek(2) is
272 used to position the io location.
274 libaio Linux native asynchronous io.
276 posixaio glibc posix asynchronous io.
278 mmap File is memory mapped and data copied
279 to/from using memcpy(3).
281 splice splice(2) is used to transfer the data and
282 vmsplice(2) to transfer data from user
285 syslet-rw Use the syslet system calls to make
286 regular read/write async.
288 sg SCSI generic sg v3 io. May either be
289 synchronous using the SG_IO ioctl, or if
290 the target is an sg character device
291 we use read(2) and write(2) for asynchronous
294 null Doesn't transfer any data, just pretends
295 to. This is mainly used to exercise fio
296 itself and for debugging/testing purposes.
298 net Transfer over the network to given host:port.
299 'filename' must be set appropriately to
300 filename=host:port regardless of send
301 or receive, if the latter only the port
304 external Prefix to specify loading an external
305 IO engine object file. Append the engine
306 filename, eg ioengine=external:/tmp/foo.o
307 to load ioengine foo.o in /tmp.
309 iodepth=int This defines how many io units to keep in flight against
310 the file. The default is 1 for each file defined in this
311 job, can be overridden with a larger value for higher
314 iodepth_batch=int This defines how many pieces of IO to submit at once.
315 It defaults to the same as iodepth, but can be set lower
318 iodepth_low=int The low water mark indicating when to start filling
319 the queue again. Defaults to the same as iodepth, meaning
320 that fio will attempt to keep the queue full at all times.
321 If iodepth is set to eg 16 and iodepth_low is set to 4, then
322 after fio has filled the queue of 16 requests, it will let
323 the depth drain down to 4 before starting to fill it again.
325 direct=bool If value is true, use non-buffered io. This is usually
328 buffered=bool If value is true, use buffered io. This is the opposite
329 of the 'direct' option. Defaults to true.
331 offset=siint Start io at the given offset in the file. The data before
332 the given offset will not be touched. This effectively
333 caps the file size at real_size - offset.
335 fsync=int If writing to a file, issue a sync of the dirty data
336 for every number of blocks given. For example, if you give
337 32 as a parameter, fio will sync the file for every 32
338 writes issued. If fio is using non-buffered io, we may
339 not sync the file. The exception is the sg io engine, which
340 synchronizes the disk cache anyway.
342 overwrite=bool If writing to a file, setup the file first and do overwrites.
344 end_fsync=bool If true, fsync file contents when the job exits.
346 rwmixcycle=int Value in milliseconds describing how often to switch between
347 reads and writes for a mixed workload. The default is
350 rwmixread=int How large a percentage of the mix should be reads.
352 rwmixwrite=int How large a percentage of the mix should be writes. If both
353 rwmixread and rwmixwrite is given and the values do not add
354 up to 100%, the latter of the two will be used to override
357 norandommap Normally fio will cover every block of the file when doing
358 random IO. If this option is given, fio will just get a
359 new random offset without looking at past io history. This
360 means that some blocks may not be read or written, and that
361 some blocks may be read/written more than once. This option
362 is mutually exclusive with verify= for that reason.
364 nice=int Run the job with the given nice value. See man nice(2).
366 prio=int Set the io priority value of this job. Linux limits us to
367 a positive value between 0 and 7, with 0 being the highest.
370 prioclass=int Set the io priority class. See man ionice(1).
372 thinktime=int Stall the job x microseconds after an io has completed before
373 issuing the next. May be used to simulate processing being
374 done by an application. See thinktime_blocks and
378 Only valid if thinktime is set - pretend to spend CPU time
379 doing something with the data received, before falling back
380 to sleeping for the rest of the period specified by
384 Only valid if thinktime is set - control how many blocks
385 to issue, before waiting 'thinktime' usecs. If not set,
386 defaults to 1 which will make fio wait 'thinktime' usecs
389 rate=int Cap the bandwidth used by this job to this number of KiB/sec.
391 ratemin=int Tell fio to do whatever it can to maintain at least this
394 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
397 cpumask=int Set the CPU affinity of this job. The parameter given is a
398 bitmask of allowed CPU's the job may run on. See man
399 sched_setaffinity(2).
401 startdelay=int Start this job the specified number of seconds after fio
402 has started. Only useful if the job file contains several
403 jobs, and you want to delay starting some jobs to a certain
406 runtime=int Tell fio to terminate processing after the specified number
407 of seconds. It can be quite hard to determine for how long
408 a specified job will run, so this parameter is handy to
409 cap the total runtime to a given time.
411 invalidate=bool Invalidate the buffer/page cache parts for this file prior
412 to starting io. Defaults to true.
414 sync=bool Use sync io for buffered writes. For the majority of the
415 io engines, this means using O_SYNC.
417 mem=str Fio can use various types of memory as the io unit buffer.
418 The allowed values are:
420 malloc Use memory from malloc(3) as the buffers.
422 shm Use shared memory as the buffers. Allocated
425 shmhuge Same as shm, but use huge pages as backing.
427 mmap Use mmap to allocate buffers. May either be
428 anonymous memory, or can be file backed if
429 a filename is given after the option. The
430 format is mem=mmap:/path/to/file.
432 mmaphuge Use a memory mapped huge file as the buffer
433 backing. Append filename after mmaphuge, ala
434 mem=mmaphuge:/hugetlbfs/file
436 The area allocated is a function of the maximum allowed
437 bs size for the job, multiplied by the io depth given. Note
438 that for shmhuge and mmaphuge to work, the system must have
439 free huge pages allocated. This can normally be checked
440 and set by reading/writing /proc/sys/vm/nr_hugepages on a
441 Linux system. Fio assumes a huge page is 4MiB in size. So
442 to calculate the number of huge pages you need for a given
443 job file, add up the io depth of all jobs (normally one unless
444 iodepth= is used) and multiply by the maximum bs set. Then
445 divide that number by the huge page size. You can see the
446 size of the huge pages in /proc/meminfo. If no huge pages
447 are allocated by having a non-zero number in nr_hugepages,
448 using mmaphuge or shmhuge will fail. Also see hugepage-size.
450 mmaphuge also needs to have hugetlbfs mounted and the file
451 location should point there. So if it's mounted in /huge,
452 you would use mem=mmaphuge:/huge/somefile.
455 Defines the size of a huge page. Must at least be equal
456 to the system setting, see /proc/meminfo. Defaults to 4MiB.
457 Should probably always be a multiple of megabytes, so using
458 hugepage-size=Xm is the preferred way to set this to avoid
459 setting a non-pow-2 bad value.
461 exitall When one job finishes, terminate the rest. The default is
462 to wait for each job to finish, sometimes that is not the
465 bwavgtime=int Average the calculated bandwidth over the given time. Value
466 is specified in milliseconds.
468 create_serialize=bool If true, serialize the file creating for the jobs.
469 This may be handy to avoid interleaving of data
470 files, which may greatly depend on the filesystem
471 used and even the number of processors in the system.
473 create_fsync=bool fsync the data file after creation. This is the
476 unlink=bool Unlink the job files when done. Not the default, as repeated
477 runs of that job would then waste time recreating the fileset
480 loops=int Run the specified number of iterations of this job. Used
481 to repeat the same workload a given number of times. Defaults
484 verify=str If writing to a file, fio can verify the file contents
485 after each iteration of the job. The allowed values are:
487 md5 Use an md5 sum of the data area and store
488 it in the header of each block.
490 crc32 Use a crc32 sum of the data area and store
491 it in the header of each block.
493 This option can be used for repeated burn-in tests of a
494 system to make sure that the written data is also
497 stonewall Wait for preceeding jobs in the job file to exit, before
498 starting this one. Can be used to insert serialization
499 points in the job file.
501 numjobs=int Create the specified number of clones of this job. May be
502 used to setup a larger number of threads/processes doing
503 the same thing. We regard that grouping of jobs as a
506 group_reporting If 'numjobs' is set, it may be interesting to display
507 statistics for the group as a whole instead of for each
508 individual job. This is especially true of 'numjobs' is
509 large, looking at individual thread/process output quickly
510 becomes unwieldy. If 'group_reporting' is specified, fio
511 will show the final report per-group instead of per-job.
513 thread fio defaults to forking jobs, however if this option is
514 given, fio will use pthread_create(3) to create threads
517 zonesize=siint Divide a file into zones of the specified size. See zoneskip.
519 zoneskip=siint Skip the specified number of bytes when zonesize data has
520 been read. The two zone options can be used to only do
521 io on zones of a file.
523 write_iolog=str Write the issued io patterns to the specified file. See
526 read_iolog=str Open an iolog with the specified file name and replay the
527 io patterns it contains. This can be used to store a
528 workload and replay it sometime later.
530 write_bw_log If given, write a bandwidth log of the jobs in this job
531 file. Can be used to store data of the bandwidth of the
532 jobs in their lifetime. The included fio_generate_plots
533 script uses gnuplot to turn these text files into nice
536 write_lat_log Same as write_bw_log, except that this option stores io
537 completion latencies instead.
539 lockmem=siint Pin down the specified amount of memory with mlock(2). Can
540 potentially be used instead of removing memory or booting
541 with less memory to simulate a smaller amount of memory.
543 exec_prerun=str Before running this job, issue the command specified
546 exec_postrun=str After the job completes, issue the command specified
549 ioscheduler=str Attempt to switch the device hosting the file to the specified
550 io scheduler before running.
552 cpuload=int If the job is a CPU cycle eater, attempt to use the specified
553 percentage of CPU cycles.
555 cpuchunks=int If the job is a CPU cycle eater, split the load into
556 cycles of the given time. In milliseconds.
559 6.0 Interpreting the output
560 ---------------------------
562 fio spits out a lot of output. While running, fio will display the
563 status of the jobs created. An example of that would be:
565 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
567 The characters inside the square brackets denote the current status of
568 each thread. The possible values (in typical life cycle order) are:
572 P Thread setup, but not started.
574 I Thread initialized, waiting.
575 R Running, doing sequential reads.
576 r Running, doing random reads.
577 W Running, doing sequential writes.
578 w Running, doing random writes.
579 M Running, doing mixed sequential reads/writes.
580 m Running, doing mixed random reads/writes.
581 F Running, currently waiting for fsync()
582 V Running, doing verification of written data.
583 E Thread exited, not reaped by main thread yet.
586 The other values are fairly self explanatory - number of threads
587 currently running and doing io, rate of io since last check, and the estimated
588 completion percentage and time for the running group. It's impossible to
589 estimate runtime of the following groups (if any).
591 When fio is done (or interrupted by ctrl-c), it will show the data for
592 each thread, group of threads, and disks in that order. For each data
593 direction, the output looks like:
595 Client1 (g=0): err= 0:
596 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
597 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
598 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
599 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
600 cpu : usr=1.49%, sys=0.25%, ctx=7969
601 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
602 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
603 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
605 The client number is printed, along with the group id and error of that
606 thread. Below is the io statistics, here for writes. In the order listed,
609 io= Number of megabytes io performed
610 bw= Average bandwidth rate
611 runt= The runtime of that thread
612 slat= Submission latency (avg being the average, dev being the
613 standard deviation). This is the time it took to submit
614 the io. For sync io, the slat is really the completion
615 latency, since queue/complete is one operation there.
616 clat= Completion latency. Same names as slat, this denotes the
617 time from submission to completion of the io pieces. For
618 sync io, clat will usually be equal (or very close) to 0,
619 as the time from submit to complete is basically just
620 CPU time (io has already been done, see slat explanation).
621 bw= Bandwidth. Same names as the xlat stats, but also includes
622 an approximate percentage of total aggregate bandwidth
623 this thread received in this group. This last value is
624 only really useful if the threads in this group are on the
625 same disk, since they are then competing for disk access.
626 cpu= CPU usage. User and system time, along with the number
627 of context switches this thread went through.
628 IO depths= The distribution of io depths over the job life time. The
629 numbers are divided into powers of 2, so for example the
630 16= entries includes depths up to that value but higher
631 than the previous entry. In other words, it covers the
633 IO latencies= The distribution of IO completion latencies. This is the
634 time from when IO leaves fio and when it gets completed.
635 The numbers follow the same pattern as the IO depths,
636 meaning that 2=1.6% means that 1.6% of the IO completed
637 within 2 msecs, 20=12.8% means that 12.8% of the IO
638 took more than 10 msecs, but less than (or equal to) 20 msecs.
640 After each client has been listed, the group statistics are printed. They
643 Run status group 0 (all jobs):
644 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
645 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
647 For each data direction, it prints:
649 io= Number of megabytes io performed.
650 aggrb= Aggregate bandwidth of threads in this group.
651 minb= The minimum average bandwidth a thread saw.
652 maxb= The maximum average bandwidth a thread saw.
653 mint= The smallest runtime of the threads in that group.
654 maxt= The longest runtime of the threads in that group.
656 And finally, the disk statistics are printed. They will look like this:
658 Disk stats (read/write):
659 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
661 Each value is printed for both reads and writes, with reads first. The
664 ios= Number of ios performed by all groups.
665 merge= Number of merges io the io scheduler.
666 ticks= Number of ticks we kept the disk busy.
667 io_queue= Total time spent in the disk queue.
668 util= The disk utilization. A value of 100% means we kept the disk
669 busy constantly, 50% would be a disk idling half of the time.
675 For scripted usage where you typically want to generate tables or graphs
676 of the results, fio can output the results in a semicolon separated format.
677 The format is one long line of values, such as:
679 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%
680 ;0.0%;0.0%;0.0%;0.0%;0.0%
682 Split up, the format is as follows:
684 jobname, groupid, error
686 KiB IO, bandwidth (KiB/sec), runtime (msec)
687 Submission latency: min, max, mean, deviation
688 Completion latency: min, max, mean, deviation
689 Bw: min, max, aggregate percentage of total, mean, deviation
691 KiB IO, bandwidth (KiB/sec), runtime (msec)
692 Submission latency: min, max, mean, deviation
693 Completion latency: min, max, mean, deviation
694 Bw: min, max, aggregate percentage of total, mean, deviation
695 CPU usage: user, system, context switches
696 IO depths: <=1, 2, 4, 8, 16, 32, >=64
697 IO latencies: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000