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 file_service_type=str Defines how fio decides which file from a job to
257 service next. The following types are defined:
259 random Just choose a file at random.
261 roundrobin Round robin over open files. This
264 ioengine=str Defines how the job issues io to the file. The following
267 sync Basic read(2) or write(2) io. lseek(2) is
268 used to position the io location.
270 libaio Linux native asynchronous io.
272 posixaio glibc posix asynchronous io.
274 mmap File is memory mapped and data copied
275 to/from using memcpy(3).
277 splice splice(2) is used to transfer the data and
278 vmsplice(2) to transfer data from user
281 syslet-rw Use the syslet system calls to make
282 regular read/write async.
284 sg SCSI generic sg v3 io. May either be
285 synchronous using the SG_IO ioctl, or if
286 the target is an sg character device
287 we use read(2) and write(2) for asynchronous
290 null Doesn't transfer any data, just pretends
291 to. This is mainly used to exercise fio
292 itself and for debugging/testing purposes.
294 net Transfer over the network to given host:port.
295 'filename' must be set appropriately to
296 filename=host:port regardless of send
297 or receive, if the latter only the port
300 external Prefix to specify loading an external
301 IO engine object file. Append the engine
302 filename, eg ioengine=external:/tmp/foo.o
303 to load ioengine foo.o in /tmp.
305 iodepth=int This defines how many io units to keep in flight against
306 the file. The default is 1 for each file defined in this
307 job, can be overridden with a larger value for higher
310 iodepth_batch=int This defines how many pieces of IO to submit at once.
311 It defaults to the same as iodepth, but can be set lower
314 iodepth_low=int The low water mark indicating when to start filling
315 the queue again. Defaults to the same as iodepth, meaning
316 that fio will attempt to keep the queue full at all times.
317 If iodepth is set to eg 16 and iodepth_low is set to 4, then
318 after fio has filled the queue of 16 requests, it will let
319 the depth drain down to 4 before starting to fill it again.
321 direct=bool If value is true, use non-buffered io. This is usually
324 buffered=bool If value is true, use buffered io. This is the opposite
325 of the 'direct' option. Defaults to true.
327 offset=siint Start io at the given offset in the file. The data before
328 the given offset will not be touched. This effectively
329 caps the file size at real_size - offset.
331 fsync=int If writing to a file, issue a sync of the dirty data
332 for every number of blocks given. For example, if you give
333 32 as a parameter, fio will sync the file for every 32
334 writes issued. If fio is using non-buffered io, we may
335 not sync the file. The exception is the sg io engine, which
336 synchronizes the disk cache anyway.
338 overwrite=bool If writing to a file, setup the file first and do overwrites.
340 end_fsync=bool If true, fsync file contents when the job exits.
342 rwmixcycle=int Value in milliseconds describing how often to switch between
343 reads and writes for a mixed workload. The default is
346 rwmixread=int How large a percentage of the mix should be reads.
348 rwmixwrite=int How large a percentage of the mix should be writes. If both
349 rwmixread and rwmixwrite is given and the values do not add
350 up to 100%, the latter of the two will be used to override
353 norandommap Normally fio will cover every block of the file when doing
354 random IO. If this option is given, fio will just get a
355 new random offset without looking at past io history. This
356 means that some blocks may not be read or written, and that
357 some blocks may be read/written more than once. This option
358 is mutually exclusive with verify= for that reason.
360 nice=int Run the job with the given nice value. See man nice(2).
362 prio=int Set the io priority value of this job. Linux limits us to
363 a positive value between 0 and 7, with 0 being the highest.
366 prioclass=int Set the io priority class. See man ionice(1).
368 thinktime=int Stall the job x microseconds after an io has completed before
369 issuing the next. May be used to simulate processing being
370 done by an application. See thinktime_blocks and
374 Only valid if thinktime is set - pretend to spend CPU time
375 doing something with the data received, before falling back
376 to sleeping for the rest of the period specified by
380 Only valid if thinktime is set - control how many blocks
381 to issue, before waiting 'thinktime' usecs. If not set,
382 defaults to 1 which will make fio wait 'thinktime' usecs
385 rate=int Cap the bandwidth used by this job to this number of KiB/sec.
387 ratemin=int Tell fio to do whatever it can to maintain at least this
390 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
393 cpumask=int Set the CPU affinity of this job. The parameter given is a
394 bitmask of allowed CPU's the job may run on. See man
395 sched_setaffinity(2).
397 startdelay=int Start this job the specified number of seconds after fio
398 has started. Only useful if the job file contains several
399 jobs, and you want to delay starting some jobs to a certain
402 runtime=int Tell fio to terminate processing after the specified number
403 of seconds. It can be quite hard to determine for how long
404 a specified job will run, so this parameter is handy to
405 cap the total runtime to a given time.
407 invalidate=bool Invalidate the buffer/page cache parts for this file prior
408 to starting io. Defaults to true.
410 sync=bool Use sync io for buffered writes. For the majority of the
411 io engines, this means using O_SYNC.
413 mem=str Fio can use various types of memory as the io unit buffer.
414 The allowed values are:
416 malloc Use memory from malloc(3) as the buffers.
418 shm Use shared memory as the buffers. Allocated
421 shmhuge Same as shm, but use huge pages as backing.
423 mmap Use mmap to allocate buffers. May either be
424 anonymous memory, or can be file backed if
425 a filename is given after the option. The
426 format is mem=mmap:/path/to/file.
428 mmaphuge Use a memory mapped huge file as the buffer
429 backing. Append filename after mmaphuge, ala
430 mem=mmaphuge:/hugetlbfs/file
432 The area allocated is a function of the maximum allowed
433 bs size for the job, multiplied by the io depth given. Note
434 that for shmhuge and mmaphuge to work, the system must have
435 free huge pages allocated. This can normally be checked
436 and set by reading/writing /proc/sys/vm/nr_hugepages on a
437 Linux system. Fio assumes a huge page is 4MiB in size. So
438 to calculate the number of huge pages you need for a given
439 job file, add up the io depth of all jobs (normally one unless
440 iodepth= is used) and multiply by the maximum bs set. Then
441 divide that number by the huge page size. You can see the
442 size of the huge pages in /proc/meminfo. If no huge pages
443 are allocated by having a non-zero number in nr_hugepages,
444 using mmaphuge or shmhuge will fail. Also see hugepage-size.
446 mmaphuge also needs to have hugetlbfs mounted and the file
447 location should point there. So if it's mounted in /huge,
448 you would use mem=mmaphuge:/huge/somefile.
451 Defines the size of a huge page. Must at least be equal
452 to the system setting, see /proc/meminfo. Defaults to 4MiB.
453 Should probably always be a multiple of megabytes, so using
454 hugepage-size=Xm is the preferred way to set this to avoid
455 setting a non-pow-2 bad value.
457 exitall When one job finishes, terminate the rest. The default is
458 to wait for each job to finish, sometimes that is not the
461 bwavgtime=int Average the calculated bandwidth over the given time. Value
462 is specified in milliseconds.
464 create_serialize=bool If true, serialize the file creating for the jobs.
465 This may be handy to avoid interleaving of data
466 files, which may greatly depend on the filesystem
467 used and even the number of processors in the system.
469 create_fsync=bool fsync the data file after creation. This is the
472 unlink=bool Unlink the job files when done. Not the default, as repeated
473 runs of that job would then waste time recreating the fileset
476 loops=int Run the specified number of iterations of this job. Used
477 to repeat the same workload a given number of times. Defaults
480 verify=str If writing to a file, fio can verify the file contents
481 after each iteration of the job. The allowed values are:
483 md5 Use an md5 sum of the data area and store
484 it in the header of each block.
486 crc32 Use a crc32 sum of the data area and store
487 it in the header of each block.
489 This option can be used for repeated burn-in tests of a
490 system to make sure that the written data is also
493 stonewall Wait for preceeding jobs in the job file to exit, before
494 starting this one. Can be used to insert serialization
495 points in the job file.
497 numjobs=int Create the specified number of clones of this job. May be
498 used to setup a larger number of threads/processes doing
499 the same thing. We regard that grouping of jobs as a
502 group_reporting If 'numjobs' is set, it may be interesting to display
503 statistics for the group as a whole instead of for each
504 individual job. This is especially true of 'numjobs' is
505 large, looking at individual thread/process output quickly
506 becomes unwieldy. If 'group_reporting' is specified, fio
507 will show the final report per-group instead of per-job.
509 thread fio defaults to forking jobs, however if this option is
510 given, fio will use pthread_create(3) to create threads
513 zonesize=siint Divide a file into zones of the specified size. See zoneskip.
515 zoneskip=siint Skip the specified number of bytes when zonesize data has
516 been read. The two zone options can be used to only do
517 io on zones of a file.
519 write_iolog=str Write the issued io patterns to the specified file. See
522 read_iolog=str Open an iolog with the specified file name and replay the
523 io patterns it contains. This can be used to store a
524 workload and replay it sometime later.
526 write_bw_log If given, write a bandwidth log of the jobs in this job
527 file. Can be used to store data of the bandwidth of the
528 jobs in their lifetime. The included fio_generate_plots
529 script uses gnuplot to turn these text files into nice
532 write_lat_log Same as write_bw_log, except that this option stores io
533 completion latencies instead.
535 lockmem=siint Pin down the specified amount of memory with mlock(2). Can
536 potentially be used instead of removing memory or booting
537 with less memory to simulate a smaller amount of memory.
539 exec_prerun=str Before running this job, issue the command specified
542 exec_postrun=str After the job completes, issue the command specified
545 ioscheduler=str Attempt to switch the device hosting the file to the specified
546 io scheduler before running.
548 cpuload=int If the job is a CPU cycle eater, attempt to use the specified
549 percentage of CPU cycles.
551 cpuchunks=int If the job is a CPU cycle eater, split the load into
552 cycles of the given time. In milliseconds.
555 6.0 Interpreting the output
556 ---------------------------
558 fio spits out a lot of output. While running, fio will display the
559 status of the jobs created. An example of that would be:
561 Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
563 The characters inside the square brackets denote the current status of
564 each thread. The possible values (in typical life cycle order) are:
568 P Thread setup, but not started.
570 I Thread initialized, waiting.
571 R Running, doing sequential reads.
572 r Running, doing random reads.
573 W Running, doing sequential writes.
574 w Running, doing random writes.
575 M Running, doing mixed sequential reads/writes.
576 m Running, doing mixed random reads/writes.
577 F Running, currently waiting for fsync()
578 V Running, doing verification of written data.
579 E Thread exited, not reaped by main thread yet.
582 The other values are fairly self explanatory - number of threads
583 currently running and doing io, rate of io since last check, and the estimated
584 completion percentage and time for the running group. It's impossible to
585 estimate runtime of the following groups (if any).
587 When fio is done (or interrupted by ctrl-c), it will show the data for
588 each thread, group of threads, and disks in that order. For each data
589 direction, the output looks like:
591 Client1 (g=0): err= 0:
592 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
593 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
594 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
595 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
596 cpu : usr=1.49%, sys=0.25%, ctx=7969
597 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
598 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
599 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
601 The client number is printed, along with the group id and error of that
602 thread. Below is the io statistics, here for writes. In the order listed,
605 io= Number of megabytes io performed
606 bw= Average bandwidth rate
607 runt= The runtime of that thread
608 slat= Submission latency (avg being the average, dev being the
609 standard deviation). This is the time it took to submit
610 the io. For sync io, the slat is really the completion
611 latency, since queue/complete is one operation there.
612 clat= Completion latency. Same names as slat, this denotes the
613 time from submission to completion of the io pieces. For
614 sync io, clat will usually be equal (or very close) to 0,
615 as the time from submit to complete is basically just
616 CPU time (io has already been done, see slat explanation).
617 bw= Bandwidth. Same names as the xlat stats, but also includes
618 an approximate percentage of total aggregate bandwidth
619 this thread received in this group. This last value is
620 only really useful if the threads in this group are on the
621 same disk, since they are then competing for disk access.
622 cpu= CPU usage. User and system time, along with the number
623 of context switches this thread went through.
624 IO depths= The distribution of io depths over the job life time. The
625 numbers are divided into powers of 2, so for example the
626 16= entries includes depths up to that value but higher
627 than the previous entry. In other words, it covers the
629 IO latencies= The distribution of IO completion latencies. This is the
630 time from when IO leaves fio and when it gets completed.
631 The numbers follow the same pattern as the IO depths,
632 meaning that 2=1.6% means that 1.6% of the IO completed
633 within 2 msecs, 20=12.8% means that 12.8% of the IO
634 took more than 10 msecs, but less than (or equal to) 20 msecs.
636 After each client has been listed, the group statistics are printed. They
639 Run status group 0 (all jobs):
640 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
641 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
643 For each data direction, it prints:
645 io= Number of megabytes io performed.
646 aggrb= Aggregate bandwidth of threads in this group.
647 minb= The minimum average bandwidth a thread saw.
648 maxb= The maximum average bandwidth a thread saw.
649 mint= The smallest runtime of the threads in that group.
650 maxt= The longest runtime of the threads in that group.
652 And finally, the disk statistics are printed. They will look like this:
654 Disk stats (read/write):
655 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
657 Each value is printed for both reads and writes, with reads first. The
660 ios= Number of ios performed by all groups.
661 merge= Number of merges io the io scheduler.
662 ticks= Number of ticks we kept the disk busy.
663 io_queue= Total time spent in the disk queue.
664 util= The disk utilization. A value of 100% means we kept the disk
665 busy constantly, 50% would be a disk idling half of the time.
671 For scripted usage where you typically want to generate tables or graphs
672 of the results, fio can output the results in a semicolon separated format.
673 The format is one long line of values, such as:
675 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%
676 ;0.0%;0.0%;0.0%;0.0%;0.0%
678 Split up, the format is as follows:
680 jobname, groupid, error
682 KiB IO, bandwidth (KiB/sec), runtime (msec)
683 Submission latency: min, max, mean, deviation
684 Completion latency: min, max, mean, deviation
685 Bw: min, max, aggregate percentage of total, mean, deviation
687 KiB IO, bandwidth (KiB/sec), runtime (msec)
688 Submission latency: min, max, mean, deviation
689 Completion latency: min, max, mean, deviation
690 Bw: min, max, aggregate percentage of total, mean, deviation
691 CPU usage: user, system, context switches
692 IO depths: <=1, 2, 4, 8, 16, 32, >=64
693 IO latencies: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000
projects / fio.git / blob