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, 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 ';', the
112 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. Also see siint.
184 With the above in mind, here follows the complete list of fio job
187 name=str ASCII name of the job. This may be used to override the
188 name printed by fio for this job. Otherwise the job
189 name is used. On the command line this parameter has the
190 special purpose of also signaling the start of a new
193 directory=str Prefix filenames with this directory. Used to places files
194 in a different location than "./".
196 filename=str Fio normally makes up a filename based on the job name,
197 thread number, and file number. If you want to share
198 files between threads in a job or several jobs, specify
199 a filename for each of them to override the default.
201 rw=str Type of io pattern. Accepted values are:
203 read Sequential reads
204 write Sequential writes
205 randwrite Random writes
206 randread Random reads
207 rw Sequential mixed reads and writes
208 randrw Random mixed reads and writes
210 For the mixed io types, the default is to split them 50/50.
211 For certain types of io the result may still be skewed a bit,
212 since the speed may be different.
214 randrepeat=bool For random IO workloads, seed the generator in a predictable
215 way so that results are repeatable across repetitions.
217 size=siint The total size of file io for this job. This may describe
218 the size of the single file the job uses, or it may be
219 divided between the number of files in the job. If the
220 file already exists, the file size will be adjusted to this
221 size if larger than the current file size. If this parameter
222 is not given and the file exists, the file size will be used.
224 bs=siint The block size used for the io units. Defaults to 4k. Values
225 can be given for both read and writes. If a single siint is
226 given, it will apply to both. If a second siint is specified
227 after a comma, it will apply to writes only. In other words,
228 the format is either bs=read_and_write or bs=read,write.
229 bs=4k,8k will thus use 4k blocks for reads, and 8k blocks
230 for writes. If you only wish to set the write size, you
231 can do so by passing an empty read size - bs=,8k will set
232 8k for writes and leave the read default value.
234 bsrange=irange Instead of giving a single block size, specify a range
235 and fio will mix the issued io block sizes. The issued
236 io unit will always be a multiple of the minimum value
237 given (also see bs_unaligned). Applies to both reads and
238 writes, however a second range can be given after a comma.
241 bs_unaligned If this option is given, any byte size value within bsrange
242 may be used as a block range. This typically wont work with
243 direct IO, as that normally requires sector alignment.
245 nrfiles=int Number of files to use for this job. Defaults to 1.
247 ioengine=str Defines how the job issues io to the file. The following
250 sync Basic read(2) or write(2) io. lseek(2) is
251 used to position the io location.
253 libaio Linux native asynchronous io.
255 posixaio glibc posix asynchronous io.
257 mmap File is memory mapped and data copied
258 to/from using memcpy(3).
260 splice splice(2) is used to transfer the data and
261 vmsplice(2) to transfer data from user
264 sg SCSI generic sg v3 io. May either be
265 synchronous using the SG_IO ioctl, or if
266 the target is an sg character device
267 we use read(2) and write(2) for asynchronous
270 null Doesn't transfer any data, just pretends
271 to. This is mainly used to exercise fio
272 itself and for debugging/testing purposes.
274 iodepth=int This defines how many io units to keep in flight against
275 the file. The default is 1 for each file defined in this
276 job, can be overridden with a larger value for higher
279 direct=bool If value is true, use non-buffered io. This is usually
282 buffered=bool If value is true, use buffered io. This is the opposite
283 of the 'direct' option. Defaults to true.
285 offset=siint Start io at the given offset in the file. The data before
286 the given offset will not be touched. This effectively
287 caps the file size at real_size - offset.
289 fsync=int If writing to a file, issue a sync of the dirty data
290 for every number of blocks given. For example, if you give
291 32 as a parameter, fio will sync the file for every 32
292 writes issued. If fio is using non-buffered io, we may
293 not sync the file. The exception is the sg io engine, which
294 synchronizes the disk cache anyway.
296 overwrite=bool If writing to a file, setup the file first and do overwrites.
298 end_fsync=bool If true, fsync file contents when the job exits.
300 rwmixcycle=int Value in milliseconds describing how often to switch between
301 reads and writes for a mixed workload. The default is
304 rwmixread=int How large a percentage of the mix should be reads.
306 rwmixwrite=int How large a percentage of the mix should be writes. If both
307 rwmixread and rwmixwrite is given and the values do not add
308 up to 100%, the latter of the two will be used to override
311 norandommap Normally fio will cover every block of the file when doing
312 random IO. If this option is given, fio will just get a
313 new random offset without looking at past io history. This
314 means that some blocks may not be read or written, and that
315 some blocks may be read/written more than once. This option
316 is mutually exclusive with verify= for that reason.
318 nice=int Run the job with the given nice value. See man nice(2).
320 prio=int Set the io priority value of this job. Linux limits us to
321 a positive value between 0 and 7, with 0 being the highest.
324 prioclass=int Set the io priority class. See man ionice(1).
326 thinktime=int Stall the job x microseconds after an io has completed before
327 issuing the next. May be used to simulate processing being
328 done by an application. See thinktime_blocks.
331 Only valid if thinktime is set - control how many blocks
332 to issue, before waiting 'thinktime' usecs. If not set,
333 defaults to 1 which will make fio wait 'thinktime' usecs
336 rate=int Cap the bandwidth used by this job to this number of KiB/sec.
338 ratemin=int Tell fio to do whatever it can to maintain at least this
341 ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
344 cpumask=int Set the CPU affinity of this job. The parameter given is a
345 bitmask of allowed CPU's the job may run on. See man
346 sched_setaffinity(2).
348 startdelay=int Start this job the specified number of seconds after fio
349 has started. Only useful if the job file contains several
350 jobs, and you want to delay starting some jobs to a certain
353 runtime=int Tell fio to terminate processing after the specified number
354 of seconds. It can be quite hard to determine for how long
355 a specified job will run, so this parameter is handy to
356 cap the total runtime to a given time.
358 invalidate=bool Invalidate the buffer/page cache parts for this file prior
359 to starting io. Defaults to true.
361 sync=bool Use sync io for buffered writes. For the majority of the
362 io engines, this means using O_SYNC.
364 mem=str Fio can use various types of memory as the io unit buffer.
365 The allowed values are:
367 malloc Use memory from malloc(3) as the buffers.
369 shm Use shared memory as the buffers. Allocated
372 shmhuge Same as shm, but use huge pages as backing.
374 mmap Use mmap to allocate buffers. May either be
375 anonymous memory, or can be file backed if
376 a filename is given after the option. The
377 format is mem=mmap:/path/to/file.
379 mmaphuge Use a memory mapped huge file as the buffer
380 backing. Append filename after mmaphuge, ala
381 mem=mmaphuge:/hugetlbfs/file
383 The area allocated is a function of the maximum allowed
384 bs size for the job, multiplied by the io depth given. Note
385 that for shmhuge and mmaphuge to work, the system must have
386 free huge pages allocated. This can normally be checked
387 and set by reading/writing /proc/sys/vm/nr_hugepages on a
388 Linux system. Fio assumes a huge page is 4MiB in size. So
389 to calculate the number of huge pages you need for a given
390 job file, add up the io depth of all jobs (normally one unless
391 iodepth= is used) and multiply by the maximum bs set. Then
392 divide that number by the huge page size. You can see the
393 size of the huge pages in /proc/meminfo. If no huge pages
394 are allocated by having a non-zero number in nr_hugepages,
395 using mmaphuge or shmhuge will fail. Also see hugepage-size.
397 mmaphuge also needs to have hugetlbfs mounted and the file
398 location should point there. So if it's mounted in /huge,
399 you would use mem=mmaphuge:/huge/somefile.
402 Defines the size of a huge page. Must at least be equal
403 to the system setting, see /proc/meminfo. Defaults to 4MiB.
404 Should probably always be a multiple of megabytes, so using
405 hugepage-size=Xm is the preferred way to set this to avoid
406 setting a non-pow-2 bad value.
408 exitall When one job finishes, terminate the rest. The default is
409 to wait for each job to finish, sometimes that is not the
412 bwavgtime=int Average the calculated bandwidth over the given time. Value
413 is specified in milliseconds.
415 create_serialize=bool If true, serialize the file creating for the jobs.
416 This may be handy to avoid interleaving of data
417 files, which may greatly depend on the filesystem
418 used and even the number of processors in the system.
420 create_fsync=bool fsync the data file after creation. This is the
423 unlink=bool Unlink the job files when done. fio defaults to doing this,
424 if it created the file itself.
426 loops=int Run the specified number of iterations of this job. Used
427 to repeat the same workload a given number of times. Defaults
430 verify=str If writing to a file, fio can verify the file contents
431 after each iteration of the job. The allowed values are:
433 md5 Use an md5 sum of the data area and store
434 it in the header of each block.
436 crc32 Use a crc32 sum of the data area and store
437 it in the header of each block.
439 This option can be used for repeated burn-in tests of a
440 system to make sure that the written data is also
443 stonewall Wait for preceeding jobs in the job file to exit, before
444 starting this one. Can be used to insert serialization
445 points in the job file.
447 numjobs=int Create the specified number of clones of this job. May be
448 used to setup a larger number of threads/processes doing
451 thread fio defaults to forking jobs, however if this option is
452 given, fio will use pthread_create(3) to create threads
455 zonesize=siint Divide a file into zones of the specified size. See zoneskip.
457 zoneskip=siint Skip the specified number of bytes when zonesize data has
458 been read. The two zone options can be used to only do
459 io on zones of a file.
461 write_iolog=str Write the issued io patterns to the specified file. See
464 read_iolog=str Open an iolog with the specified file name and replay the
465 io patterns it contains. This can be used to store a
466 workload and replay it sometime later.
468 write_bw_log If given, write a bandwidth log of the jobs in this job
469 file. Can be used to store data of the bandwidth of the
470 jobs in their lifetime. The included fio_generate_plots
471 script uses gnuplot to turn these text files into nice
474 write_lat_log Same as write_bw_log, except that this option stores io
475 completion latencies instead.
477 lockmem=siint Pin down the specified amount of memory with mlock(2). Can
478 potentially be used instead of removing memory or booting
479 with less memory to simulate a smaller amount of memory.
481 exec_prerun=str Before running this job, issue the command specified
484 exec_postrun=str After the job completes, issue the command specified
487 ioscheduler=str Attempt to switch the device hosting the file to the specified
488 io scheduler before running.
490 cpuload=int If the job is a CPU cycle eater, attempt to use the specified
491 percentage of CPU cycles.
493 cpuchunks=int If the job is a CPU cycle eater, split the load into
494 cycles of the given time. In milliseconds.
497 6.0 Interpreting the output
498 ---------------------------
500 fio spits out a lot of output. While running, fio will display the
501 status of the jobs created. An example of that would be:
503 Threads running: 1: [_r] [24.79% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
505 The characters inside the square brackets denote the current status of
506 each thread. The possible values (in typical life cycle order) are:
510 P Thread setup, but not started.
512 I Thread initialized, waiting.
513 R Running, doing sequential reads.
514 r Running, doing random reads.
515 W Running, doing sequential writes.
516 w Running, doing random writes.
517 M Running, doing mixed sequential reads/writes.
518 m Running, doing mixed random reads/writes.
519 F Running, currently waiting for fsync()
520 V Running, doing verification of written data.
521 E Thread exited, not reaped by main thread yet.
524 The other values are fairly self explanatory - number of threads
525 currently running and doing io, rate of io since last check, and the estimated
526 completion percentage and time for the running group. It's impossible to
527 estimate runtime of the following groups (if any).
529 When fio is done (or interrupted by ctrl-c), it will show the data for
530 each thread, group of threads, and disks in that order. For each data
531 direction, the output looks like:
533 Client1 (g=0): err= 0:
534 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
535 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
536 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
537 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
538 cpu : usr=1.49%, sys=0.25%, ctx=7969
540 The client number is printed, along with the group id and error of that
541 thread. Below is the io statistics, here for writes. In the order listed,
544 io= Number of megabytes io performed
545 bw= Average bandwidth rate
546 runt= The runtime of that thread
547 slat= Submission latency (avg being the average, dev being the
548 standard deviation). This is the time it took to submit
549 the io. For sync io, the slat is really the completion
550 latency, since queue/complete is one operation there.
551 clat= Completion latency. Same names as slat, this denotes the
552 time from submission to completion of the io pieces. For
553 sync io, clat will usually be equal (or very close) to 0,
554 as the time from submit to complete is basically just
555 CPU time (io has already been done, see slat explanation).
556 bw= Bandwidth. Same names as the xlat stats, but also includes
557 an approximate percentage of total aggregate bandwidth
558 this thread received in this group. This last value is
559 only really useful if the threads in this group are on the
560 same disk, since they are then competing for disk access.
561 cpu= CPU usage. User and system time, along with the number
562 of context switches this thread went through.
564 After each client has been listed, the group statistics are printed. They
567 Run status group 0 (all jobs):
568 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
569 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
571 For each data direction, it prints:
573 io= Number of megabytes io performed.
574 aggrb= Aggregate bandwidth of threads in this group.
575 minb= The minimum average bandwidth a thread saw.
576 maxb= The maximum average bandwidth a thread saw.
577 mint= The smallest runtime of the threads in that group.
578 maxt= The longest runtime of the threads in that group.
580 And finally, the disk statistics are printed. They will look like this:
582 Disk stats (read/write):
583 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
585 Each value is printed for both reads and writes, with reads first. The
588 ios= Number of ios performed by all groups.
589 merge= Number of merges io the io scheduler.
590 ticks= Number of ticks we kept the disk busy.
591 io_queue= Total time spent in the disk queue.
592 util= The disk utilization. A value of 100% means we kept the disk
593 busy constantly, 50% would be a disk idling half of the time.
599 For scripted usage where you typically want to generate tables or graphs
600 of the results, fio can output the results in a comma separated format.
601 The format is one long line of values, such as:
603 client1,0,0,936,331,2894,0,0,0.000000,0.000000,1,170,22.115385,34.290410,16,714,84.252874%,366.500000,566.417819,3496,1237,2894,0,0,0.000000,0.000000,0,246,6.671625,21.436952,0,2534,55.465300%,1406.600000,2008.044216,0.000000%,0.431928%,1109
605 Split up, the format is as follows:
607 jobname, groupid, error
609 KiB IO, bandwidth (KiB/sec), runtime (msec)
610 Submission latency: min, max, mean, deviation
611 Completion latency: min, max, mean, deviation
612 Bw: min, max, aggregate percentage of total, mean, deviation
614 KiB IO, bandwidth (KiB/sec), runtime (msec)
615 Submission latency: min, max, mean, deviation
616 Completion latency: min, max, mean, deviation
617 Bw: min, max, aggregate percentage of total, mean, deviation
618 CPU usage: user, system, context switches