Add 'fsync_on_close' option
[fio.git] / HOWTO
CommitLineData
71bfa161
JA
1Table of contents
2-----------------
3
41. Overview
52. How fio works
63. Running fio
74. Job file format
85. Detailed list of parameters
96. Normal output
107. Terse output
11
12
131.0 Overview and history
14------------------------
15fio was originally written to save me the hassle of writing special test
16case programs when I wanted to test a specific workload, either for
17performance reasons or to find/reproduce a bug. The process of writing
18such a test app can be tiresome, especially if you have to do it often.
19Hence I needed a tool that would be able to simulate a given io workload
20without resorting to writing a tailored test case again and again.
21
22A test work load is difficult to define, though. There can be any number
23of processes or threads involved, and they can each be using their own
24way of generating io. You could have someone dirtying large amounts of
25memory in an memory mapped file, or maybe several threads issuing
26reads using asynchronous io. fio needed to be flexible enough to
27simulate both of these cases, and many more.
28
292.0 How fio works
30-----------------
31The first step in getting fio to simulate a desired io workload, is
32writing a job file describing that specific setup. A job file may contain
33any number of threads and/or files - the typical contents of the job file
34is a global section defining shared parameters, and one or more job
35sections describing the jobs involved. When run, fio parses this file
36and sets everything up as described. If we break down a job from top to
37bottom, it contains the following basic parameters:
38
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.
43
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
46 block sizes.
47
48 IO size How much data are we going to be reading/writing.
49
50 IO engine How do we issue io? We could be memory mapping the
51 file, we could be using regular read/write, we
d0ff85df 52 could be using splice, async io, syslet, or even
71bfa161
JA
53 SG (SCSI generic sg).
54
6c219763 55 IO depth If the io engine is async, how large a queuing
71bfa161
JA
56 depth do we want to maintain?
57
58 IO type Should we be doing buffered io, or direct/raw io?
59
60 Num files How many files are we spreading the workload over.
61
62 Num threads How many threads or processes should we spread
63 this workload over.
64
65The above are the basic parameters defined for a workload, in addition
66there's a multitude of parameters that modify other aspects of how this
67job behaves.
68
69
703.0 Running fio
71---------------
72See the README file for command line parameters, there are only a few
73of them.
74
75Running fio is normally the easiest part - you just give it the job file
76(or job files) as parameters:
77
78$ fio job_file
79
80and it will start doing what the job_file tells it to do. You can give
81more than one job file on the command line, fio will serialize the running
82of those files. Internally that is the same as using the 'stonewall'
83parameter described the the parameter section.
84
b4692828
JA
85If the job file contains only one job, you may as well just give the
86parameters on the command line. The command line parameters are identical
87to the job parameters, with a few extra that control global parameters
88(see README). For example, for the job file parameter iodepth=2, the
c2b1e753
JA
89mirror command line option would be --iodepth 2 or --iodepth=2. You can
90also 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.
92Command line entries following a --name entry will apply to that job,
93until there are no more entries or a new --name entry is seen. This is
94similar to the job file options, where each option applies to the current
95job until a new [] job entry is seen.
b4692828 96
71bfa161
JA
97fio does not need to run as root, except if the files or devices specified
98in the job section requires that. Some other options may also be restricted,
6c219763 99such as memory locking, io scheduler switching, and decreasing the nice value.
71bfa161
JA
100
101
1024.0 Job file format
103-------------------
104As previously described, fio accepts one or more job files describing
105what it is supposed to do. The job file format is the classic ini file,
106where the names enclosed in [] brackets define the job name. You are free
107to use any ascii name you want, except 'global' which has special meaning.
108A global section sets defaults for the jobs described in that file. A job
109may override a global section parameter, and a job file may even have
110several global sections if so desired. A job is only affected by a global
65db0851
JA
111section residing above it. If the first character in a line is a ';' or a
112'#', the entire line is discarded as a comment.
71bfa161
JA
113
114So lets look at a really simple job file that define to threads, each
115randomly reading from a 128MiB file.
116
117; -- start job file --
118[global]
119rw=randread
120size=128m
121
122[job1]
123
124[job2]
125
126; -- end job file --
127
128As you can see, the job file sections themselves are empty as all the
129described parameters are shared. As no filename= option is given, fio
c2b1e753
JA
130makes up a filename for each of the jobs as it sees fit. On the command
131line, this job would look as follows:
132
133$ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
134
71bfa161
JA
135
136Lets look at an example that have a number of processes writing randomly
137to files.
138
139; -- start job file --
140[random-writers]
141ioengine=libaio
142iodepth=4
143rw=randwrite
144bs=32k
145direct=0
146size=64m
147numjobs=4
148
149; -- end job file --
150
151Here we have no global section, as we only have one job defined anyway.
152We want to use async io here, with a depth of 4 for each file. We also
153increased the buffer size used to 32KiB and define numjobs to 4 to
154fork 4 identical jobs. The result is 4 processes each randomly writing
b4692828
JA
155to their own 64MiB file. Instead of using the above job file, you could
156have given the parameters on the command line. For this case, you would
157specify:
158
159$ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
71bfa161
JA
160
161fio ships with a few example job files, you can also look there for
162inspiration.
163
164
1655.0 Detailed list of parameters
166-------------------------------
167
168This section describes in details each parameter associated with a job.
169Some parameters take an option of a given type, such as an integer or
170a string. The following types are used:
171
172str String. This is a sequence of alpha characters.
173int Integer. A whole number value, may be negative.
174siint SI integer. A whole number value, which may contain a postfix
175 describing the base of the number. Accepted postfixes are k/m/g,
6c219763 176 meaning kilo, mega, and giga. So if you want to specify 4096,
71bfa161
JA
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.
179bool Boolean. Usually parsed as an integer, however only defined for
180 true and false (1 and 0).
181irange Integer range with postfix. Allows value range to be given, such
0c9baf91
JA
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
185 siint.
71bfa161
JA
186
187With the above in mind, here follows the complete list of fio job
188parameters.
189
190name=str ASCII name of the job. This may be used to override the
191 name printed by fio for this job. Otherwise the job
c2b1e753 192 name is used. On the command line this parameter has the
6c219763 193 special purpose of also signaling the start of a new
c2b1e753 194 job.
71bfa161 195
61697c37
JA
196description=str Text description of the job. Doesn't do anything except
197 dump this text description when this job is run. It's
198 not parsed.
199
71bfa161
JA
200directory=str Prefix filenames with this directory. Used to places files
201 in a different location than "./".
202
203filename=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
ed92ac0c
JA
206 a filename for each of them to override the default. If
207 the ioengine used is 'net', the filename is the host and
9f9214f2 208 port to connect to in the format of =host/port. If the
af52b345
JA
209 ioengine is file based, you can specify a number of files
210 by seperating the names with a ':' colon. So if you wanted
211 a job to open /dev/sda and /dev/sdb as the two working files,
212 you would use filename=/dev/sda:/dev/sdb
71bfa161
JA
213
214rw=str Type of io pattern. Accepted values are:
215
216 read Sequential reads
217 write Sequential writes
218 randwrite Random writes
219 randread Random reads
220 rw Sequential mixed reads and writes
221 randrw Random mixed reads and writes
222
223 For the mixed io types, the default is to split them 50/50.
224 For certain types of io the result may still be skewed a bit,
225 since the speed may be different.
226
ee738499
JA
227randrepeat=bool For random IO workloads, seed the generator in a predictable
228 way so that results are repeatable across repetitions.
229
71bfa161
JA
230size=siint The total size of file io for this job. This may describe
231 the size of the single file the job uses, or it may be
232 divided between the number of files in the job. If the
233 file already exists, the file size will be adjusted to this
234 size if larger than the current file size. If this parameter
235 is not given and the file exists, the file size will be used.
236
f90eff5a
JA
237bs=siint The block size used for the io units. Defaults to 4k. Values
238 can be given for both read and writes. If a single siint is
239 given, it will apply to both. If a second siint is specified
240 after a comma, it will apply to writes only. In other words,
241 the format is either bs=read_and_write or bs=read,write.
242 bs=4k,8k will thus use 4k blocks for reads, and 8k blocks
787f7e95
JA
243 for writes. If you only wish to set the write size, you
244 can do so by passing an empty read size - bs=,8k will set
245 8k for writes and leave the read default value.
a00735e6 246
71bfa161
JA
247bsrange=irange Instead of giving a single block size, specify a range
248 and fio will mix the issued io block sizes. The issued
249 io unit will always be a multiple of the minimum value
f90eff5a
JA
250 given (also see bs_unaligned). Applies to both reads and
251 writes, however a second range can be given after a comma.
252 See bs=.
a00735e6 253
690adba3
JA
254bs_unaligned If this option is given, any byte size value within bsrange
255 may be used as a block range. This typically wont work with
256 direct IO, as that normally requires sector alignment.
71bfa161
JA
257
258nrfiles=int Number of files to use for this job. Defaults to 1.
259
390b1537
JA
260openfiles=int Number of files to keep open at the same time. Defaults to
261 the same as nrfiles, can be set smaller to limit the number
262 simultaneous opens.
263
5af1c6f3
JA
264file_service_type=str Defines how fio decides which file from a job to
265 service next. The following types are defined:
266
267 random Just choose a file at random.
268
269 roundrobin Round robin over open files. This
270 is the default.
271
1907dbc6
JA
272 The string can have a number appended, indicating how
273 often to switch to a new file. So if option random:4 is
274 given, fio will switch to a new random file after 4 ios
275 have been issued.
276
71bfa161
JA
277ioengine=str Defines how the job issues io to the file. The following
278 types are defined:
279
280 sync Basic read(2) or write(2) io. lseek(2) is
281 used to position the io location.
282
283 libaio Linux native asynchronous io.
284
285 posixaio glibc posix asynchronous io.
286
287 mmap File is memory mapped and data copied
288 to/from using memcpy(3).
289
290 splice splice(2) is used to transfer the data and
291 vmsplice(2) to transfer data from user
292 space to the kernel.
293
d0ff85df
JA
294 syslet-rw Use the syslet system calls to make
295 regular read/write async.
296
71bfa161 297 sg SCSI generic sg v3 io. May either be
6c219763 298 synchronous using the SG_IO ioctl, or if
71bfa161
JA
299 the target is an sg character device
300 we use read(2) and write(2) for asynchronous
301 io.
302
a94ea28b
JA
303 null Doesn't transfer any data, just pretends
304 to. This is mainly used to exercise fio
305 itself and for debugging/testing purposes.
306
ed92ac0c
JA
307 net Transfer over the network to given host:port.
308 'filename' must be set appropriately to
9f9214f2 309 filename=host/port regardless of send
ed92ac0c
JA
310 or receive, if the latter only the port
311 argument is used.
312
ba0fbe10
JA
313 cpu Doesn't transfer any data, but burns CPU
314 cycles according to the cpuload= and
315 cpucycle= options. Setting cpuload=85
316 will cause that job to do nothing but burn
317 85% of the CPU.
318
8a7bd877
JA
319 external Prefix to specify loading an external
320 IO engine object file. Append the engine
321 filename, eg ioengine=external:/tmp/foo.o
322 to load ioengine foo.o in /tmp.
323
71bfa161
JA
324iodepth=int This defines how many io units to keep in flight against
325 the file. The default is 1 for each file defined in this
326 job, can be overridden with a larger value for higher
327 concurrency.
328
cb5ab512
JA
329iodepth_batch=int This defines how many pieces of IO to submit at once.
330 It defaults to the same as iodepth, but can be set lower
331 if one so desires.
332
e916b390
JA
333iodepth_low=int The low water mark indicating when to start filling
334 the queue again. Defaults to the same as iodepth, meaning
335 that fio will attempt to keep the queue full at all times.
336 If iodepth is set to eg 16 and iodepth_low is set to 4, then
337 after fio has filled the queue of 16 requests, it will let
338 the depth drain down to 4 before starting to fill it again.
339
71bfa161 340direct=bool If value is true, use non-buffered io. This is usually
76a43db4
JA
341 O_DIRECT.
342
343buffered=bool If value is true, use buffered io. This is the opposite
344 of the 'direct' option. Defaults to true.
71bfa161
JA
345
346offset=siint Start io at the given offset in the file. The data before
347 the given offset will not be touched. This effectively
348 caps the file size at real_size - offset.
349
350fsync=int If writing to a file, issue a sync of the dirty data
351 for every number of blocks given. For example, if you give
352 32 as a parameter, fio will sync the file for every 32
353 writes issued. If fio is using non-buffered io, we may
354 not sync the file. The exception is the sg io engine, which
6c219763 355 synchronizes the disk cache anyway.
71bfa161
JA
356
357overwrite=bool If writing to a file, setup the file first and do overwrites.
358
359end_fsync=bool If true, fsync file contents when the job exits.
360
ebb1415f
JA
361fsync_on_close=bool If true, fio will fsync() a dirty file on close.
362 This differs from end_fsync in that it will happen on every
363 file close, not just at the end of the job.
364
6c219763 365rwmixcycle=int Value in milliseconds describing how often to switch between
71bfa161
JA
366 reads and writes for a mixed workload. The default is
367 500 msecs.
368
369rwmixread=int How large a percentage of the mix should be reads.
370
371rwmixwrite=int How large a percentage of the mix should be writes. If both
372 rwmixread and rwmixwrite is given and the values do not add
373 up to 100%, the latter of the two will be used to override
374 the first.
375
bb8895e0
JA
376norandommap Normally fio will cover every block of the file when doing
377 random IO. If this option is given, fio will just get a
378 new random offset without looking at past io history. This
379 means that some blocks may not be read or written, and that
380 some blocks may be read/written more than once. This option
381 is mutually exclusive with verify= for that reason.
382
71bfa161
JA
383nice=int Run the job with the given nice value. See man nice(2).
384
385prio=int Set the io priority value of this job. Linux limits us to
386 a positive value between 0 and 7, with 0 being the highest.
387 See man ionice(1).
388
389prioclass=int Set the io priority class. See man ionice(1).
390
391thinktime=int Stall the job x microseconds after an io has completed before
392 issuing the next. May be used to simulate processing being
48097d5c
JA
393 done by an application. See thinktime_blocks and
394 thinktime_spin.
395
396thinktime_spin=int
397 Only valid if thinktime is set - pretend to spend CPU time
398 doing something with the data received, before falling back
399 to sleeping for the rest of the period specified by
400 thinktime.
9c1f7434
JA
401
402thinktime_blocks
403 Only valid if thinktime is set - control how many blocks
404 to issue, before waiting 'thinktime' usecs. If not set,
405 defaults to 1 which will make fio wait 'thinktime' usecs
406 after every block.
71bfa161
JA
407
408rate=int Cap the bandwidth used by this job to this number of KiB/sec.
409
410ratemin=int Tell fio to do whatever it can to maintain at least this
411 bandwidth.
412
413ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
6c219763 414 of milliseconds.
71bfa161
JA
415
416cpumask=int Set the CPU affinity of this job. The parameter given is a
417 bitmask of allowed CPU's the job may run on. See man
418 sched_setaffinity(2).
419
420startdelay=int Start this job the specified number of seconds after fio
421 has started. Only useful if the job file contains several
422 jobs, and you want to delay starting some jobs to a certain
423 time.
424
03b74b3e 425runtime=int Tell fio to terminate processing after the specified number
71bfa161
JA
426 of seconds. It can be quite hard to determine for how long
427 a specified job will run, so this parameter is handy to
428 cap the total runtime to a given time.
429
430invalidate=bool Invalidate the buffer/page cache parts for this file prior
431 to starting io. Defaults to true.
432
433sync=bool Use sync io for buffered writes. For the majority of the
434 io engines, this means using O_SYNC.
435
436mem=str Fio can use various types of memory as the io unit buffer.
437 The allowed values are:
438
439 malloc Use memory from malloc(3) as the buffers.
440
441 shm Use shared memory as the buffers. Allocated
442 through shmget(2).
443
74b025b0
JA
444 shmhuge Same as shm, but use huge pages as backing.
445
313cb206
JA
446 mmap Use mmap to allocate buffers. May either be
447 anonymous memory, or can be file backed if
448 a filename is given after the option. The
449 format is mem=mmap:/path/to/file.
71bfa161 450
d0bdaf49
JA
451 mmaphuge Use a memory mapped huge file as the buffer
452 backing. Append filename after mmaphuge, ala
453 mem=mmaphuge:/hugetlbfs/file
454
71bfa161 455 The area allocated is a function of the maximum allowed
5394ae5f
JA
456 bs size for the job, multiplied by the io depth given. Note
457 that for shmhuge and mmaphuge to work, the system must have
458 free huge pages allocated. This can normally be checked
459 and set by reading/writing /proc/sys/vm/nr_hugepages on a
460 Linux system. Fio assumes a huge page is 4MiB in size. So
461 to calculate the number of huge pages you need for a given
462 job file, add up the io depth of all jobs (normally one unless
463 iodepth= is used) and multiply by the maximum bs set. Then
464 divide that number by the huge page size. You can see the
465 size of the huge pages in /proc/meminfo. If no huge pages
466 are allocated by having a non-zero number in nr_hugepages,
56bb17f2 467 using mmaphuge or shmhuge will fail. Also see hugepage-size.
5394ae5f
JA
468
469 mmaphuge also needs to have hugetlbfs mounted and the file
470 location should point there. So if it's mounted in /huge,
471 you would use mem=mmaphuge:/huge/somefile.
71bfa161 472
56bb17f2
JA
473hugepage-size=siint
474 Defines the size of a huge page. Must at least be equal
475 to the system setting, see /proc/meminfo. Defaults to 4MiB.
c51074e7
JA
476 Should probably always be a multiple of megabytes, so using
477 hugepage-size=Xm is the preferred way to set this to avoid
478 setting a non-pow-2 bad value.
56bb17f2 479
71bfa161
JA
480exitall When one job finishes, terminate the rest. The default is
481 to wait for each job to finish, sometimes that is not the
482 desired action.
483
484bwavgtime=int Average the calculated bandwidth over the given time. Value
6c219763 485 is specified in milliseconds.
71bfa161
JA
486
487create_serialize=bool If true, serialize the file creating for the jobs.
488 This may be handy to avoid interleaving of data
489 files, which may greatly depend on the filesystem
490 used and even the number of processors in the system.
491
492create_fsync=bool fsync the data file after creation. This is the
493 default.
494
e545a6ce
JA
495unlink=bool Unlink the job files when done. Not the default, as repeated
496 runs of that job would then waste time recreating the fileset
497 again and again.
71bfa161
JA
498
499loops=int Run the specified number of iterations of this job. Used
500 to repeat the same workload a given number of times. Defaults
501 to 1.
502
503verify=str If writing to a file, fio can verify the file contents
504 after each iteration of the job. The allowed values are:
505
506 md5 Use an md5 sum of the data area and store
507 it in the header of each block.
508
509 crc32 Use a crc32 sum of the data area and store
510 it in the header of each block.
511
6c219763 512 This option can be used for repeated burn-in tests of a
71bfa161
JA
513 system to make sure that the written data is also
514 correctly read back.
515
516stonewall Wait for preceeding jobs in the job file to exit, before
517 starting this one. Can be used to insert serialization
518 points in the job file.
519
520numjobs=int Create the specified number of clones of this job. May be
521 used to setup a larger number of threads/processes doing
fa28c85a
JA
522 the same thing. We regard that grouping of jobs as a
523 specific group.
524
525group_reporting If 'numjobs' is set, it may be interesting to display
526 statistics for the group as a whole instead of for each
527 individual job. This is especially true of 'numjobs' is
528 large, looking at individual thread/process output quickly
529 becomes unwieldy. If 'group_reporting' is specified, fio
530 will show the final report per-group instead of per-job.
71bfa161
JA
531
532thread fio defaults to forking jobs, however if this option is
533 given, fio will use pthread_create(3) to create threads
534 instead.
535
536zonesize=siint Divide a file into zones of the specified size. See zoneskip.
537
538zoneskip=siint Skip the specified number of bytes when zonesize data has
539 been read. The two zone options can be used to only do
540 io on zones of a file.
541
076efc7c
JA
542write_iolog=str Write the issued io patterns to the specified file. See
543 read_iolog.
71bfa161 544
076efc7c 545read_iolog=str Open an iolog with the specified file name and replay the
71bfa161
JA
546 io patterns it contains. This can be used to store a
547 workload and replay it sometime later.
548
549write_bw_log If given, write a bandwidth log of the jobs in this job
550 file. Can be used to store data of the bandwidth of the
e0da9bc2
JA
551 jobs in their lifetime. The included fio_generate_plots
552 script uses gnuplot to turn these text files into nice
553 graphs.
71bfa161
JA
554
555write_lat_log Same as write_bw_log, except that this option stores io
556 completion latencies instead.
557
558lockmem=siint Pin down the specified amount of memory with mlock(2). Can
559 potentially be used instead of removing memory or booting
560 with less memory to simulate a smaller amount of memory.
561
562exec_prerun=str Before running this job, issue the command specified
563 through system(3).
564
565exec_postrun=str After the job completes, issue the command specified
566 though system(3).
567
568ioscheduler=str Attempt to switch the device hosting the file to the specified
569 io scheduler before running.
570
571cpuload=int If the job is a CPU cycle eater, attempt to use the specified
572 percentage of CPU cycles.
573
574cpuchunks=int If the job is a CPU cycle eater, split the load into
6c219763 575 cycles of the given time. In milliseconds.
71bfa161
JA
576
577
5786.0 Interpreting the output
579---------------------------
580
581fio spits out a lot of output. While running, fio will display the
582status of the jobs created. An example of that would be:
583
73c8b082 584Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
71bfa161
JA
585
586The characters inside the square brackets denote the current status of
587each thread. The possible values (in typical life cycle order) are:
588
589Idle Run
590---- ---
591P Thread setup, but not started.
592C Thread created.
593I Thread initialized, waiting.
594 R Running, doing sequential reads.
595 r Running, doing random reads.
596 W Running, doing sequential writes.
597 w Running, doing random writes.
598 M Running, doing mixed sequential reads/writes.
599 m Running, doing mixed random reads/writes.
600 F Running, currently waiting for fsync()
601V Running, doing verification of written data.
602E Thread exited, not reaped by main thread yet.
603_ Thread reaped.
604
605The other values are fairly self explanatory - number of threads
6043c579
JA
606currently running and doing io, rate of io since last check, and the estimated
607completion percentage and time for the running group. It's impossible to
608estimate runtime of the following groups (if any).
71bfa161
JA
609
610When fio is done (or interrupted by ctrl-c), it will show the data for
611each thread, group of threads, and disks in that order. For each data
612direction, the output looks like:
613
614Client1 (g=0): err= 0:
615 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
6104ddb6
JA
616 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
617 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
618 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
71bfa161 619 cpu : usr=1.49%, sys=0.25%, ctx=7969
71619dc2 620 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
8abdce66
JA
621 lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%,
622 lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0%
71bfa161
JA
623
624The client number is printed, along with the group id and error of that
625thread. Below is the io statistics, here for writes. In the order listed,
626they denote:
627
628io= Number of megabytes io performed
629bw= Average bandwidth rate
630runt= The runtime of that thread
631 slat= Submission latency (avg being the average, dev being the
632 standard deviation). This is the time it took to submit
633 the io. For sync io, the slat is really the completion
634 latency, since queue/complete is one operation there.
635 clat= Completion latency. Same names as slat, this denotes the
636 time from submission to completion of the io pieces. For
637 sync io, clat will usually be equal (or very close) to 0,
638 as the time from submit to complete is basically just
639 CPU time (io has already been done, see slat explanation).
640 bw= Bandwidth. Same names as the xlat stats, but also includes
641 an approximate percentage of total aggregate bandwidth
642 this thread received in this group. This last value is
643 only really useful if the threads in this group are on the
644 same disk, since they are then competing for disk access.
645cpu= CPU usage. User and system time, along with the number
646 of context switches this thread went through.
71619dc2
JA
647IO depths= The distribution of io depths over the job life time. The
648 numbers are divided into powers of 2, so for example the
649 16= entries includes depths up to that value but higher
650 than the previous entry. In other words, it covers the
651 range from 16 to 31.
ec118304
JA
652IO latencies= The distribution of IO completion latencies. This is the
653 time from when IO leaves fio and when it gets completed.
654 The numbers follow the same pattern as the IO depths,
655 meaning that 2=1.6% means that 1.6% of the IO completed
8abdce66
JA
656 within 2 msecs, 20=12.8% means that 12.8% of the IO
657 took more than 10 msecs, but less than (or equal to) 20 msecs.
71bfa161
JA
658
659After each client has been listed, the group statistics are printed. They
660will look like this:
661
662Run status group 0 (all jobs):
663 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
664 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
665
666For each data direction, it prints:
667
668io= Number of megabytes io performed.
669aggrb= Aggregate bandwidth of threads in this group.
670minb= The minimum average bandwidth a thread saw.
671maxb= The maximum average bandwidth a thread saw.
672mint= The smallest runtime of the threads in that group.
673maxt= The longest runtime of the threads in that group.
674
675And finally, the disk statistics are printed. They will look like this:
676
677Disk stats (read/write):
678 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
679
680Each value is printed for both reads and writes, with reads first. The
681numbers denote:
682
683ios= Number of ios performed by all groups.
684merge= Number of merges io the io scheduler.
685ticks= Number of ticks we kept the disk busy.
686io_queue= Total time spent in the disk queue.
687util= The disk utilization. A value of 100% means we kept the disk
688 busy constantly, 50% would be a disk idling half of the time.
689
690
6917.0 Terse output
692----------------
693
694For scripted usage where you typically want to generate tables or graphs
6af019c9 695of the results, fio can output the results in a semicolon separated format.
71bfa161
JA
696The format is one long line of values, such as:
697
6af019c9
JA
698client1;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%
699;0.0%;0.0%;0.0%;0.0%;0.0%
71bfa161
JA
700
701Split up, the format is as follows:
702
703 jobname, groupid, error
704 READ status:
705 KiB IO, bandwidth (KiB/sec), runtime (msec)
706 Submission latency: min, max, mean, deviation
707 Completion latency: min, max, mean, deviation
6c219763 708 Bw: min, max, aggregate percentage of total, mean, deviation
71bfa161
JA
709 WRITE status:
710 KiB IO, bandwidth (KiB/sec), runtime (msec)
711 Submission latency: min, max, mean, deviation
712 Completion latency: min, max, mean, deviation
6c219763 713 Bw: min, max, aggregate percentage of total, mean, deviation
71bfa161 714 CPU usage: user, system, context switches
2270890c
JA
715 IO depths: <=1, 2, 4, 8, 16, 32, >=64
716 IO latencies: <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, >=2000
717 Text description
71bfa161 718