[PATCH] syslet engine: use pread instead lseek+read
[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
111section residing above it. If the first character in a line is a ';', the
112entire line is discarded as a comment.
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
208 port to connect to in the format of =host:port.
71bfa161
JA
209
210rw=str Type of io pattern. Accepted values are:
211
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
218
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.
222
ee738499
JA
223randrepeat=bool For random IO workloads, seed the generator in a predictable
224 way so that results are repeatable across repetitions.
225
71bfa161
JA
226size=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.
232
f90eff5a
JA
233bs=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
787f7e95
JA
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.
a00735e6 242
71bfa161
JA
243bsrange=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
f90eff5a
JA
246 given (also see bs_unaligned). Applies to both reads and
247 writes, however a second range can be given after a comma.
248 See bs=.
a00735e6 249
690adba3
JA
250bs_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.
71bfa161
JA
253
254nrfiles=int Number of files to use for this job. Defaults to 1.
255
256ioengine=str Defines how the job issues io to the file. The following
257 types are defined:
258
259 sync Basic read(2) or write(2) io. lseek(2) is
260 used to position the io location.
261
262 libaio Linux native asynchronous io.
263
264 posixaio glibc posix asynchronous io.
265
266 mmap File is memory mapped and data copied
267 to/from using memcpy(3).
268
269 splice splice(2) is used to transfer the data and
270 vmsplice(2) to transfer data from user
271 space to the kernel.
272
d0ff85df
JA
273 syslet-rw Use the syslet system calls to make
274 regular read/write async.
275
71bfa161 276 sg SCSI generic sg v3 io. May either be
6c219763 277 synchronous using the SG_IO ioctl, or if
71bfa161
JA
278 the target is an sg character device
279 we use read(2) and write(2) for asynchronous
280 io.
281
a94ea28b
JA
282 null Doesn't transfer any data, just pretends
283 to. This is mainly used to exercise fio
284 itself and for debugging/testing purposes.
285
ed92ac0c
JA
286 net Transfer over the network to given host:port.
287 'filename' must be set appropriately to
288 filename=host:port regardless of send
289 or receive, if the latter only the port
290 argument is used.
291
71bfa161
JA
292iodepth=int This defines how many io units to keep in flight against
293 the file. The default is 1 for each file defined in this
294 job, can be overridden with a larger value for higher
295 concurrency.
296
297direct=bool If value is true, use non-buffered io. This is usually
76a43db4
JA
298 O_DIRECT.
299
300buffered=bool If value is true, use buffered io. This is the opposite
301 of the 'direct' option. Defaults to true.
71bfa161
JA
302
303offset=siint Start io at the given offset in the file. The data before
304 the given offset will not be touched. This effectively
305 caps the file size at real_size - offset.
306
307fsync=int If writing to a file, issue a sync of the dirty data
308 for every number of blocks given. For example, if you give
309 32 as a parameter, fio will sync the file for every 32
310 writes issued. If fio is using non-buffered io, we may
311 not sync the file. The exception is the sg io engine, which
6c219763 312 synchronizes the disk cache anyway.
71bfa161
JA
313
314overwrite=bool If writing to a file, setup the file first and do overwrites.
315
316end_fsync=bool If true, fsync file contents when the job exits.
317
6c219763 318rwmixcycle=int Value in milliseconds describing how often to switch between
71bfa161
JA
319 reads and writes for a mixed workload. The default is
320 500 msecs.
321
322rwmixread=int How large a percentage of the mix should be reads.
323
324rwmixwrite=int How large a percentage of the mix should be writes. If both
325 rwmixread and rwmixwrite is given and the values do not add
326 up to 100%, the latter of the two will be used to override
327 the first.
328
bb8895e0
JA
329norandommap Normally fio will cover every block of the file when doing
330 random IO. If this option is given, fio will just get a
331 new random offset without looking at past io history. This
332 means that some blocks may not be read or written, and that
333 some blocks may be read/written more than once. This option
334 is mutually exclusive with verify= for that reason.
335
71bfa161
JA
336nice=int Run the job with the given nice value. See man nice(2).
337
338prio=int Set the io priority value of this job. Linux limits us to
339 a positive value between 0 and 7, with 0 being the highest.
340 See man ionice(1).
341
342prioclass=int Set the io priority class. See man ionice(1).
343
344thinktime=int Stall the job x microseconds after an io has completed before
345 issuing the next. May be used to simulate processing being
9c1f7434
JA
346 done by an application. See thinktime_blocks.
347
348thinktime_blocks
349 Only valid if thinktime is set - control how many blocks
350 to issue, before waiting 'thinktime' usecs. If not set,
351 defaults to 1 which will make fio wait 'thinktime' usecs
352 after every block.
71bfa161
JA
353
354rate=int Cap the bandwidth used by this job to this number of KiB/sec.
355
356ratemin=int Tell fio to do whatever it can to maintain at least this
357 bandwidth.
358
359ratecycle=int Average bandwidth for 'rate' and 'ratemin' over this number
6c219763 360 of milliseconds.
71bfa161
JA
361
362cpumask=int Set the CPU affinity of this job. The parameter given is a
363 bitmask of allowed CPU's the job may run on. See man
364 sched_setaffinity(2).
365
366startdelay=int Start this job the specified number of seconds after fio
367 has started. Only useful if the job file contains several
368 jobs, and you want to delay starting some jobs to a certain
369 time.
370
03b74b3e 371runtime=int Tell fio to terminate processing after the specified number
71bfa161
JA
372 of seconds. It can be quite hard to determine for how long
373 a specified job will run, so this parameter is handy to
374 cap the total runtime to a given time.
375
376invalidate=bool Invalidate the buffer/page cache parts for this file prior
377 to starting io. Defaults to true.
378
379sync=bool Use sync io for buffered writes. For the majority of the
380 io engines, this means using O_SYNC.
381
382mem=str Fio can use various types of memory as the io unit buffer.
383 The allowed values are:
384
385 malloc Use memory from malloc(3) as the buffers.
386
387 shm Use shared memory as the buffers. Allocated
388 through shmget(2).
389
74b025b0
JA
390 shmhuge Same as shm, but use huge pages as backing.
391
313cb206
JA
392 mmap Use mmap to allocate buffers. May either be
393 anonymous memory, or can be file backed if
394 a filename is given after the option. The
395 format is mem=mmap:/path/to/file.
71bfa161 396
d0bdaf49
JA
397 mmaphuge Use a memory mapped huge file as the buffer
398 backing. Append filename after mmaphuge, ala
399 mem=mmaphuge:/hugetlbfs/file
400
71bfa161 401 The area allocated is a function of the maximum allowed
5394ae5f
JA
402 bs size for the job, multiplied by the io depth given. Note
403 that for shmhuge and mmaphuge to work, the system must have
404 free huge pages allocated. This can normally be checked
405 and set by reading/writing /proc/sys/vm/nr_hugepages on a
406 Linux system. Fio assumes a huge page is 4MiB in size. So
407 to calculate the number of huge pages you need for a given
408 job file, add up the io depth of all jobs (normally one unless
409 iodepth= is used) and multiply by the maximum bs set. Then
410 divide that number by the huge page size. You can see the
411 size of the huge pages in /proc/meminfo. If no huge pages
412 are allocated by having a non-zero number in nr_hugepages,
56bb17f2 413 using mmaphuge or shmhuge will fail. Also see hugepage-size.
5394ae5f
JA
414
415 mmaphuge also needs to have hugetlbfs mounted and the file
416 location should point there. So if it's mounted in /huge,
417 you would use mem=mmaphuge:/huge/somefile.
71bfa161 418
56bb17f2
JA
419hugepage-size=siint
420 Defines the size of a huge page. Must at least be equal
421 to the system setting, see /proc/meminfo. Defaults to 4MiB.
c51074e7
JA
422 Should probably always be a multiple of megabytes, so using
423 hugepage-size=Xm is the preferred way to set this to avoid
424 setting a non-pow-2 bad value.
56bb17f2 425
71bfa161
JA
426exitall When one job finishes, terminate the rest. The default is
427 to wait for each job to finish, sometimes that is not the
428 desired action.
429
430bwavgtime=int Average the calculated bandwidth over the given time. Value
6c219763 431 is specified in milliseconds.
71bfa161
JA
432
433create_serialize=bool If true, serialize the file creating for the jobs.
434 This may be handy to avoid interleaving of data
435 files, which may greatly depend on the filesystem
436 used and even the number of processors in the system.
437
438create_fsync=bool fsync the data file after creation. This is the
439 default.
440
e545a6ce
JA
441unlink=bool Unlink the job files when done. Not the default, as repeated
442 runs of that job would then waste time recreating the fileset
443 again and again.
71bfa161
JA
444
445loops=int Run the specified number of iterations of this job. Used
446 to repeat the same workload a given number of times. Defaults
447 to 1.
448
449verify=str If writing to a file, fio can verify the file contents
450 after each iteration of the job. The allowed values are:
451
452 md5 Use an md5 sum of the data area and store
453 it in the header of each block.
454
455 crc32 Use a crc32 sum of the data area and store
456 it in the header of each block.
457
6c219763 458 This option can be used for repeated burn-in tests of a
71bfa161
JA
459 system to make sure that the written data is also
460 correctly read back.
461
462stonewall Wait for preceeding jobs in the job file to exit, before
463 starting this one. Can be used to insert serialization
464 points in the job file.
465
466numjobs=int Create the specified number of clones of this job. May be
467 used to setup a larger number of threads/processes doing
468 the same thing.
469
470thread fio defaults to forking jobs, however if this option is
471 given, fio will use pthread_create(3) to create threads
472 instead.
473
474zonesize=siint Divide a file into zones of the specified size. See zoneskip.
475
476zoneskip=siint Skip the specified number of bytes when zonesize data has
477 been read. The two zone options can be used to only do
478 io on zones of a file.
479
076efc7c
JA
480write_iolog=str Write the issued io patterns to the specified file. See
481 read_iolog.
71bfa161 482
076efc7c 483read_iolog=str Open an iolog with the specified file name and replay the
71bfa161
JA
484 io patterns it contains. This can be used to store a
485 workload and replay it sometime later.
486
487write_bw_log If given, write a bandwidth log of the jobs in this job
488 file. Can be used to store data of the bandwidth of the
e0da9bc2
JA
489 jobs in their lifetime. The included fio_generate_plots
490 script uses gnuplot to turn these text files into nice
491 graphs.
71bfa161
JA
492
493write_lat_log Same as write_bw_log, except that this option stores io
494 completion latencies instead.
495
496lockmem=siint Pin down the specified amount of memory with mlock(2). Can
497 potentially be used instead of removing memory or booting
498 with less memory to simulate a smaller amount of memory.
499
500exec_prerun=str Before running this job, issue the command specified
501 through system(3).
502
503exec_postrun=str After the job completes, issue the command specified
504 though system(3).
505
506ioscheduler=str Attempt to switch the device hosting the file to the specified
507 io scheduler before running.
508
509cpuload=int If the job is a CPU cycle eater, attempt to use the specified
510 percentage of CPU cycles.
511
512cpuchunks=int If the job is a CPU cycle eater, split the load into
6c219763 513 cycles of the given time. In milliseconds.
71bfa161
JA
514
515
5166.0 Interpreting the output
517---------------------------
518
519fio spits out a lot of output. While running, fio will display the
520status of the jobs created. An example of that would be:
521
73c8b082 522Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s]
71bfa161
JA
523
524The characters inside the square brackets denote the current status of
525each thread. The possible values (in typical life cycle order) are:
526
527Idle Run
528---- ---
529P Thread setup, but not started.
530C Thread created.
531I Thread initialized, waiting.
532 R Running, doing sequential reads.
533 r Running, doing random reads.
534 W Running, doing sequential writes.
535 w Running, doing random writes.
536 M Running, doing mixed sequential reads/writes.
537 m Running, doing mixed random reads/writes.
538 F Running, currently waiting for fsync()
539V Running, doing verification of written data.
540E Thread exited, not reaped by main thread yet.
541_ Thread reaped.
542
543The other values are fairly self explanatory - number of threads
6043c579
JA
544currently running and doing io, rate of io since last check, and the estimated
545completion percentage and time for the running group. It's impossible to
546estimate runtime of the following groups (if any).
71bfa161
JA
547
548When fio is done (or interrupted by ctrl-c), it will show the data for
549each thread, group of threads, and disks in that order. For each data
550direction, the output looks like:
551
552Client1 (g=0): err= 0:
553 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
6104ddb6
JA
554 slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92
555 clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82
556 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68
71bfa161 557 cpu : usr=1.49%, sys=0.25%, ctx=7969
71619dc2 558 IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0%
71bfa161
JA
559
560The client number is printed, along with the group id and error of that
561thread. Below is the io statistics, here for writes. In the order listed,
562they denote:
563
564io= Number of megabytes io performed
565bw= Average bandwidth rate
566runt= The runtime of that thread
567 slat= Submission latency (avg being the average, dev being the
568 standard deviation). This is the time it took to submit
569 the io. For sync io, the slat is really the completion
570 latency, since queue/complete is one operation there.
571 clat= Completion latency. Same names as slat, this denotes the
572 time from submission to completion of the io pieces. For
573 sync io, clat will usually be equal (or very close) to 0,
574 as the time from submit to complete is basically just
575 CPU time (io has already been done, see slat explanation).
576 bw= Bandwidth. Same names as the xlat stats, but also includes
577 an approximate percentage of total aggregate bandwidth
578 this thread received in this group. This last value is
579 only really useful if the threads in this group are on the
580 same disk, since they are then competing for disk access.
581cpu= CPU usage. User and system time, along with the number
582 of context switches this thread went through.
71619dc2
JA
583IO depths= The distribution of io depths over the job life time. The
584 numbers are divided into powers of 2, so for example the
585 16= entries includes depths up to that value but higher
586 than the previous entry. In other words, it covers the
587 range from 16 to 31.
71bfa161
JA
588
589After each client has been listed, the group statistics are printed. They
590will look like this:
591
592Run status group 0 (all jobs):
593 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
594 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
595
596For each data direction, it prints:
597
598io= Number of megabytes io performed.
599aggrb= Aggregate bandwidth of threads in this group.
600minb= The minimum average bandwidth a thread saw.
601maxb= The maximum average bandwidth a thread saw.
602mint= The smallest runtime of the threads in that group.
603maxt= The longest runtime of the threads in that group.
604
605And finally, the disk statistics are printed. They will look like this:
606
607Disk stats (read/write):
608 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
609
610Each value is printed for both reads and writes, with reads first. The
611numbers denote:
612
613ios= Number of ios performed by all groups.
614merge= Number of merges io the io scheduler.
615ticks= Number of ticks we kept the disk busy.
616io_queue= Total time spent in the disk queue.
617util= The disk utilization. A value of 100% means we kept the disk
618 busy constantly, 50% would be a disk idling half of the time.
619
620
6217.0 Terse output
622----------------
623
624For scripted usage where you typically want to generate tables or graphs
6c219763 625of the results, fio can output the results in a comma separated format.
71bfa161
JA
626The format is one long line of values, such as:
627
628client1,0,0,936,331,2894,0,0,0.000000,0.000000,1,170,22.115385,34.290410,16,714,84.252874%,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
629
630Split up, the format is as follows:
631
632 jobname, groupid, error
633 READ status:
634 KiB IO, bandwidth (KiB/sec), runtime (msec)
635 Submission latency: min, max, mean, deviation
636 Completion latency: min, max, mean, deviation
6c219763 637 Bw: min, max, aggregate percentage of total, mean, deviation
71bfa161
JA
638 WRITE status:
639 KiB IO, bandwidth (KiB/sec), runtime (msec)
640 Submission latency: min, max, mean, deviation
641 Completion latency: min, max, mean, deviation
6c219763 642 Bw: min, max, aggregate percentage of total, mean, deviation
71bfa161
JA
643 CPU usage: user, system, context switches
644