Commit | Line | Data |
---|---|---|
f80dba8d MT |
1 | How fio works |
2 | ------------- | |
3 | ||
4 | The first step in getting fio to simulate a desired I/O workload, is writing a | |
5 | job file describing that specific setup. A job file may contain any number of | |
6 | threads and/or files -- the typical contents of the job file is a *global* | |
7 | section defining shared parameters, and one or more job sections describing the | |
8 | jobs involved. When run, fio parses this file and sets everything up as | |
9 | described. If we break down a job from top to bottom, it contains the following | |
10 | basic parameters: | |
11 | ||
12 | `I/O type`_ | |
13 | ||
14 | Defines the I/O pattern issued to the file(s). We may only be reading | |
15 | sequentially from this file(s), or we may be writing randomly. Or even | |
16 | mixing reads and writes, sequentially or randomly. | |
17 | Should we be doing buffered I/O, or direct/raw I/O? | |
18 | ||
19 | `Block size`_ | |
20 | ||
21 | In how large chunks are we issuing I/O? This may be a single value, | |
22 | or it may describe a range of block sizes. | |
23 | ||
24 | `I/O size`_ | |
25 | ||
26 | How much data are we going to be reading/writing. | |
27 | ||
28 | `I/O engine`_ | |
29 | ||
30 | How do we issue I/O? We could be memory mapping the file, we could be | |
31 | using regular read/write, we could be using splice, async I/O, or even | |
32 | SG (SCSI generic sg). | |
33 | ||
34 | `I/O depth`_ | |
35 | ||
36 | If the I/O engine is async, how large a queuing depth do we want to | |
37 | maintain? | |
38 | ||
39 | ||
40 | `Target file/device`_ | |
41 | ||
42 | How many files are we spreading the workload over. | |
43 | ||
44 | `Threads, processes and job synchronization`_ | |
45 | ||
46 | How many threads or processes should we spread this workload over. | |
47 | ||
48 | The above are the basic parameters defined for a workload, in addition there's a | |
49 | multitude of parameters that modify other aspects of how this job behaves. | |
50 | ||
51 | ||
52 | Command line options | |
53 | -------------------- | |
54 | ||
55 | .. option:: --debug=type | |
56 | ||
57 | Enable verbose tracing of various fio actions. May be ``all`` for all types | |
c60ebc45 | 58 | or individual types separated by a comma (e.g. ``--debug=file,mem`` will |
f80dba8d MT |
59 | enable file and memory debugging). Currently, additional logging is |
60 | available for: | |
61 | ||
62 | *process* | |
63 | Dump info related to processes. | |
64 | *file* | |
65 | Dump info related to file actions. | |
66 | *io* | |
67 | Dump info related to I/O queuing. | |
68 | *mem* | |
69 | Dump info related to memory allocations. | |
70 | *blktrace* | |
71 | Dump info related to blktrace setup. | |
72 | *verify* | |
73 | Dump info related to I/O verification. | |
74 | *all* | |
75 | Enable all debug options. | |
76 | *random* | |
77 | Dump info related to random offset generation. | |
78 | *parse* | |
79 | Dump info related to option matching and parsing. | |
80 | *diskutil* | |
81 | Dump info related to disk utilization updates. | |
82 | *job:x* | |
83 | Dump info only related to job number x. | |
84 | *mutex* | |
85 | Dump info only related to mutex up/down ops. | |
86 | *profile* | |
87 | Dump info related to profile extensions. | |
88 | *time* | |
89 | Dump info related to internal time keeping. | |
90 | *net* | |
91 | Dump info related to networking connections. | |
92 | *rate* | |
93 | Dump info related to I/O rate switching. | |
94 | *compress* | |
95 | Dump info related to log compress/decompress. | |
96 | *?* or *help* | |
97 | Show available debug options. | |
98 | ||
99 | .. option:: --parse-only | |
100 | ||
101 | Parse options only, don\'t start any I/O. | |
102 | ||
103 | .. option:: --output=filename | |
104 | ||
105 | Write output to file `filename`. | |
106 | ||
107 | .. option:: --bandwidth-log | |
108 | ||
109 | Generate aggregate bandwidth logs. | |
110 | ||
111 | .. option:: --minimal | |
112 | ||
113 | Print statistics in a terse, semicolon-delimited format. | |
114 | ||
115 | .. option:: --append-terse | |
116 | ||
117 | Print statistics in selected mode AND terse, semicolon-delimited format. | |
118 | **deprecated**, use :option:`--output-format` instead to select multiple | |
119 | formats. | |
120 | ||
121 | .. option:: --output-format=type | |
122 | ||
123 | Set the reporting format to `normal`, `terse`, `json`, or `json+`. Multiple | |
124 | formats can be selected, separate by a comma. `terse` is a CSV based | |
125 | format. `json+` is like `json`, except it adds a full dump of the latency | |
126 | buckets. | |
127 | ||
128 | .. option:: --terse-version=type | |
129 | ||
130 | Set terse version output format (default 3, or 2 or 4). | |
131 | ||
132 | .. option:: --version | |
133 | ||
134 | Print version info and exit. | |
135 | ||
136 | .. option:: --help | |
137 | ||
138 | Print this page. | |
139 | ||
140 | .. option:: --cpuclock-test | |
141 | ||
142 | Perform test and validation of internal CPU clock. | |
143 | ||
144 | .. option:: --crctest=test | |
145 | ||
146 | Test the speed of the builtin checksumming functions. If no argument is | |
147 | given, all of them are tested. Or a comma separated list can be passed, in | |
148 | which case the given ones are tested. | |
149 | ||
150 | .. option:: --cmdhelp=command | |
151 | ||
152 | Print help information for `command`. May be ``all`` for all commands. | |
153 | ||
154 | .. option:: --enghelp=[ioengine[,command]] | |
155 | ||
156 | List all commands defined by :option:`ioengine`, or print help for `command` | |
157 | defined by :option:`ioengine`. If no :option:`ioengine` is given, list all | |
158 | available ioengines. | |
159 | ||
160 | .. option:: --showcmd=jobfile | |
161 | ||
162 | Turn a job file into command line options. | |
163 | ||
164 | .. option:: --readonly | |
165 | ||
166 | Turn on safety read-only checks, preventing writes. The ``--readonly`` | |
167 | option is an extra safety guard to prevent users from accidentally starting | |
168 | a write workload when that is not desired. Fio will only write if | |
169 | `rw=write/randwrite/rw/randrw` is given. This extra safety net can be used | |
170 | as an extra precaution as ``--readonly`` will also enable a write check in | |
171 | the I/O engine core to prevent writes due to unknown user space bug(s). | |
172 | ||
173 | .. option:: --eta=when | |
174 | ||
175 | When real-time ETA estimate should be printed. May be `always`, `never` or | |
176 | `auto`. | |
177 | ||
178 | .. option:: --eta-newline=time | |
179 | ||
180 | Force a new line for every `time` period passed. | |
181 | ||
182 | .. option:: --status-interval=time | |
183 | ||
184 | Force full status dump every `time` period passed. | |
185 | ||
186 | .. option:: --section=name | |
187 | ||
188 | Only run specified section in job file. Multiple sections can be specified. | |
189 | The ``--section`` option allows one to combine related jobs into one file. | |
190 | E.g. one job file could define light, moderate, and heavy sections. Tell | |
191 | fio to run only the "heavy" section by giving ``--section=heavy`` | |
192 | command line option. One can also specify the "write" operations in one | |
193 | section and "verify" operation in another section. The ``--section`` option | |
194 | only applies to job sections. The reserved *global* section is always | |
195 | parsed and used. | |
196 | ||
197 | .. option:: --alloc-size=kb | |
198 | ||
199 | Set the internal smalloc pool to this size in kb (def 1024). The | |
200 | ``--alloc-size`` switch allows one to use a larger pool size for smalloc. | |
201 | If running large jobs with randommap enabled, fio can run out of memory. | |
202 | Smalloc is an internal allocator for shared structures from a fixed size | |
203 | memory pool. The pool size defaults to 16M and can grow to 8 pools. | |
204 | ||
205 | NOTE: While running :file:`.fio_smalloc.*` backing store files are visible | |
206 | in :file:`/tmp`. | |
207 | ||
208 | .. option:: --warnings-fatal | |
209 | ||
210 | All fio parser warnings are fatal, causing fio to exit with an | |
211 | error. | |
212 | ||
213 | .. option:: --max-jobs=nr | |
214 | ||
215 | Maximum number of threads/processes to support. | |
216 | ||
217 | .. option:: --server=args | |
218 | ||
219 | Start a backend server, with `args` specifying what to listen to. | |
220 | See `Client/Server`_ section. | |
221 | ||
222 | .. option:: --daemonize=pidfile | |
223 | ||
224 | Background a fio server, writing the pid to the given `pidfile` file. | |
225 | ||
226 | .. option:: --client=hostname | |
227 | ||
228 | Instead of running the jobs locally, send and run them on the given host or | |
229 | set of hosts. See `Client/Server`_ section. | |
230 | ||
231 | .. option:: --remote-config=file | |
232 | ||
233 | Tell fio server to load this local file. | |
234 | ||
235 | .. option:: --idle-prof=option | |
236 | ||
237 | Report cpu idleness on a system or percpu basis | |
238 | ``--idle-prof=system,percpu`` or | |
239 | run unit work calibration only ``--idle-prof=calibrate``. | |
240 | ||
241 | .. option:: --inflate-log=log | |
242 | ||
243 | Inflate and output compressed log. | |
244 | ||
245 | .. option:: --trigger-file=file | |
246 | ||
247 | Execute trigger cmd when file exists. | |
248 | ||
249 | .. option:: --trigger-timeout=t | |
250 | ||
251 | Execute trigger at this time. | |
252 | ||
253 | .. option:: --trigger=cmd | |
254 | ||
255 | Set this command as local trigger. | |
256 | ||
257 | .. option:: --trigger-remote=cmd | |
258 | ||
259 | Set this command as remote trigger. | |
260 | ||
261 | .. option:: --aux-path=path | |
262 | ||
263 | Use this path for fio state generated files. | |
264 | ||
265 | Any parameters following the options will be assumed to be job files, unless | |
266 | they match a job file parameter. Multiple job files can be listed and each job | |
267 | file will be regarded as a separate group. Fio will :option:`stonewall` | |
268 | execution between each group. | |
269 | ||
270 | ||
271 | Job file format | |
272 | --------------- | |
273 | ||
274 | As previously described, fio accepts one or more job files describing what it is | |
275 | supposed to do. The job file format is the classic ini file, where the names | |
c60ebc45 | 276 | enclosed in [] brackets define the job name. You are free to use any ASCII name |
f80dba8d MT |
277 | you want, except *global* which has special meaning. Following the job name is |
278 | a sequence of zero or more parameters, one per line, that define the behavior of | |
279 | the job. If the first character in a line is a ';' or a '#', the entire line is | |
280 | discarded as a comment. | |
281 | ||
282 | A *global* section sets defaults for the jobs described in that file. A job may | |
283 | override a *global* section parameter, and a job file may even have several | |
284 | *global* sections if so desired. A job is only affected by a *global* section | |
285 | residing above it. | |
286 | ||
287 | The :option:`--cmdhelp` option also lists all options. If used with an `option` | |
288 | argument, :option:`--cmdhelp` will detail the given `option`. | |
289 | ||
290 | See the `examples/` directory for inspiration on how to write job files. Note | |
291 | the copyright and license requirements currently apply to `examples/` files. | |
292 | ||
293 | So let's look at a really simple job file that defines two processes, each | |
294 | randomly reading from a 128MiB file: | |
295 | ||
296 | .. code-block:: ini | |
297 | ||
298 | ; -- start job file -- | |
299 | [global] | |
300 | rw=randread | |
301 | size=128m | |
302 | ||
303 | [job1] | |
304 | ||
305 | [job2] | |
306 | ||
307 | ; -- end job file -- | |
308 | ||
309 | As you can see, the job file sections themselves are empty as all the described | |
310 | parameters are shared. As no :option:`filename` option is given, fio makes up a | |
311 | `filename` for each of the jobs as it sees fit. On the command line, this job | |
312 | would look as follows:: | |
313 | ||
314 | $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2 | |
315 | ||
316 | ||
317 | Let's look at an example that has a number of processes writing randomly to | |
318 | files: | |
319 | ||
320 | .. code-block:: ini | |
321 | ||
322 | ; -- start job file -- | |
323 | [random-writers] | |
324 | ioengine=libaio | |
325 | iodepth=4 | |
326 | rw=randwrite | |
327 | bs=32k | |
328 | direct=0 | |
329 | size=64m | |
330 | numjobs=4 | |
331 | ; -- end job file -- | |
332 | ||
333 | Here we have no *global* section, as we only have one job defined anyway. We | |
334 | want to use async I/O here, with a depth of 4 for each file. We also increased | |
335 | the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical | |
336 | jobs. The result is 4 processes each randomly writing to their own 64MiB | |
337 | file. Instead of using the above job file, you could have given the parameters | |
338 | on the command line. For this case, you would specify:: | |
339 | ||
340 | $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4 | |
341 | ||
342 | When fio is utilized as a basis of any reasonably large test suite, it might be | |
343 | desirable to share a set of standardized settings across multiple job files. | |
344 | Instead of copy/pasting such settings, any section may pull in an external | |
345 | :file:`filename.fio` file with *include filename* directive, as in the following | |
346 | example:: | |
347 | ||
348 | ; -- start job file including.fio -- | |
349 | [global] | |
350 | filename=/tmp/test | |
351 | filesize=1m | |
352 | include glob-include.fio | |
353 | ||
354 | [test] | |
355 | rw=randread | |
356 | bs=4k | |
357 | time_based=1 | |
358 | runtime=10 | |
359 | include test-include.fio | |
360 | ; -- end job file including.fio -- | |
361 | ||
362 | .. code-block:: ini | |
363 | ||
364 | ; -- start job file glob-include.fio -- | |
365 | thread=1 | |
366 | group_reporting=1 | |
367 | ; -- end job file glob-include.fio -- | |
368 | ||
369 | .. code-block:: ini | |
370 | ||
371 | ; -- start job file test-include.fio -- | |
372 | ioengine=libaio | |
373 | iodepth=4 | |
374 | ; -- end job file test-include.fio -- | |
375 | ||
376 | Settings pulled into a section apply to that section only (except *global* | |
377 | section). Include directives may be nested in that any included file may contain | |
378 | further include directive(s). Include files may not contain [] sections. | |
379 | ||
380 | ||
381 | Environment variables | |
382 | ~~~~~~~~~~~~~~~~~~~~~ | |
383 | ||
384 | Fio also supports environment variable expansion in job files. Any sub-string of | |
385 | the form ``${VARNAME}`` as part of an option value (in other words, on the right | |
386 | of the '='), will be expanded to the value of the environment variable called | |
387 | `VARNAME`. If no such environment variable is defined, or `VARNAME` is the | |
388 | empty string, the empty string will be substituted. | |
389 | ||
390 | As an example, let's look at a sample fio invocation and job file:: | |
391 | ||
392 | $ SIZE=64m NUMJOBS=4 fio jobfile.fio | |
393 | ||
394 | .. code-block:: ini | |
395 | ||
396 | ; -- start job file -- | |
397 | [random-writers] | |
398 | rw=randwrite | |
399 | size=${SIZE} | |
400 | numjobs=${NUMJOBS} | |
401 | ; -- end job file -- | |
402 | ||
403 | This will expand to the following equivalent job file at runtime: | |
404 | ||
405 | .. code-block:: ini | |
406 | ||
407 | ; -- start job file -- | |
408 | [random-writers] | |
409 | rw=randwrite | |
410 | size=64m | |
411 | numjobs=4 | |
412 | ; -- end job file -- | |
413 | ||
414 | Fio ships with a few example job files, you can also look there for inspiration. | |
415 | ||
416 | Reserved keywords | |
417 | ~~~~~~~~~~~~~~~~~ | |
418 | ||
419 | Additionally, fio has a set of reserved keywords that will be replaced | |
420 | internally with the appropriate value. Those keywords are: | |
421 | ||
422 | **$pagesize** | |
423 | ||
424 | The architecture page size of the running system. | |
425 | ||
426 | **$mb_memory** | |
427 | ||
428 | Megabytes of total memory in the system. | |
429 | ||
430 | **$ncpus** | |
431 | ||
432 | Number of online available CPUs. | |
433 | ||
434 | These can be used on the command line or in the job file, and will be | |
435 | automatically substituted with the current system values when the job is | |
436 | run. Simple math is also supported on these keywords, so you can perform actions | |
437 | like:: | |
438 | ||
439 | size=8*$mb_memory | |
440 | ||
441 | and get that properly expanded to 8 times the size of memory in the machine. | |
442 | ||
443 | ||
444 | Job file parameters | |
445 | ------------------- | |
446 | ||
447 | This section describes in details each parameter associated with a job. Some | |
448 | parameters take an option of a given type, such as an integer or a | |
449 | string. Anywhere a numeric value is required, an arithmetic expression may be | |
450 | used, provided it is surrounded by parentheses. Supported operators are: | |
451 | ||
452 | - addition (+) | |
453 | - subtraction (-) | |
454 | - multiplication (*) | |
455 | - division (/) | |
456 | - modulus (%) | |
457 | - exponentiation (^) | |
458 | ||
459 | For time values in expressions, units are microseconds by default. This is | |
460 | different than for time values not in expressions (not enclosed in | |
461 | parentheses). The following types are used: | |
462 | ||
463 | ||
464 | Parameter types | |
465 | ~~~~~~~~~~~~~~~ | |
466 | ||
467 | **str** | |
468 | String. This is a sequence of alpha characters. | |
469 | ||
470 | **time** | |
471 | Integer with possible time suffix. In seconds unless otherwise | |
c60ebc45 | 472 | specified, use e.g. 10m for 10 minutes. Accepts s/m/h for seconds, minutes, |
f80dba8d MT |
473 | and hours, and accepts 'ms' (or 'msec') for milliseconds, and 'us' (or |
474 | 'usec') for microseconds. | |
475 | ||
476 | .. _int: | |
477 | ||
478 | **int** | |
479 | Integer. A whole number value, which may contain an integer prefix | |
480 | and an integer suffix: | |
481 | ||
482 | [*integer prefix*] **number** [*integer suffix*] | |
483 | ||
484 | The optional *integer prefix* specifies the number's base. The default | |
485 | is decimal. *0x* specifies hexadecimal. | |
486 | ||
487 | The optional *integer suffix* specifies the number's units, and includes an | |
488 | optional unit prefix and an optional unit. For quantities of data, the | |
489 | default unit is bytes. For quantities of time, the default unit is seconds. | |
490 | ||
491 | With :option:`kb_base` =1000, fio follows international standards for unit | |
492 | prefixes. To specify power-of-10 decimal values defined in the | |
493 | International System of Units (SI): | |
494 | ||
495 | * *Ki* -- means kilo (K) or 1000 | |
496 | * *Mi* -- means mega (M) or 1000**2 | |
497 | * *Gi* -- means giga (G) or 1000**3 | |
498 | * *Ti* -- means tera (T) or 1000**4 | |
499 | * *Pi* -- means peta (P) or 1000**5 | |
500 | ||
501 | To specify power-of-2 binary values defined in IEC 80000-13: | |
502 | ||
503 | * *k* -- means kibi (Ki) or 1024 | |
504 | * *M* -- means mebi (Mi) or 1024**2 | |
505 | * *G* -- means gibi (Gi) or 1024**3 | |
506 | * *T* -- means tebi (Ti) or 1024**4 | |
507 | * *P* -- means pebi (Pi) or 1024**5 | |
508 | ||
509 | With :option:`kb_base` =1024 (the default), the unit prefixes are opposite | |
510 | from those specified in the SI and IEC 80000-13 standards to provide | |
511 | compatibility with old scripts. For example, 4k means 4096. | |
512 | ||
513 | For quantities of data, an optional unit of 'B' may be included | |
514 | (e.g., 'kB' is the same as 'k'). | |
515 | ||
516 | The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega, | |
517 | not milli). 'b' and 'B' both mean byte, not bit. | |
518 | ||
519 | Examples with :option:`kb_base` =1000: | |
520 | ||
521 | * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB | |
522 | * *1 MiB*: 1048576, 1mi, 1024ki | |
523 | * *1 MB*: 1000000, 1m, 1000k | |
524 | * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi | |
525 | * *1 TB*: 1000000000, 1t, 1000m, 1000000k | |
526 | ||
527 | Examples with :option:`kb_base` =1024 (default): | |
528 | ||
529 | * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB | |
530 | * *1 MiB*: 1048576, 1m, 1024k | |
531 | * *1 MB*: 1000000, 1mi, 1000ki | |
532 | * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m | |
533 | * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki | |
534 | ||
535 | To specify times (units are not case sensitive): | |
536 | ||
537 | * *D* -- means days | |
538 | * *H* -- means hours | |
539 | * *M* -- mean minutes | |
540 | * *s* -- or sec means seconds (default) | |
541 | * *ms* -- or *msec* means milliseconds | |
542 | * *us* -- or *usec* means microseconds | |
543 | ||
544 | If the option accepts an upper and lower range, use a colon ':' or | |
545 | minus '-' to separate such values. See :ref:`irange <irange>`. | |
dab41128 TK |
546 | If the lower value specified happens to be larger than the upper value, |
547 | two values are swapped. | |
f80dba8d MT |
548 | |
549 | .. _bool: | |
550 | ||
551 | **bool** | |
552 | Boolean. Usually parsed as an integer, however only defined for | |
553 | true and false (1 and 0). | |
554 | ||
555 | .. _irange: | |
556 | ||
557 | **irange** | |
558 | Integer range with suffix. Allows value range to be given, such as | |
c60ebc45 | 559 | 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the |
f80dba8d MT |
560 | option allows two sets of ranges, they can be specified with a ',' or '/' |
561 | delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`. | |
562 | ||
563 | **float_list** | |
564 | A list of floating point numbers, separated by a ':' character. | |
565 | ||
566 | ||
567 | Units | |
568 | ~~~~~ | |
569 | ||
570 | .. option:: kb_base=int | |
571 | ||
572 | Select the interpretation of unit prefixes in input parameters. | |
573 | ||
574 | **1000** | |
575 | Inputs comply with IEC 80000-13 and the International | |
576 | System of Units (SI). Use: | |
577 | ||
578 | - power-of-2 values with IEC prefixes (e.g., KiB) | |
579 | - power-of-10 values with SI prefixes (e.g., kB) | |
580 | ||
581 | **1024** | |
582 | Compatibility mode (default). To avoid breaking old scripts: | |
583 | ||
584 | - power-of-2 values with SI prefixes | |
585 | - power-of-10 values with IEC prefixes | |
586 | ||
587 | See :option:`bs` for more details on input parameters. | |
588 | ||
589 | Outputs always use correct prefixes. Most outputs include both | |
590 | side-by-side, like:: | |
591 | ||
592 | bw=2383.3kB/s (2327.4KiB/s) | |
593 | ||
594 | If only one value is reported, then kb_base selects the one to use: | |
595 | ||
596 | **1000** -- SI prefixes | |
597 | ||
598 | **1024** -- IEC prefixes | |
599 | ||
600 | .. option:: unit_base=int | |
601 | ||
602 | Base unit for reporting. Allowed values are: | |
603 | ||
604 | **0** | |
605 | Use auto-detection (default). | |
606 | **8** | |
607 | Byte based. | |
608 | **1** | |
609 | Bit based. | |
610 | ||
611 | ||
612 | With the above in mind, here follows the complete list of fio job parameters. | |
613 | ||
614 | ||
615 | Job description | |
616 | ~~~~~~~~~~~~~~~ | |
617 | ||
618 | .. option:: name=str | |
619 | ||
620 | ASCII name of the job. This may be used to override the name printed by fio | |
621 | for this job. Otherwise the job name is used. On the command line this | |
622 | parameter has the special purpose of also signaling the start of a new job. | |
623 | ||
624 | .. option:: description=str | |
625 | ||
626 | Text description of the job. Doesn't do anything except dump this text | |
627 | description when this job is run. It's not parsed. | |
628 | ||
629 | .. option:: loops=int | |
630 | ||
631 | Run the specified number of iterations of this job. Used to repeat the same | |
632 | workload a given number of times. Defaults to 1. | |
633 | ||
634 | .. option:: numjobs=int | |
635 | ||
79591fa9 TK |
636 | Create the specified number of clones of this job. Each clone of job |
637 | is spawned as an independent thread or process. May be used to setup a | |
f80dba8d MT |
638 | larger number of threads/processes doing the same thing. Each thread is |
639 | reported separately; to see statistics for all clones as a whole, use | |
640 | :option:`group_reporting` in conjunction with :option:`new_group`. | |
641 | See :option:`--max-jobs`. | |
642 | ||
643 | ||
644 | Time related parameters | |
645 | ~~~~~~~~~~~~~~~~~~~~~~~ | |
646 | ||
647 | .. option:: runtime=time | |
648 | ||
f75ede1d | 649 | Tell fio to terminate processing after the specified period of time. It |
f80dba8d | 650 | can be quite hard to determine for how long a specified job will run, so |
f75ede1d SW |
651 | this parameter is handy to cap the total runtime to a given time. When |
652 | the unit is omitted, the value is given in seconds. | |
f80dba8d MT |
653 | |
654 | .. option:: time_based | |
655 | ||
656 | If set, fio will run for the duration of the :option:`runtime` specified | |
657 | even if the file(s) are completely read or written. It will simply loop over | |
658 | the same workload as many times as the :option:`runtime` allows. | |
659 | ||
a881438b | 660 | .. option:: startdelay=irange(time) |
f80dba8d MT |
661 | |
662 | Delay start of job for the specified number of seconds. Supports all time | |
663 | suffixes to allow specification of hours, minutes, seconds and milliseconds | |
664 | -- seconds are the default if a unit is omitted. Can be given as a range | |
665 | which causes each thread to choose randomly out of the range. | |
666 | ||
667 | .. option:: ramp_time=time | |
668 | ||
669 | If set, fio will run the specified workload for this amount of time before | |
670 | logging any performance numbers. Useful for letting performance settle | |
671 | before logging results, thus minimizing the runtime required for stable | |
672 | results. Note that the ``ramp_time`` is considered lead in time for a job, | |
673 | thus it will increase the total runtime if a special timeout or | |
f75ede1d SW |
674 | :option:`runtime` is specified. When the unit is omitted, the value is |
675 | given in seconds. | |
f80dba8d MT |
676 | |
677 | .. option:: clocksource=str | |
678 | ||
679 | Use the given clocksource as the base of timing. The supported options are: | |
680 | ||
681 | **gettimeofday** | |
682 | :manpage:`gettimeofday(2)` | |
683 | ||
684 | **clock_gettime** | |
685 | :manpage:`clock_gettime(2)` | |
686 | ||
687 | **cpu** | |
688 | Internal CPU clock source | |
689 | ||
690 | cpu is the preferred clocksource if it is reliable, as it is very fast (and | |
691 | fio is heavy on time calls). Fio will automatically use this clocksource if | |
692 | it's supported and considered reliable on the system it is running on, | |
693 | unless another clocksource is specifically set. For x86/x86-64 CPUs, this | |
694 | means supporting TSC Invariant. | |
695 | ||
696 | .. option:: gtod_reduce=bool | |
697 | ||
698 | Enable all of the :manpage:`gettimeofday(2)` reducing options | |
f75ede1d | 699 | (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus |
f80dba8d MT |
700 | reduce precision of the timeout somewhat to really shrink the |
701 | :manpage:`gettimeofday(2)` call count. With this option enabled, we only do | |
702 | about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all | |
703 | time keeping was enabled. | |
704 | ||
705 | .. option:: gtod_cpu=int | |
706 | ||
707 | Sometimes it's cheaper to dedicate a single thread of execution to just | |
708 | getting the current time. Fio (and databases, for instance) are very | |
709 | intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set | |
710 | one CPU aside for doing nothing but logging current time to a shared memory | |
711 | location. Then the other threads/processes that run I/O workloads need only | |
712 | copy that segment, instead of entering the kernel with a | |
713 | :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time | |
714 | calls will be excluded from other uses. Fio will manually clear it from the | |
715 | CPU mask of other jobs. | |
716 | ||
717 | ||
718 | Target file/device | |
719 | ~~~~~~~~~~~~~~~~~~ | |
720 | ||
721 | .. option:: directory=str | |
722 | ||
723 | Prefix filenames with this directory. Used to place files in a different | |
724 | location than :file:`./`. You can specify a number of directories by | |
725 | separating the names with a ':' character. These directories will be | |
726 | assigned equally distributed to job clones creates with :option:`numjobs` as | |
727 | long as they are using generated filenames. If specific `filename(s)` are | |
728 | set fio will use the first listed directory, and thereby matching the | |
729 | `filename` semantic which generates a file each clone if not specified, but | |
730 | let all clones use the same if set. | |
731 | ||
732 | See the :option:`filename` option for escaping certain characters. | |
733 | ||
734 | .. option:: filename=str | |
735 | ||
736 | Fio normally makes up a `filename` based on the job name, thread number, and | |
737 | file number. If you want to share files between threads in a job or several | |
79591fa9 TK |
738 | jobs with fixed file paths, specify a `filename` for each of them to override |
739 | the default. If the ioengine is file based, you can specify a number of files | |
740 | by separating the names with a ':' colon. So if you wanted a job to open | |
741 | :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use | |
742 | ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is | |
743 | specified, :option:`nrfiles` is ignored. The size of regular files specified | |
744 | by this option will be :option:`size` divided by number of files unless | |
745 | explicit size is specified by :option:`filesize`. | |
746 | ||
f80dba8d MT |
747 | On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for |
748 | the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc. | |
749 | Note: Windows and FreeBSD prevent write access to areas | |
750 | of the disk containing in-use data (e.g. filesystems). If the wanted | |
751 | `filename` does need to include a colon, then escape that with a ``\`` | |
752 | character. For instance, if the `filename` is :file:`/dev/dsk/foo@3,0:c`, | |
753 | then you would use ``filename="/dev/dsk/foo@3,0\:c"``. The | |
754 | :file:`-` is a reserved name, meaning stdin or stdout. Which of the two | |
755 | depends on the read/write direction set. | |
756 | ||
757 | .. option:: filename_format=str | |
758 | ||
759 | If sharing multiple files between jobs, it is usually necessary to have fio | |
760 | generate the exact names that you want. By default, fio will name a file | |
761 | based on the default file format specification of | |
762 | :file:`jobname.jobnumber.filenumber`. With this option, that can be | |
763 | customized. Fio will recognize and replace the following keywords in this | |
764 | string: | |
765 | ||
766 | **$jobname** | |
767 | The name of the worker thread or process. | |
768 | **$jobnum** | |
769 | The incremental number of the worker thread or process. | |
770 | **$filenum** | |
771 | The incremental number of the file for that worker thread or | |
772 | process. | |
773 | ||
774 | To have dependent jobs share a set of files, this option can be set to have | |
775 | fio generate filenames that are shared between the two. For instance, if | |
776 | :file:`testfiles.$filenum` is specified, file number 4 for any job will be | |
777 | named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum` | |
778 | will be used if no other format specifier is given. | |
779 | ||
780 | .. option:: unique_filename=bool | |
781 | ||
782 | To avoid collisions between networked clients, fio defaults to prefixing any | |
783 | generated filenames (with a directory specified) with the source of the | |
784 | client connecting. To disable this behavior, set this option to 0. | |
785 | ||
786 | .. option:: opendir=str | |
787 | ||
788 | Recursively open any files below directory `str`. | |
789 | ||
790 | .. option:: lockfile=str | |
791 | ||
792 | Fio defaults to not locking any files before it does I/O to them. If a file | |
793 | or file descriptor is shared, fio can serialize I/O to that file to make the | |
794 | end result consistent. This is usual for emulating real workloads that share | |
795 | files. The lock modes are: | |
796 | ||
797 | **none** | |
798 | No locking. The default. | |
799 | **exclusive** | |
800 | Only one thread or process may do I/O at a time, excluding all | |
801 | others. | |
802 | **readwrite** | |
803 | Read-write locking on the file. Many readers may | |
804 | access the file at the same time, but writes get exclusive access. | |
805 | ||
806 | .. option:: nrfiles=int | |
807 | ||
79591fa9 TK |
808 | Number of files to use for this job. Defaults to 1. The size of files |
809 | will be :option:`size` divided by this unless explicit size is specified by | |
810 | :option:`filesize`. Files are created for each thread separately, and each | |
811 | file will have a file number within its name by default, as explained in | |
812 | :option:`filename` section. | |
813 | ||
f80dba8d MT |
814 | |
815 | .. option:: openfiles=int | |
816 | ||
817 | Number of files to keep open at the same time. Defaults to the same as | |
818 | :option:`nrfiles`, can be set smaller to limit the number simultaneous | |
819 | opens. | |
820 | ||
821 | .. option:: file_service_type=str | |
822 | ||
823 | Defines how fio decides which file from a job to service next. The following | |
824 | types are defined: | |
825 | ||
826 | **random** | |
827 | Choose a file at random. | |
828 | ||
829 | **roundrobin** | |
830 | Round robin over opened files. This is the default. | |
831 | ||
832 | **sequential** | |
833 | Finish one file before moving on to the next. Multiple files can | |
834 | still be open depending on 'openfiles'. | |
835 | ||
836 | **zipf** | |
c60ebc45 | 837 | Use a *Zipf* distribution to decide what file to access. |
f80dba8d MT |
838 | |
839 | **pareto** | |
c60ebc45 | 840 | Use a *Pareto* distribution to decide what file to access. |
f80dba8d MT |
841 | |
842 | **gauss** | |
c60ebc45 | 843 | Use a *Gaussian* (normal) distribution to decide what file to |
f80dba8d MT |
844 | access. |
845 | ||
846 | For *random*, *roundrobin*, and *sequential*, a postfix can be appended to | |
847 | tell fio how many I/Os to issue before switching to a new file. For example, | |
848 | specifying ``file_service_type=random:8`` would cause fio to issue | |
849 | 8 I/Os before selecting a new file at random. For the non-uniform | |
850 | distributions, a floating point postfix can be given to influence how the | |
851 | distribution is skewed. See :option:`random_distribution` for a description | |
852 | of how that would work. | |
853 | ||
854 | .. option:: ioscheduler=str | |
855 | ||
856 | Attempt to switch the device hosting the file to the specified I/O scheduler | |
857 | before running. | |
858 | ||
859 | .. option:: create_serialize=bool | |
860 | ||
861 | If true, serialize the file creation for the jobs. This may be handy to | |
862 | avoid interleaving of data files, which may greatly depend on the filesystem | |
863 | used and even the number of processors in the system. | |
864 | ||
865 | .. option:: create_fsync=bool | |
866 | ||
867 | fsync the data file after creation. This is the default. | |
868 | ||
869 | .. option:: create_on_open=bool | |
870 | ||
871 | Don't pre-setup the files for I/O, just create open() when it's time to do | |
872 | I/O to that file. | |
873 | ||
874 | .. option:: create_only=bool | |
875 | ||
876 | If true, fio will only run the setup phase of the job. If files need to be | |
877 | laid out or updated on disk, only that will be done. The actual job contents | |
878 | are not executed. | |
879 | ||
880 | .. option:: allow_file_create=bool | |
881 | ||
882 | If true, fio is permitted to create files as part of its workload. This is | |
883 | the default behavior. If this option is false, then fio will error out if | |
884 | the files it needs to use don't already exist. Default: true. | |
885 | ||
886 | .. option:: allow_mounted_write=bool | |
887 | ||
c60ebc45 | 888 | If this isn't set, fio will abort jobs that are destructive (e.g. that write) |
f80dba8d MT |
889 | to what appears to be a mounted device or partition. This should help catch |
890 | creating inadvertently destructive tests, not realizing that the test will | |
b1db0375 TK |
891 | destroy data on the mounted file system. Note that some platforms don't allow |
892 | writing against a mounted device regardless of this option. Default: false. | |
f80dba8d MT |
893 | |
894 | .. option:: pre_read=bool | |
895 | ||
896 | If this is given, files will be pre-read into memory before starting the | |
897 | given I/O operation. This will also clear the :option:`invalidate` flag, | |
898 | since it is pointless to pre-read and then drop the cache. This will only | |
899 | work for I/O engines that are seek-able, since they allow you to read the | |
c60ebc45 | 900 | same data multiple times. Thus it will not work on e.g. network or splice I/O. |
f80dba8d MT |
901 | |
902 | .. option:: unlink=bool | |
903 | ||
904 | Unlink the job files when done. Not the default, as repeated runs of that | |
905 | job would then waste time recreating the file set again and again. | |
906 | ||
907 | .. option:: unlink_each_loop=bool | |
908 | ||
909 | Unlink job files after each iteration or loop. | |
910 | ||
911 | .. option:: zonesize=int | |
912 | ||
913 | Divide a file into zones of the specified size. See :option:`zoneskip`. | |
914 | ||
915 | .. option:: zonerange=int | |
916 | ||
917 | Give size of an I/O zone. See :option:`zoneskip`. | |
918 | ||
919 | .. option:: zoneskip=int | |
920 | ||
921 | Skip the specified number of bytes when :option:`zonesize` data has been | |
922 | read. The two zone options can be used to only do I/O on zones of a file. | |
923 | ||
924 | ||
925 | I/O type | |
926 | ~~~~~~~~ | |
927 | ||
928 | .. option:: direct=bool | |
929 | ||
930 | If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that | |
931 | ZFS on Solaris doesn't support direct I/O. On Windows the synchronous | |
932 | ioengines don't support direct I/O. Default: false. | |
933 | ||
934 | .. option:: atomic=bool | |
935 | ||
936 | If value is true, attempt to use atomic direct I/O. Atomic writes are | |
937 | guaranteed to be stable once acknowledged by the operating system. Only | |
938 | Linux supports O_ATOMIC right now. | |
939 | ||
940 | .. option:: buffered=bool | |
941 | ||
942 | If value is true, use buffered I/O. This is the opposite of the | |
943 | :option:`direct` option. Defaults to true. | |
944 | ||
945 | .. option:: readwrite=str, rw=str | |
946 | ||
947 | Type of I/O pattern. Accepted values are: | |
948 | ||
949 | **read** | |
950 | Sequential reads. | |
951 | **write** | |
952 | Sequential writes. | |
953 | **trim** | |
954 | Sequential trims (Linux block devices only). | |
955 | **randwrite** | |
956 | Random writes. | |
957 | **randread** | |
958 | Random reads. | |
959 | **randtrim** | |
960 | Random trims (Linux block devices only). | |
961 | **rw,readwrite** | |
962 | Sequential mixed reads and writes. | |
963 | **randrw** | |
964 | Random mixed reads and writes. | |
965 | **trimwrite** | |
966 | Sequential trim+write sequences. Blocks will be trimmed first, | |
967 | then the same blocks will be written to. | |
968 | ||
969 | Fio defaults to read if the option is not specified. For the mixed I/O | |
970 | types, the default is to split them 50/50. For certain types of I/O the | |
971 | result may still be skewed a bit, since the speed may be different. It is | |
972 | possible to specify a number of I/O's to do before getting a new offset, | |
973 | this is done by appending a ``:<nr>`` to the end of the string given. For a | |
974 | random read, it would look like ``rw=randread:8`` for passing in an offset | |
975 | modifier with a value of 8. If the suffix is used with a sequential I/O | |
976 | pattern, then the value specified will be added to the generated offset for | |
977 | each I/O. For instance, using ``rw=write:4k`` will skip 4k for every | |
978 | write. It turns sequential I/O into sequential I/O with holes. See the | |
979 | :option:`rw_sequencer` option. | |
980 | ||
981 | .. option:: rw_sequencer=str | |
982 | ||
983 | If an offset modifier is given by appending a number to the ``rw=<str>`` | |
984 | line, then this option controls how that number modifies the I/O offset | |
985 | being generated. Accepted values are: | |
986 | ||
987 | **sequential** | |
988 | Generate sequential offset. | |
989 | **identical** | |
990 | Generate the same offset. | |
991 | ||
992 | ``sequential`` is only useful for random I/O, where fio would normally | |
c60ebc45 | 993 | generate a new random offset for every I/O. If you append e.g. 8 to randread, |
f80dba8d MT |
994 | you would get a new random offset for every 8 I/O's. The result would be a |
995 | seek for only every 8 I/O's, instead of for every I/O. Use ``rw=randread:8`` | |
996 | to specify that. As sequential I/O is already sequential, setting | |
997 | ``sequential`` for that would not result in any differences. ``identical`` | |
998 | behaves in a similar fashion, except it sends the same offset 8 number of | |
999 | times before generating a new offset. | |
1000 | ||
1001 | .. option:: unified_rw_reporting=bool | |
1002 | ||
1003 | Fio normally reports statistics on a per data direction basis, meaning that | |
1004 | reads, writes, and trims are accounted and reported separately. If this | |
1005 | option is set fio sums the results and report them as "mixed" instead. | |
1006 | ||
1007 | .. option:: randrepeat=bool | |
1008 | ||
1009 | Seed the random number generator used for random I/O patterns in a | |
1010 | predictable way so the pattern is repeatable across runs. Default: true. | |
1011 | ||
1012 | .. option:: allrandrepeat=bool | |
1013 | ||
1014 | Seed all random number generators in a predictable way so results are | |
1015 | repeatable across runs. Default: false. | |
1016 | ||
1017 | .. option:: randseed=int | |
1018 | ||
1019 | Seed the random number generators based on this seed value, to be able to | |
1020 | control what sequence of output is being generated. If not set, the random | |
1021 | sequence depends on the :option:`randrepeat` setting. | |
1022 | ||
1023 | .. option:: fallocate=str | |
1024 | ||
1025 | Whether pre-allocation is performed when laying down files. | |
1026 | Accepted values are: | |
1027 | ||
1028 | **none** | |
1029 | Do not pre-allocate space. | |
1030 | ||
1031 | **posix** | |
1032 | Pre-allocate via :manpage:`posix_fallocate(3)`. | |
1033 | ||
1034 | **keep** | |
1035 | Pre-allocate via :manpage:`fallocate(2)` with | |
1036 | FALLOC_FL_KEEP_SIZE set. | |
1037 | ||
1038 | **0** | |
1039 | Backward-compatible alias for **none**. | |
1040 | ||
1041 | **1** | |
1042 | Backward-compatible alias for **posix**. | |
1043 | ||
1044 | May not be available on all supported platforms. **keep** is only available | |
1045 | on Linux. If using ZFS on Solaris this must be set to **none** because ZFS | |
1046 | doesn't support it. Default: **posix**. | |
1047 | ||
1048 | .. option:: fadvise_hint=str | |
1049 | ||
1050 | Use :manpage:`posix_fadvise(2)` to advise the kernel on what I/O patterns | |
1051 | are likely to be issued. Accepted values are: | |
1052 | ||
1053 | **0** | |
1054 | Backwards-compatible hint for "no hint". | |
1055 | ||
1056 | **1** | |
1057 | Backwards compatible hint for "advise with fio workload type". This | |
1058 | uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL** | |
1059 | for a sequential workload. | |
1060 | ||
1061 | **sequential** | |
1062 | Advise using **FADV_SEQUENTIAL**. | |
1063 | ||
1064 | **random** | |
1065 | Advise using **FADV_RANDOM**. | |
1066 | ||
1067 | .. option:: fadvise_stream=int | |
1068 | ||
1069 | Use :manpage:`posix_fadvise(2)` to advise the kernel what stream ID the | |
1070 | writes issued belong to. Only supported on Linux. Note, this option may | |
1071 | change going forward. | |
1072 | ||
1073 | .. option:: offset=int | |
1074 | ||
89978a6b BW |
1075 | Start I/O at the provided offset in the file, given as either a fixed size or |
1076 | a percentage. If a percentage is given, the next ``blockalign``-ed offset | |
1077 | will be used. Data before the given offset will not be touched. This | |
1078 | effectively caps the file size at `real_size - offset`. Can be combined with | |
1079 | :option:`size` to constrain the start and end range of the I/O workload. | |
f80dba8d MT |
1080 | |
1081 | .. option:: offset_increment=int | |
1082 | ||
1083 | If this is provided, then the real offset becomes `offset + offset_increment | |
1084 | * thread_number`, where the thread number is a counter that starts at 0 and | |
1085 | is incremented for each sub-job (i.e. when :option:`numjobs` option is | |
1086 | specified). This option is useful if there are several jobs which are | |
1087 | intended to operate on a file in parallel disjoint segments, with even | |
1088 | spacing between the starting points. | |
1089 | ||
1090 | .. option:: number_ios=int | |
1091 | ||
c60ebc45 | 1092 | Fio will normally perform I/Os until it has exhausted the size of the region |
f80dba8d MT |
1093 | set by :option:`size`, or if it exhaust the allocated time (or hits an error |
1094 | condition). With this setting, the range/size can be set independently of | |
c60ebc45 | 1095 | the number of I/Os to perform. When fio reaches this number, it will exit |
f80dba8d MT |
1096 | normally and report status. Note that this does not extend the amount of I/O |
1097 | that will be done, it will only stop fio if this condition is met before | |
1098 | other end-of-job criteria. | |
1099 | ||
1100 | .. option:: fsync=int | |
1101 | ||
1102 | If writing to a file, issue a sync of the dirty data for every number of | |
1103 | blocks given. For example, if you give 32 as a parameter, fio will sync the | |
1104 | file for every 32 writes issued. If fio is using non-buffered I/O, we may | |
1105 | not sync the file. The exception is the sg I/O engine, which synchronizes | |
54227e6b TK |
1106 | the disk cache anyway. Defaults to 0, which means no sync every certain |
1107 | number of writes. | |
f80dba8d MT |
1108 | |
1109 | .. option:: fdatasync=int | |
1110 | ||
1111 | Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and | |
000a5f1c | 1112 | not metadata blocks. In Windows, FreeBSD, and DragonFlyBSD there is no |
f80dba8d | 1113 | :manpage:`fdatasync(2)`, this falls back to using :manpage:`fsync(2)`. |
54227e6b | 1114 | Defaults to 0, which means no sync data every certain number of writes. |
f80dba8d MT |
1115 | |
1116 | .. option:: write_barrier=int | |
1117 | ||
1118 | Make every `N-th` write a barrier write. | |
1119 | ||
1120 | .. option:: sync_file_range=str:val | |
1121 | ||
1122 | Use :manpage:`sync_file_range(2)` for every `val` number of write | |
1123 | operations. Fio will track range of writes that have happened since the last | |
1124 | :manpage:`sync_file_range(2)` call. `str` can currently be one or more of: | |
1125 | ||
1126 | **wait_before** | |
1127 | SYNC_FILE_RANGE_WAIT_BEFORE | |
1128 | **write** | |
1129 | SYNC_FILE_RANGE_WRITE | |
1130 | **wait_after** | |
1131 | SYNC_FILE_RANGE_WAIT_AFTER | |
1132 | ||
1133 | So if you do ``sync_file_range=wait_before,write:8``, fio would use | |
1134 | ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8 | |
1135 | writes. Also see the :manpage:`sync_file_range(2)` man page. This option is | |
1136 | Linux specific. | |
1137 | ||
1138 | .. option:: overwrite=bool | |
1139 | ||
1140 | If true, writes to a file will always overwrite existing data. If the file | |
1141 | doesn't already exist, it will be created before the write phase begins. If | |
1142 | the file exists and is large enough for the specified write phase, nothing | |
1143 | will be done. | |
1144 | ||
1145 | .. option:: end_fsync=bool | |
1146 | ||
1147 | If true, fsync file contents when a write stage has completed. | |
1148 | ||
1149 | .. option:: fsync_on_close=bool | |
1150 | ||
1151 | If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs | |
1152 | from end_fsync in that it will happen on every file close, not just at the | |
1153 | end of the job. | |
1154 | ||
1155 | .. option:: rwmixread=int | |
1156 | ||
1157 | Percentage of a mixed workload that should be reads. Default: 50. | |
1158 | ||
1159 | .. option:: rwmixwrite=int | |
1160 | ||
1161 | Percentage of a mixed workload that should be writes. If both | |
1162 | :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not | |
1163 | add up to 100%, the latter of the two will be used to override the | |
1164 | first. This may interfere with a given rate setting, if fio is asked to | |
1165 | limit reads or writes to a certain rate. If that is the case, then the | |
1166 | distribution may be skewed. Default: 50. | |
1167 | ||
1168 | .. option:: random_distribution=str:float[,str:float][,str:float] | |
1169 | ||
1170 | By default, fio will use a completely uniform random distribution when asked | |
1171 | to perform random I/O. Sometimes it is useful to skew the distribution in | |
1172 | specific ways, ensuring that some parts of the data is more hot than others. | |
1173 | fio includes the following distribution models: | |
1174 | ||
1175 | **random** | |
1176 | Uniform random distribution | |
1177 | ||
1178 | **zipf** | |
1179 | Zipf distribution | |
1180 | ||
1181 | **pareto** | |
1182 | Pareto distribution | |
1183 | ||
1184 | **gauss** | |
c60ebc45 | 1185 | Normal (Gaussian) distribution |
f80dba8d MT |
1186 | |
1187 | **zoned** | |
1188 | Zoned random distribution | |
1189 | ||
1190 | When using a **zipf** or **pareto** distribution, an input value is also | |
1191 | needed to define the access pattern. For **zipf**, this is the `zipf | |
c60ebc45 | 1192 | theta`. For **pareto**, it's the `Pareto power`. Fio includes a test |
f80dba8d MT |
1193 | program, :command:`genzipf`, that can be used visualize what the given input |
1194 | values will yield in terms of hit rates. If you wanted to use **zipf** with | |
1195 | a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the | |
1196 | option. If a non-uniform model is used, fio will disable use of the random | |
1197 | map. For the **gauss** distribution, a normal deviation is supplied as a | |
1198 | value between 0 and 100. | |
1199 | ||
1200 | For a **zoned** distribution, fio supports specifying percentages of I/O | |
1201 | access that should fall within what range of the file or device. For | |
1202 | example, given a criteria of: | |
1203 | ||
1204 | * 60% of accesses should be to the first 10% | |
1205 | * 30% of accesses should be to the next 20% | |
1206 | * 8% of accesses should be to to the next 30% | |
1207 | * 2% of accesses should be to the next 40% | |
1208 | ||
1209 | we can define that through zoning of the random accesses. For the above | |
1210 | example, the user would do:: | |
1211 | ||
1212 | random_distribution=zoned:60/10:30/20:8/30:2/40 | |
1213 | ||
1214 | similarly to how :option:`bssplit` works for setting ranges and percentages | |
1215 | of block sizes. Like :option:`bssplit`, it's possible to specify separate | |
1216 | zones for reads, writes, and trims. If just one set is given, it'll apply to | |
1217 | all of them. | |
1218 | ||
1219 | .. option:: percentage_random=int[,int][,int] | |
1220 | ||
1221 | For a random workload, set how big a percentage should be random. This | |
1222 | defaults to 100%, in which case the workload is fully random. It can be set | |
1223 | from anywhere from 0 to 100. Setting it to 0 would make the workload fully | |
1224 | sequential. Any setting in between will result in a random mix of sequential | |
1225 | and random I/O, at the given percentages. Comma-separated values may be | |
1226 | specified for reads, writes, and trims as described in :option:`blocksize`. | |
1227 | ||
1228 | .. option:: norandommap | |
1229 | ||
1230 | Normally fio will cover every block of the file when doing random I/O. If | |
1231 | this option is given, fio will just get a new random offset without looking | |
1232 | at past I/O history. This means that some blocks may not be read or written, | |
1233 | and that some blocks may be read/written more than once. If this option is | |
1234 | used with :option:`verify` and multiple blocksizes (via :option:`bsrange`), | |
1235 | only intact blocks are verified, i.e., partially-overwritten blocks are | |
1236 | ignored. | |
1237 | ||
1238 | .. option:: softrandommap=bool | |
1239 | ||
1240 | See :option:`norandommap`. If fio runs with the random block map enabled and | |
1241 | it fails to allocate the map, if this option is set it will continue without | |
1242 | a random block map. As coverage will not be as complete as with random maps, | |
1243 | this option is disabled by default. | |
1244 | ||
1245 | .. option:: random_generator=str | |
1246 | ||
1247 | Fio supports the following engines for generating | |
1248 | I/O offsets for random I/O: | |
1249 | ||
1250 | **tausworthe** | |
1251 | Strong 2^88 cycle random number generator | |
1252 | **lfsr** | |
1253 | Linear feedback shift register generator | |
1254 | **tausworthe64** | |
1255 | Strong 64-bit 2^258 cycle random number generator | |
1256 | ||
1257 | **tausworthe** is a strong random number generator, but it requires tracking | |
1258 | on the side if we want to ensure that blocks are only read or written | |
1259 | once. **LFSR** guarantees that we never generate the same offset twice, and | |
1260 | it's also less computationally expensive. It's not a true random generator, | |
1261 | however, though for I/O purposes it's typically good enough. **LFSR** only | |
1262 | works with single block sizes, not with workloads that use multiple block | |
1263 | sizes. If used with such a workload, fio may read or write some blocks | |
1264 | multiple times. The default value is **tausworthe**, unless the required | |
1265 | space exceeds 2^32 blocks. If it does, then **tausworthe64** is | |
1266 | selected automatically. | |
1267 | ||
1268 | ||
1269 | Block size | |
1270 | ~~~~~~~~~~ | |
1271 | ||
1272 | .. option:: blocksize=int[,int][,int], bs=int[,int][,int] | |
1273 | ||
1274 | The block size in bytes used for I/O units. Default: 4096. A single value | |
1275 | applies to reads, writes, and trims. Comma-separated values may be | |
1276 | specified for reads, writes, and trims. A value not terminated in a comma | |
1277 | applies to subsequent types. | |
1278 | ||
1279 | Examples: | |
1280 | ||
1281 | **bs=256k** | |
1282 | means 256k for reads, writes and trims. | |
1283 | ||
1284 | **bs=8k,32k** | |
1285 | means 8k for reads, 32k for writes and trims. | |
1286 | ||
1287 | **bs=8k,32k,** | |
1288 | means 8k for reads, 32k for writes, and default for trims. | |
1289 | ||
1290 | **bs=,8k** | |
1291 | means default for reads, 8k for writes and trims. | |
1292 | ||
1293 | **bs=,8k,** | |
b443ae44 | 1294 | means default for reads, 8k for writes, and default for trims. |
f80dba8d MT |
1295 | |
1296 | .. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange] | |
1297 | ||
1298 | A range of block sizes in bytes for I/O units. The issued I/O unit will | |
1299 | always be a multiple of the minimum size, unless | |
1300 | :option:`blocksize_unaligned` is set. | |
1301 | ||
1302 | Comma-separated ranges may be specified for reads, writes, and trims as | |
1303 | described in :option:`blocksize`. | |
1304 | ||
1305 | Example: ``bsrange=1k-4k,2k-8k``. | |
1306 | ||
1307 | .. option:: bssplit=str[,str][,str] | |
1308 | ||
1309 | Sometimes you want even finer grained control of the block sizes issued, not | |
1310 | just an even split between them. This option allows you to weight various | |
1311 | block sizes, so that you are able to define a specific amount of block sizes | |
1312 | issued. The format for this option is:: | |
1313 | ||
1314 | bssplit=blocksize/percentage:blocksize/percentage | |
1315 | ||
1316 | for as many block sizes as needed. So if you want to define a workload that | |
1317 | has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would write:: | |
1318 | ||
1319 | bssplit=4k/10:64k/50:32k/40 | |
1320 | ||
1321 | Ordering does not matter. If the percentage is left blank, fio will fill in | |
1322 | the remaining values evenly. So a bssplit option like this one:: | |
1323 | ||
1324 | bssplit=4k/50:1k/:32k/ | |
1325 | ||
1326 | would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always add up | |
1327 | to 100, if bssplit is given a range that adds up to more, it will error out. | |
1328 | ||
1329 | Comma-separated values may be specified for reads, writes, and trims as | |
1330 | described in :option:`blocksize`. | |
1331 | ||
1332 | If you want a workload that has 50% 2k reads and 50% 4k reads, while having | |
1333 | 90% 4k writes and 10% 8k writes, you would specify:: | |
1334 | ||
1335 | bssplit=2k/50:4k/50,4k/90,8k/10 | |
1336 | ||
1337 | .. option:: blocksize_unaligned, bs_unaligned | |
1338 | ||
1339 | If set, fio will issue I/O units with any size within | |
1340 | :option:`blocksize_range`, not just multiples of the minimum size. This | |
1341 | typically won't work with direct I/O, as that normally requires sector | |
1342 | alignment. | |
1343 | ||
1344 | .. option:: bs_is_seq_rand | |
1345 | ||
1346 | If this option is set, fio will use the normal read,write blocksize settings | |
1347 | as sequential,random blocksize settings instead. Any random read or write | |
1348 | will use the WRITE blocksize settings, and any sequential read or write will | |
1349 | use the READ blocksize settings. | |
1350 | ||
1351 | .. option:: blockalign=int[,int][,int], ba=int[,int][,int] | |
1352 | ||
1353 | Boundary to which fio will align random I/O units. Default: | |
1354 | :option:`blocksize`. Minimum alignment is typically 512b for using direct | |
1355 | I/O, though it usually depends on the hardware block size. This option is | |
1356 | mutually exclusive with using a random map for files, so it will turn off | |
1357 | that option. Comma-separated values may be specified for reads, writes, and | |
1358 | trims as described in :option:`blocksize`. | |
1359 | ||
1360 | ||
1361 | Buffers and memory | |
1362 | ~~~~~~~~~~~~~~~~~~ | |
1363 | ||
1364 | .. option:: zero_buffers | |
1365 | ||
1366 | Initialize buffers with all zeros. Default: fill buffers with random data. | |
1367 | ||
1368 | .. option:: refill_buffers | |
1369 | ||
1370 | If this option is given, fio will refill the I/O buffers on every | |
1371 | submit. The default is to only fill it at init time and reuse that | |
1372 | data. Only makes sense if zero_buffers isn't specified, naturally. If data | |
1373 | verification is enabled, `refill_buffers` is also automatically enabled. | |
1374 | ||
1375 | .. option:: scramble_buffers=bool | |
1376 | ||
1377 | If :option:`refill_buffers` is too costly and the target is using data | |
1378 | deduplication, then setting this option will slightly modify the I/O buffer | |
1379 | contents to defeat normal de-dupe attempts. This is not enough to defeat | |
1380 | more clever block compression attempts, but it will stop naive dedupe of | |
1381 | blocks. Default: true. | |
1382 | ||
1383 | .. option:: buffer_compress_percentage=int | |
1384 | ||
1385 | If this is set, then fio will attempt to provide I/O buffer content (on | |
1386 | WRITEs) that compress to the specified level. Fio does this by providing a | |
1387 | mix of random data and a fixed pattern. The fixed pattern is either zeroes, | |
1388 | or the pattern specified by :option:`buffer_pattern`. If the pattern option | |
1389 | is used, it might skew the compression ratio slightly. Note that this is per | |
1390 | block size unit, for file/disk wide compression level that matches this | |
1391 | setting, you'll also want to set :option:`refill_buffers`. | |
1392 | ||
1393 | .. option:: buffer_compress_chunk=int | |
1394 | ||
1395 | See :option:`buffer_compress_percentage`. This setting allows fio to manage | |
1396 | how big the ranges of random data and zeroed data is. Without this set, fio | |
1397 | will provide :option:`buffer_compress_percentage` of blocksize random data, | |
1398 | followed by the remaining zeroed. With this set to some chunk size smaller | |
1399 | than the block size, fio can alternate random and zeroed data throughout the | |
1400 | I/O buffer. | |
1401 | ||
1402 | .. option:: buffer_pattern=str | |
1403 | ||
a1554f65 SB |
1404 | If set, fio will fill the I/O buffers with this pattern or with the contents |
1405 | of a file. If not set, the contents of I/O buffers are defined by the other | |
1406 | options related to buffer contents. The setting can be any pattern of bytes, | |
1407 | and can be prefixed with 0x for hex values. It may also be a string, where | |
1408 | the string must then be wrapped with ``""``. Or it may also be a filename, | |
1409 | where the filename must be wrapped with ``''`` in which case the file is | |
1410 | opened and read. Note that not all the file contents will be read if that | |
1411 | would cause the buffers to overflow. So, for example:: | |
1412 | ||
1413 | buffer_pattern='filename' | |
1414 | ||
1415 | or:: | |
f80dba8d MT |
1416 | |
1417 | buffer_pattern="abcd" | |
1418 | ||
1419 | or:: | |
1420 | ||
1421 | buffer_pattern=-12 | |
1422 | ||
1423 | or:: | |
1424 | ||
1425 | buffer_pattern=0xdeadface | |
1426 | ||
1427 | Also you can combine everything together in any order:: | |
1428 | ||
a1554f65 | 1429 | buffer_pattern=0xdeadface"abcd"-12'filename' |
f80dba8d MT |
1430 | |
1431 | .. option:: dedupe_percentage=int | |
1432 | ||
1433 | If set, fio will generate this percentage of identical buffers when | |
1434 | writing. These buffers will be naturally dedupable. The contents of the | |
1435 | buffers depend on what other buffer compression settings have been set. It's | |
1436 | possible to have the individual buffers either fully compressible, or not at | |
1437 | all. This option only controls the distribution of unique buffers. | |
1438 | ||
1439 | .. option:: invalidate=bool | |
1440 | ||
1441 | Invalidate the buffer/page cache parts for this file prior to starting | |
21c1b29e TK |
1442 | I/O if the platform and file type support it. Defaults to true. |
1443 | This will be ignored if :option:`pre_read` is also specified for the | |
1444 | same job. | |
f80dba8d MT |
1445 | |
1446 | .. option:: sync=bool | |
1447 | ||
1448 | Use synchronous I/O for buffered writes. For the majority of I/O engines, | |
1449 | this means using O_SYNC. Default: false. | |
1450 | ||
1451 | .. option:: iomem=str, mem=str | |
1452 | ||
1453 | Fio can use various types of memory as the I/O unit buffer. The allowed | |
1454 | values are: | |
1455 | ||
1456 | **malloc** | |
1457 | Use memory from :manpage:`malloc(3)` as the buffers. Default memory | |
1458 | type. | |
1459 | ||
1460 | **shm** | |
1461 | Use shared memory as the buffers. Allocated through | |
1462 | :manpage:`shmget(2)`. | |
1463 | ||
1464 | **shmhuge** | |
1465 | Same as shm, but use huge pages as backing. | |
1466 | ||
1467 | **mmap** | |
1468 | Use mmap to allocate buffers. May either be anonymous memory, or can | |
1469 | be file backed if a filename is given after the option. The format | |
1470 | is `mem=mmap:/path/to/file`. | |
1471 | ||
1472 | **mmaphuge** | |
1473 | Use a memory mapped huge file as the buffer backing. Append filename | |
1474 | after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`. | |
1475 | ||
1476 | **mmapshared** | |
1477 | Same as mmap, but use a MMAP_SHARED mapping. | |
1478 | ||
03553853 YR |
1479 | **cudamalloc** |
1480 | Use GPU memory as the buffers for GPUDirect RDMA benchmark. | |
1481 | ||
f80dba8d MT |
1482 | The area allocated is a function of the maximum allowed bs size for the job, |
1483 | multiplied by the I/O depth given. Note that for **shmhuge** and | |
1484 | **mmaphuge** to work, the system must have free huge pages allocated. This | |
1485 | can normally be checked and set by reading/writing | |
1486 | :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page | |
1487 | is 4MiB in size. So to calculate the number of huge pages you need for a | |
1488 | given job file, add up the I/O depth of all jobs (normally one unless | |
1489 | :option:`iodepth` is used) and multiply by the maximum bs set. Then divide | |
1490 | that number by the huge page size. You can see the size of the huge pages in | |
1491 | :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero | |
1492 | number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also | |
1493 | see :option:`hugepage-size`. | |
1494 | ||
1495 | **mmaphuge** also needs to have hugetlbfs mounted and the file location | |
1496 | should point there. So if it's mounted in :file:`/huge`, you would use | |
1497 | `mem=mmaphuge:/huge/somefile`. | |
1498 | ||
1499 | .. option:: iomem_align=int | |
1500 | ||
1501 | This indicates the memory alignment of the I/O memory buffers. Note that | |
1502 | the given alignment is applied to the first I/O unit buffer, if using | |
1503 | :option:`iodepth` the alignment of the following buffers are given by the | |
1504 | :option:`bs` used. In other words, if using a :option:`bs` that is a | |
1505 | multiple of the page sized in the system, all buffers will be aligned to | |
1506 | this value. If using a :option:`bs` that is not page aligned, the alignment | |
1507 | of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and | |
1508 | :option:`bs` used. | |
1509 | ||
1510 | .. option:: hugepage-size=int | |
1511 | ||
1512 | Defines the size of a huge page. Must at least be equal to the system | |
1513 | setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably | |
1514 | always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the | |
1515 | preferred way to set this to avoid setting a non-pow-2 bad value. | |
1516 | ||
1517 | .. option:: lockmem=int | |
1518 | ||
1519 | Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to | |
1520 | simulate a smaller amount of memory. The amount specified is per worker. | |
1521 | ||
1522 | ||
1523 | I/O size | |
1524 | ~~~~~~~~ | |
1525 | ||
1526 | .. option:: size=int | |
1527 | ||
79591fa9 TK |
1528 | The total size of file I/O for each thread of this job. Fio will run until |
1529 | this many bytes has been transferred, unless runtime is limited by other options | |
1530 | (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`). | |
1531 | Fio will divide this size between the available files determined by options | |
1532 | such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is | |
1533 | specified by the job. If the result of division happens to be 0, the size is | |
c4aa2d08 | 1534 | set to the physical size of the given files or devices if they exist. |
79591fa9 | 1535 | If this option is not specified, fio will use the full size of the given |
f80dba8d MT |
1536 | files or devices. If the files do not exist, size must be given. It is also |
1537 | possible to give size as a percentage between 1 and 100. If ``size=20%`` is | |
1538 | given, fio will use 20% of the full size of the given files or devices. | |
9d25d068 SW |
1539 | Can be combined with :option:`offset` to constrain the start and end range |
1540 | that I/O will be done within. | |
f80dba8d MT |
1541 | |
1542 | .. option:: io_size=int, io_limit=int | |
1543 | ||
1544 | Normally fio operates within the region set by :option:`size`, which means | |
1545 | that the :option:`size` option sets both the region and size of I/O to be | |
1546 | performed. Sometimes that is not what you want. With this option, it is | |
1547 | possible to define just the amount of I/O that fio should do. For instance, | |
1548 | if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio | |
1549 | will perform I/O within the first 20GiB but exit when 5GiB have been | |
1550 | done. The opposite is also possible -- if :option:`size` is set to 20GiB, | |
1551 | and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within | |
1552 | the 0..20GiB region. | |
1553 | ||
1554 | .. option:: filesize=int | |
1555 | ||
1556 | Individual file sizes. May be a range, in which case fio will select sizes | |
1557 | for files at random within the given range and limited to :option:`size` in | |
1558 | total (if that is given). If not given, each created file is the same size. | |
79591fa9 TK |
1559 | This option overrides :option:`size` in terms of file size, which means |
1560 | this value is used as a fixed size or possible range of each file. | |
f80dba8d MT |
1561 | |
1562 | .. option:: file_append=bool | |
1563 | ||
1564 | Perform I/O after the end of the file. Normally fio will operate within the | |
1565 | size of a file. If this option is set, then fio will append to the file | |
1566 | instead. This has identical behavior to setting :option:`offset` to the size | |
1567 | of a file. This option is ignored on non-regular files. | |
1568 | ||
1569 | .. option:: fill_device=bool, fill_fs=bool | |
1570 | ||
1571 | Sets size to something really large and waits for ENOSPC (no space left on | |
1572 | device) as the terminating condition. Only makes sense with sequential | |
1573 | write. For a read workload, the mount point will be filled first then I/O | |
1574 | started on the result. This option doesn't make sense if operating on a raw | |
1575 | device node, since the size of that is already known by the file system. | |
1576 | Additionally, writing beyond end-of-device will not return ENOSPC there. | |
1577 | ||
1578 | ||
1579 | I/O engine | |
1580 | ~~~~~~~~~~ | |
1581 | ||
1582 | .. option:: ioengine=str | |
1583 | ||
1584 | Defines how the job issues I/O to the file. The following types are defined: | |
1585 | ||
1586 | **sync** | |
1587 | Basic :manpage:`read(2)` or :manpage:`write(2)` | |
1588 | I/O. :manpage:`lseek(2)` is used to position the I/O location. | |
54227e6b | 1589 | See :option:`fsync` and :option:`fdatasync` for syncing write I/Os. |
f80dba8d MT |
1590 | |
1591 | **psync** | |
1592 | Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on | |
1593 | all supported operating systems except for Windows. | |
1594 | ||
1595 | **vsync** | |
1596 | Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate | |
c60ebc45 | 1597 | queuing by coalescing adjacent I/Os into a single submission. |
f80dba8d MT |
1598 | |
1599 | **pvsync** | |
1600 | Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O. | |
1601 | ||
1602 | **pvsync2** | |
1603 | Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O. | |
1604 | ||
1605 | **libaio** | |
1606 | Linux native asynchronous I/O. Note that Linux may only support | |
1607 | queued behaviour with non-buffered I/O (set ``direct=1`` or | |
1608 | ``buffered=0``). | |
1609 | This engine defines engine specific options. | |
1610 | ||
1611 | **posixaio** | |
1612 | POSIX asynchronous I/O using :manpage:`aio_read(3)` and | |
1613 | :manpage:`aio_write(3)`. | |
1614 | ||
1615 | **solarisaio** | |
1616 | Solaris native asynchronous I/O. | |
1617 | ||
1618 | **windowsaio** | |
1619 | Windows native asynchronous I/O. Default on Windows. | |
1620 | ||
1621 | **mmap** | |
1622 | File is memory mapped with :manpage:`mmap(2)` and data copied | |
1623 | to/from using :manpage:`memcpy(3)`. | |
1624 | ||
1625 | **splice** | |
1626 | :manpage:`splice(2)` is used to transfer the data and | |
1627 | :manpage:`vmsplice(2)` to transfer data from user space to the | |
1628 | kernel. | |
1629 | ||
1630 | **sg** | |
1631 | SCSI generic sg v3 I/O. May either be synchronous using the SG_IO | |
1632 | ioctl, or if the target is an sg character device we use | |
1633 | :manpage:`read(2)` and :manpage:`write(2)` for asynchronous | |
1634 | I/O. Requires filename option to specify either block or character | |
1635 | devices. | |
1636 | ||
1637 | **null** | |
1638 | Doesn't transfer any data, just pretends to. This is mainly used to | |
1639 | exercise fio itself and for debugging/testing purposes. | |
1640 | ||
1641 | **net** | |
1642 | Transfer over the network to given ``host:port``. Depending on the | |
1643 | :option:`protocol` used, the :option:`hostname`, :option:`port`, | |
1644 | :option:`listen` and :option:`filename` options are used to specify | |
1645 | what sort of connection to make, while the :option:`protocol` option | |
1646 | determines which protocol will be used. This engine defines engine | |
1647 | specific options. | |
1648 | ||
1649 | **netsplice** | |
1650 | Like **net**, but uses :manpage:`splice(2)` and | |
1651 | :manpage:`vmsplice(2)` to map data and send/receive. | |
1652 | This engine defines engine specific options. | |
1653 | ||
1654 | **cpuio** | |
1655 | Doesn't transfer any data, but burns CPU cycles according to the | |
1656 | :option:`cpuload` and :option:`cpuchunks` options. Setting | |
1657 | :option:`cpuload` =85 will cause that job to do nothing but burn 85% | |
1658 | of the CPU. In case of SMP machines, use :option:`numjobs` | |
1659 | =<no_of_cpu> to get desired CPU usage, as the cpuload only loads a | |
1660 | single CPU at the desired rate. A job never finishes unless there is | |
1661 | at least one non-cpuio job. | |
1662 | ||
1663 | **guasi** | |
1664 | The GUASI I/O engine is the Generic Userspace Asyncronous Syscall | |
1665 | Interface approach to async I/O. See | |
1666 | ||
1667 | http://www.xmailserver.org/guasi-lib.html | |
1668 | ||
1669 | for more info on GUASI. | |
1670 | ||
1671 | **rdma** | |
1672 | The RDMA I/O engine supports both RDMA memory semantics | |
1673 | (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the | |
1674 | InfiniBand, RoCE and iWARP protocols. | |
1675 | ||
1676 | **falloc** | |
1677 | I/O engine that does regular fallocate to simulate data transfer as | |
1678 | fio ioengine. | |
1679 | ||
1680 | DDIR_READ | |
1681 | does fallocate(,mode = FALLOC_FL_KEEP_SIZE,). | |
1682 | ||
1683 | DDIR_WRITE | |
1684 | does fallocate(,mode = 0). | |
1685 | ||
1686 | DDIR_TRIM | |
1687 | does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE). | |
1688 | ||
761cd093 SW |
1689 | **ftruncate** |
1690 | I/O engine that sends :manpage:`ftruncate(2)` operations in response | |
1691 | to write (DDIR_WRITE) events. Each ftruncate issued sets the file's | |
1692 | size to the current block offset. Block size is ignored. | |
1693 | ||
f80dba8d MT |
1694 | **e4defrag** |
1695 | I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate | |
1696 | defragment activity in request to DDIR_WRITE event. | |
1697 | ||
1698 | **rbd** | |
1699 | I/O engine supporting direct access to Ceph Rados Block Devices | |
1700 | (RBD) via librbd without the need to use the kernel rbd driver. This | |
1701 | ioengine defines engine specific options. | |
1702 | ||
1703 | **gfapi** | |
1704 | Using Glusterfs libgfapi sync interface to direct access to | |
1705 | Glusterfs volumes without having to go through FUSE. This ioengine | |
1706 | defines engine specific options. | |
1707 | ||
1708 | **gfapi_async** | |
1709 | Using Glusterfs libgfapi async interface to direct access to | |
1710 | Glusterfs volumes without having to go through FUSE. This ioengine | |
1711 | defines engine specific options. | |
1712 | ||
1713 | **libhdfs** | |
1714 | Read and write through Hadoop (HDFS). The :file:`filename` option | |
1715 | is used to specify host,port of the hdfs name-node to connect. This | |
1716 | engine interprets offsets a little differently. In HDFS, files once | |
1717 | created cannot be modified. So random writes are not possible. To | |
1718 | imitate this, libhdfs engine expects bunch of small files to be | |
1719 | created over HDFS, and engine will randomly pick a file out of those | |
1720 | files based on the offset generated by fio backend. (see the example | |
1721 | job file to create such files, use ``rw=write`` option). Please | |
1722 | note, you might want to set necessary environment variables to work | |
9d25d068 | 1723 | with hdfs/libhdfs properly. Each job uses its own connection to |
f80dba8d MT |
1724 | HDFS. |
1725 | ||
1726 | **mtd** | |
1727 | Read, write and erase an MTD character device (e.g., | |
1728 | :file:`/dev/mtd0`). Discards are treated as erases. Depending on the | |
1729 | underlying device type, the I/O may have to go in a certain pattern, | |
1730 | e.g., on NAND, writing sequentially to erase blocks and discarding | |
1731 | before overwriting. The writetrim mode works well for this | |
1732 | constraint. | |
1733 | ||
1734 | **pmemblk** | |
1735 | Read and write using filesystem DAX to a file on a filesystem | |
1736 | mounted with DAX on a persistent memory device through the NVML | |
1737 | libpmemblk library. | |
1738 | ||
1739 | **dev-dax** | |
1740 | Read and write using device DAX to a persistent memory device (e.g., | |
1741 | /dev/dax0.0) through the NVML libpmem library. | |
1742 | ||
1743 | **external** | |
1744 | Prefix to specify loading an external I/O engine object file. Append | |
c60ebc45 | 1745 | the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load |
f80dba8d MT |
1746 | ioengine :file:`foo.o` in :file:`/tmp`. |
1747 | ||
1748 | ||
1749 | I/O engine specific parameters | |
1750 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
1751 | ||
1752 | In addition, there are some parameters which are only valid when a specific | |
1753 | ioengine is in use. These are used identically to normal parameters, with the | |
1754 | caveat that when used on the command line, they must come after the | |
1755 | :option:`ioengine` that defines them is selected. | |
1756 | ||
1757 | .. option:: userspace_reap : [libaio] | |
1758 | ||
1759 | Normally, with the libaio engine in use, fio will use the | |
1760 | :manpage:`io_getevents(2)` system call to reap newly returned events. With | |
1761 | this flag turned on, the AIO ring will be read directly from user-space to | |
1762 | reap events. The reaping mode is only enabled when polling for a minimum of | |
c60ebc45 | 1763 | 0 events (e.g. when :option:`iodepth_batch_complete` `=0`). |
f80dba8d | 1764 | |
9d25d068 | 1765 | .. option:: hipri : [pvsync2] |
f80dba8d MT |
1766 | |
1767 | Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority | |
1768 | than normal. | |
1769 | ||
1770 | .. option:: cpuload=int : [cpuio] | |
1771 | ||
da19cdb4 TK |
1772 | Attempt to use the specified percentage of CPU cycles. This is a mandatory |
1773 | option when using cpuio I/O engine. | |
f80dba8d MT |
1774 | |
1775 | .. option:: cpuchunks=int : [cpuio] | |
1776 | ||
1777 | Split the load into cycles of the given time. In microseconds. | |
1778 | ||
1779 | .. option:: exit_on_io_done=bool : [cpuio] | |
1780 | ||
1781 | Detect when I/O threads are done, then exit. | |
1782 | ||
1783 | .. option:: hostname=str : [netsplice] [net] | |
1784 | ||
1785 | The host name or IP address to use for TCP or UDP based I/O. If the job is | |
1786 | a TCP listener or UDP reader, the host name is not used and must be omitted | |
1787 | unless it is a valid UDP multicast address. | |
1788 | ||
1789 | .. option:: namenode=str : [libhdfs] | |
1790 | ||
1791 | The host name or IP address of a HDFS cluster namenode to contact. | |
1792 | ||
1793 | .. option:: port=int | |
1794 | ||
1795 | [netsplice], [net] | |
1796 | ||
1797 | The TCP or UDP port to bind to or connect to. If this is used with | |
1798 | :option:`numjobs` to spawn multiple instances of the same job type, then | |
1799 | this will be the starting port number since fio will use a range of | |
1800 | ports. | |
1801 | ||
1802 | [libhdfs] | |
1803 | ||
1804 | the listening port of the HFDS cluster namenode. | |
1805 | ||
1806 | .. option:: interface=str : [netsplice] [net] | |
1807 | ||
1808 | The IP address of the network interface used to send or receive UDP | |
1809 | multicast. | |
1810 | ||
1811 | .. option:: ttl=int : [netsplice] [net] | |
1812 | ||
1813 | Time-to-live value for outgoing UDP multicast packets. Default: 1. | |
1814 | ||
1815 | .. option:: nodelay=bool : [netsplice] [net] | |
1816 | ||
1817 | Set TCP_NODELAY on TCP connections. | |
1818 | ||
1819 | .. option:: protocol=str : [netsplice] [net] | |
1820 | ||
1821 | .. option:: proto=str : [netsplice] [net] | |
1822 | ||
1823 | The network protocol to use. Accepted values are: | |
1824 | ||
1825 | **tcp** | |
1826 | Transmission control protocol. | |
1827 | **tcpv6** | |
1828 | Transmission control protocol V6. | |
1829 | **udp** | |
1830 | User datagram protocol. | |
1831 | **udpv6** | |
1832 | User datagram protocol V6. | |
1833 | **unix** | |
1834 | UNIX domain socket. | |
1835 | ||
1836 | When the protocol is TCP or UDP, the port must also be given, as well as the | |
1837 | hostname if the job is a TCP listener or UDP reader. For unix sockets, the | |
1838 | normal filename option should be used and the port is invalid. | |
1839 | ||
1840 | .. option:: listen : [net] | |
1841 | ||
1842 | For TCP network connections, tell fio to listen for incoming connections | |
1843 | rather than initiating an outgoing connection. The :option:`hostname` must | |
1844 | be omitted if this option is used. | |
1845 | ||
1846 | .. option:: pingpong : [net] | |
1847 | ||
1848 | Normally a network writer will just continue writing data, and a network | |
1849 | reader will just consume packages. If ``pingpong=1`` is set, a writer will | |
1850 | send its normal payload to the reader, then wait for the reader to send the | |
1851 | same payload back. This allows fio to measure network latencies. The | |
1852 | submission and completion latencies then measure local time spent sending or | |
1853 | receiving, and the completion latency measures how long it took for the | |
1854 | other end to receive and send back. For UDP multicast traffic | |
1855 | ``pingpong=1`` should only be set for a single reader when multiple readers | |
1856 | are listening to the same address. | |
1857 | ||
1858 | .. option:: window_size : [net] | |
1859 | ||
1860 | Set the desired socket buffer size for the connection. | |
1861 | ||
1862 | .. option:: mss : [net] | |
1863 | ||
1864 | Set the TCP maximum segment size (TCP_MAXSEG). | |
1865 | ||
1866 | .. option:: donorname=str : [e4defrag] | |
1867 | ||
1868 | File will be used as a block donor(swap extents between files). | |
1869 | ||
1870 | .. option:: inplace=int : [e4defrag] | |
1871 | ||
1872 | Configure donor file blocks allocation strategy: | |
1873 | ||
1874 | **0** | |
1875 | Default. Preallocate donor's file on init. | |
1876 | **1** | |
1877 | Allocate space immediately inside defragment event, and free right | |
1878 | after event. | |
1879 | ||
1880 | .. option:: clustername=str : [rbd] | |
1881 | ||
1882 | Specifies the name of the Ceph cluster. | |
1883 | ||
1884 | .. option:: rbdname=str : [rbd] | |
1885 | ||
1886 | Specifies the name of the RBD. | |
1887 | ||
1888 | .. option:: pool=str : [rbd] | |
1889 | ||
1890 | Specifies the name of the Ceph pool containing RBD. | |
1891 | ||
1892 | .. option:: clientname=str : [rbd] | |
1893 | ||
1894 | Specifies the username (without the 'client.' prefix) used to access the | |
1895 | Ceph cluster. If the *clustername* is specified, the *clientname* shall be | |
1896 | the full *type.id* string. If no type. prefix is given, fio will add | |
1897 | 'client.' by default. | |
1898 | ||
1899 | .. option:: skip_bad=bool : [mtd] | |
1900 | ||
1901 | Skip operations against known bad blocks. | |
1902 | ||
1903 | .. option:: hdfsdirectory : [libhdfs] | |
1904 | ||
1905 | libhdfs will create chunk in this HDFS directory. | |
1906 | ||
1907 | .. option:: chunk_size : [libhdfs] | |
1908 | ||
1909 | the size of the chunk to use for each file. | |
1910 | ||
1911 | ||
1912 | I/O depth | |
1913 | ~~~~~~~~~ | |
1914 | ||
1915 | .. option:: iodepth=int | |
1916 | ||
1917 | Number of I/O units to keep in flight against the file. Note that | |
1918 | increasing *iodepth* beyond 1 will not affect synchronous ioengines (except | |
c60ebc45 | 1919 | for small degrees when :option:`verify_async` is in use). Even async |
f80dba8d MT |
1920 | engines may impose OS restrictions causing the desired depth not to be |
1921 | achieved. This may happen on Linux when using libaio and not setting | |
1922 | :option:`direct` =1, since buffered I/O is not async on that OS. Keep an | |
1923 | eye on the I/O depth distribution in the fio output to verify that the | |
1924 | achieved depth is as expected. Default: 1. | |
1925 | ||
1926 | .. option:: iodepth_batch_submit=int, iodepth_batch=int | |
1927 | ||
1928 | This defines how many pieces of I/O to submit at once. It defaults to 1 | |
1929 | which means that we submit each I/O as soon as it is available, but can be | |
1930 | raised to submit bigger batches of I/O at the time. If it is set to 0 the | |
1931 | :option:`iodepth` value will be used. | |
1932 | ||
1933 | .. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int | |
1934 | ||
1935 | This defines how many pieces of I/O to retrieve at once. It defaults to 1 | |
1936 | which means that we'll ask for a minimum of 1 I/O in the retrieval process | |
1937 | from the kernel. The I/O retrieval will go on until we hit the limit set by | |
1938 | :option:`iodepth_low`. If this variable is set to 0, then fio will always | |
1939 | check for completed events before queuing more I/O. This helps reduce I/O | |
1940 | latency, at the cost of more retrieval system calls. | |
1941 | ||
1942 | .. option:: iodepth_batch_complete_max=int | |
1943 | ||
1944 | This defines maximum pieces of I/O to retrieve at once. This variable should | |
1945 | be used along with :option:`iodepth_batch_complete_min` =int variable, | |
1946 | specifying the range of min and max amount of I/O which should be | |
1947 | retrieved. By default it is equal to :option:`iodepth_batch_complete_min` | |
1948 | value. | |
1949 | ||
1950 | Example #1:: | |
1951 | ||
1952 | iodepth_batch_complete_min=1 | |
1953 | iodepth_batch_complete_max=<iodepth> | |
1954 | ||
1955 | which means that we will retrieve at least 1 I/O and up to the whole | |
1956 | submitted queue depth. If none of I/O has been completed yet, we will wait. | |
1957 | ||
1958 | Example #2:: | |
1959 | ||
1960 | iodepth_batch_complete_min=0 | |
1961 | iodepth_batch_complete_max=<iodepth> | |
1962 | ||
1963 | which means that we can retrieve up to the whole submitted queue depth, but | |
1964 | if none of I/O has been completed yet, we will NOT wait and immediately exit | |
1965 | the system call. In this example we simply do polling. | |
1966 | ||
1967 | .. option:: iodepth_low=int | |
1968 | ||
1969 | The low water mark indicating when to start filling the queue | |
1970 | again. Defaults to the same as :option:`iodepth`, meaning that fio will | |
1971 | attempt to keep the queue full at all times. If :option:`iodepth` is set to | |
c60ebc45 | 1972 | e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of |
f80dba8d MT |
1973 | 16 requests, it will let the depth drain down to 4 before starting to fill |
1974 | it again. | |
1975 | ||
1976 | .. option:: io_submit_mode=str | |
1977 | ||
1978 | This option controls how fio submits the I/O to the I/O engine. The default | |
1979 | is `inline`, which means that the fio job threads submit and reap I/O | |
1980 | directly. If set to `offload`, the job threads will offload I/O submission | |
1981 | to a dedicated pool of I/O threads. This requires some coordination and thus | |
1982 | has a bit of extra overhead, especially for lower queue depth I/O where it | |
1983 | can increase latencies. The benefit is that fio can manage submission rates | |
1984 | independently of the device completion rates. This avoids skewed latency | |
1985 | reporting if I/O gets back up on the device side (the coordinated omission | |
1986 | problem). | |
1987 | ||
1988 | ||
1989 | I/O rate | |
1990 | ~~~~~~~~ | |
1991 | ||
a881438b | 1992 | .. option:: thinktime=time |
f80dba8d | 1993 | |
f75ede1d SW |
1994 | Stall the job for the specified period of time after an I/O has completed before issuing the |
1995 | next. May be used to simulate processing being done by an application. | |
1996 | When the unit is omitted, the value is given in microseconds. See | |
f80dba8d MT |
1997 | :option:`thinktime_blocks` and :option:`thinktime_spin`. |
1998 | ||
a881438b | 1999 | .. option:: thinktime_spin=time |
f80dba8d MT |
2000 | |
2001 | Only valid if :option:`thinktime` is set - pretend to spend CPU time doing | |
2002 | something with the data received, before falling back to sleeping for the | |
f75ede1d SW |
2003 | rest of the period specified by :option:`thinktime`. When the unit is |
2004 | omitted, the value is given in microseconds. | |
f80dba8d MT |
2005 | |
2006 | .. option:: thinktime_blocks=int | |
2007 | ||
2008 | Only valid if :option:`thinktime` is set - control how many blocks to issue, | |
2009 | before waiting `thinktime` usecs. If not set, defaults to 1 which will make | |
2010 | fio wait `thinktime` usecs after every block. This effectively makes any | |
2011 | queue depth setting redundant, since no more than 1 I/O will be queued | |
2012 | before we have to complete it and do our thinktime. In other words, this | |
2013 | setting effectively caps the queue depth if the latter is larger. | |
71bfa161 | 2014 | |
f80dba8d | 2015 | .. option:: rate=int[,int][,int] |
71bfa161 | 2016 | |
f80dba8d MT |
2017 | Cap the bandwidth used by this job. The number is in bytes/sec, the normal |
2018 | suffix rules apply. Comma-separated values may be specified for reads, | |
2019 | writes, and trims as described in :option:`blocksize`. | |
71bfa161 | 2020 | |
f80dba8d | 2021 | .. option:: rate_min=int[,int][,int] |
71bfa161 | 2022 | |
f80dba8d MT |
2023 | Tell fio to do whatever it can to maintain at least this bandwidth. Failing |
2024 | to meet this requirement will cause the job to exit. Comma-separated values | |
2025 | may be specified for reads, writes, and trims as described in | |
2026 | :option:`blocksize`. | |
71bfa161 | 2027 | |
f80dba8d | 2028 | .. option:: rate_iops=int[,int][,int] |
71bfa161 | 2029 | |
f80dba8d MT |
2030 | Cap the bandwidth to this number of IOPS. Basically the same as |
2031 | :option:`rate`, just specified independently of bandwidth. If the job is | |
2032 | given a block size range instead of a fixed value, the smallest block size | |
2033 | is used as the metric. Comma-separated values may be specified for reads, | |
2034 | writes, and trims as described in :option:`blocksize`. | |
71bfa161 | 2035 | |
f80dba8d | 2036 | .. option:: rate_iops_min=int[,int][,int] |
71bfa161 | 2037 | |
f80dba8d MT |
2038 | If fio doesn't meet this rate of I/O, it will cause the job to exit. |
2039 | Comma-separated values may be specified for reads, writes, and trims as | |
2040 | described in :option:`blocksize`. | |
71bfa161 | 2041 | |
f80dba8d | 2042 | .. option:: rate_process=str |
66c098b8 | 2043 | |
f80dba8d MT |
2044 | This option controls how fio manages rated I/O submissions. The default is |
2045 | `linear`, which submits I/O in a linear fashion with fixed delays between | |
c60ebc45 | 2046 | I/Os that gets adjusted based on I/O completion rates. If this is set to |
f80dba8d MT |
2047 | `poisson`, fio will submit I/O based on a more real world random request |
2048 | flow, known as the Poisson process | |
2049 | (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be | |
2050 | 10^6 / IOPS for the given workload. | |
71bfa161 JA |
2051 | |
2052 | ||
f80dba8d MT |
2053 | I/O latency |
2054 | ~~~~~~~~~~~ | |
71bfa161 | 2055 | |
a881438b | 2056 | .. option:: latency_target=time |
71bfa161 | 2057 | |
f80dba8d | 2058 | If set, fio will attempt to find the max performance point that the given |
f75ede1d SW |
2059 | workload will run at while maintaining a latency below this target. When |
2060 | the unit is omitted, the value is given in microseconds. See | |
2061 | :option:`latency_window` and :option:`latency_percentile`. | |
71bfa161 | 2062 | |
a881438b | 2063 | .. option:: latency_window=time |
71bfa161 | 2064 | |
f80dba8d | 2065 | Used with :option:`latency_target` to specify the sample window that the job |
f75ede1d SW |
2066 | is run at varying queue depths to test the performance. When the unit is |
2067 | omitted, the value is given in microseconds. | |
b4692828 | 2068 | |
f80dba8d | 2069 | .. option:: latency_percentile=float |
71bfa161 | 2070 | |
c60ebc45 | 2071 | The percentage of I/Os that must fall within the criteria specified by |
f80dba8d | 2072 | :option:`latency_target` and :option:`latency_window`. If not set, this |
c60ebc45 | 2073 | defaults to 100.0, meaning that all I/Os must be equal or below to the value |
f80dba8d | 2074 | set by :option:`latency_target`. |
71bfa161 | 2075 | |
a881438b | 2076 | .. option:: max_latency=time |
71bfa161 | 2077 | |
f75ede1d SW |
2078 | If set, fio will exit the job with an ETIMEDOUT error if it exceeds this |
2079 | maximum latency. When the unit is omitted, the value is given in | |
2080 | microseconds. | |
71bfa161 | 2081 | |
f80dba8d | 2082 | .. option:: rate_cycle=int |
71bfa161 | 2083 | |
f80dba8d MT |
2084 | Average bandwidth for :option:`rate` and :option:`rate_min` over this number |
2085 | of milliseconds. | |
71bfa161 | 2086 | |
71bfa161 | 2087 | |
f80dba8d MT |
2088 | I/O replay |
2089 | ~~~~~~~~~~ | |
71bfa161 | 2090 | |
f80dba8d | 2091 | .. option:: write_iolog=str |
c2b1e753 | 2092 | |
f80dba8d MT |
2093 | Write the issued I/O patterns to the specified file. See |
2094 | :option:`read_iolog`. Specify a separate file for each job, otherwise the | |
2095 | iologs will be interspersed and the file may be corrupt. | |
c2b1e753 | 2096 | |
f80dba8d | 2097 | .. option:: read_iolog=str |
71bfa161 | 2098 | |
f80dba8d MT |
2099 | Open an iolog with the specified file name and replay the I/O patterns it |
2100 | contains. This can be used to store a workload and replay it sometime | |
2101 | later. The iolog given may also be a blktrace binary file, which allows fio | |
2102 | to replay a workload captured by :command:`blktrace`. See | |
2103 | :manpage:`blktrace(8)` for how to capture such logging data. For blktrace | |
2104 | replay, the file needs to be turned into a blkparse binary data file first | |
2105 | (``blkparse <device> -o /dev/null -d file_for_fio.bin``). | |
71bfa161 | 2106 | |
f80dba8d | 2107 | .. option:: replay_no_stall=int |
71bfa161 | 2108 | |
f80dba8d MT |
2109 | When replaying I/O with :option:`read_iolog` the default behavior is to |
2110 | attempt to respect the time stamps within the log and replay them with the | |
2111 | appropriate delay between IOPS. By setting this variable fio will not | |
2112 | respect the timestamps and attempt to replay them as fast as possible while | |
2113 | still respecting ordering. The result is the same I/O pattern to a given | |
2114 | device, but different timings. | |
71bfa161 | 2115 | |
f80dba8d | 2116 | .. option:: replay_redirect=str |
b4692828 | 2117 | |
f80dba8d MT |
2118 | While replaying I/O patterns using :option:`read_iolog` the default behavior |
2119 | is to replay the IOPS onto the major/minor device that each IOP was recorded | |
2120 | from. This is sometimes undesirable because on a different machine those | |
2121 | major/minor numbers can map to a different device. Changing hardware on the | |
2122 | same system can also result in a different major/minor mapping. | |
2123 | ``replay_redirect`` causes all IOPS to be replayed onto the single specified | |
2124 | device regardless of the device it was recorded | |
2125 | from. i.e. :option:`replay_redirect` = :file:`/dev/sdc` would cause all I/O | |
2126 | in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means | |
2127 | multiple devices will be replayed onto a single device, if the trace | |
2128 | contains multiple devices. If you want multiple devices to be replayed | |
2129 | concurrently to multiple redirected devices you must blkparse your trace | |
2130 | into separate traces and replay them with independent fio invocations. | |
2131 | Unfortunately this also breaks the strict time ordering between multiple | |
2132 | device accesses. | |
71bfa161 | 2133 | |
f80dba8d | 2134 | .. option:: replay_align=int |
74929ac2 | 2135 | |
f80dba8d MT |
2136 | Force alignment of I/O offsets and lengths in a trace to this power of 2 |
2137 | value. | |
3c54bc46 | 2138 | |
f80dba8d | 2139 | .. option:: replay_scale=int |
3c54bc46 | 2140 | |
f80dba8d | 2141 | Scale sector offsets down by this factor when replaying traces. |
3c54bc46 | 2142 | |
3c54bc46 | 2143 | |
f80dba8d MT |
2144 | Threads, processes and job synchronization |
2145 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
3c54bc46 | 2146 | |
f80dba8d | 2147 | .. option:: thread |
3c54bc46 | 2148 | |
f80dba8d | 2149 | Fio defaults to forking jobs, however if this option is given, fio will use |
79591fa9 TK |
2150 | POSIX Threads function :manpage:`pthread_create(3)` to create threads instead |
2151 | of forking processes. | |
71bfa161 | 2152 | |
f80dba8d | 2153 | .. option:: wait_for=str |
74929ac2 | 2154 | |
f80dba8d MT |
2155 | Specifies the name of the already defined job to wait for. Single waitee |
2156 | name only may be specified. If set, the job won't be started until all | |
2157 | workers of the waitee job are done. | |
74929ac2 | 2158 | |
f80dba8d MT |
2159 | ``wait_for`` operates on the job name basis, so there are a few |
2160 | limitations. First, the waitee must be defined prior to the waiter job | |
2161 | (meaning no forward references). Second, if a job is being referenced as a | |
2162 | waitee, it must have a unique name (no duplicate waitees). | |
74929ac2 | 2163 | |
f80dba8d | 2164 | .. option:: nice=int |
892a6ffc | 2165 | |
f80dba8d | 2166 | Run the job with the given nice value. See man :manpage:`nice(2)`. |
892a6ffc | 2167 | |
f80dba8d MT |
2168 | On Windows, values less than -15 set the process class to "High"; -1 through |
2169 | -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle" | |
2170 | priority class. | |
74929ac2 | 2171 | |
f80dba8d | 2172 | .. option:: prio=int |
71bfa161 | 2173 | |
f80dba8d MT |
2174 | Set the I/O priority value of this job. Linux limits us to a positive value |
2175 | between 0 and 7, with 0 being the highest. See man | |
2176 | :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating | |
2177 | systems since meaning of priority may differ. | |
71bfa161 | 2178 | |
f80dba8d | 2179 | .. option:: prioclass=int |
d59aa780 | 2180 | |
f80dba8d | 2181 | Set the I/O priority class. See man :manpage:`ionice(1)`. |
d59aa780 | 2182 | |
f80dba8d | 2183 | .. option:: cpumask=int |
71bfa161 | 2184 | |
f80dba8d MT |
2185 | Set the CPU affinity of this job. The parameter given is a bitmask of |
2186 | allowed CPU's the job may run on. So if you want the allowed CPUs to be 1 | |
2187 | and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man | |
2188 | :manpage:`sched_setaffinity(2)`. This may not work on all supported | |
2189 | operating systems or kernel versions. This option doesn't work well for a | |
2190 | higher CPU count than what you can store in an integer mask, so it can only | |
2191 | control cpus 1-32. For boxes with larger CPU counts, use | |
2192 | :option:`cpus_allowed`. | |
6d500c2e | 2193 | |
f80dba8d | 2194 | .. option:: cpus_allowed=str |
6d500c2e | 2195 | |
f80dba8d MT |
2196 | Controls the same options as :option:`cpumask`, but it allows a text setting |
2197 | of the permitted CPUs instead. So to use CPUs 1 and 5, you would specify | |
2198 | ``cpus_allowed=1,5``. This options also allows a range of CPUs. Say you | |
2199 | wanted a binding to CPUs 1, 5, and 8-15, you would set | |
2200 | ``cpus_allowed=1,5,8-15``. | |
6d500c2e | 2201 | |
f80dba8d | 2202 | .. option:: cpus_allowed_policy=str |
6d500c2e | 2203 | |
f80dba8d MT |
2204 | Set the policy of how fio distributes the CPUs specified by |
2205 | :option:`cpus_allowed` or cpumask. Two policies are supported: | |
6d500c2e | 2206 | |
f80dba8d MT |
2207 | **shared** |
2208 | All jobs will share the CPU set specified. | |
2209 | **split** | |
2210 | Each job will get a unique CPU from the CPU set. | |
6d500c2e | 2211 | |
f80dba8d MT |
2212 | **shared** is the default behaviour, if the option isn't specified. If |
2213 | **split** is specified, then fio will will assign one cpu per job. If not | |
2214 | enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs | |
2215 | in the set. | |
6d500c2e | 2216 | |
f80dba8d | 2217 | .. option:: numa_cpu_nodes=str |
6d500c2e | 2218 | |
f80dba8d MT |
2219 | Set this job running on specified NUMA nodes' CPUs. The arguments allow |
2220 | comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable | |
2221 | numa options support, fio must be built on a system with libnuma-dev(el) | |
2222 | installed. | |
61b9861d | 2223 | |
f80dba8d | 2224 | .. option:: numa_mem_policy=str |
61b9861d | 2225 | |
f80dba8d MT |
2226 | Set this job's memory policy and corresponding NUMA nodes. Format of the |
2227 | arguments:: | |
5c94b008 | 2228 | |
f80dba8d | 2229 | <mode>[:<nodelist>] |
ce35b1ec | 2230 | |
f80dba8d MT |
2231 | ``mode`` is one of the following memory policy: ``default``, ``prefer``, |
2232 | ``bind``, ``interleave``, ``local`` For ``default`` and ``local`` memory | |
2233 | policy, no node is needed to be specified. For ``prefer``, only one node is | |
2234 | allowed. For ``bind`` and ``interleave``, it allow comma delimited list of | |
2235 | numbers, A-B ranges, or `all`. | |
71bfa161 | 2236 | |
f80dba8d | 2237 | .. option:: cgroup=str |
390b1537 | 2238 | |
f80dba8d MT |
2239 | Add job to this control group. If it doesn't exist, it will be created. The |
2240 | system must have a mounted cgroup blkio mount point for this to work. If | |
2241 | your system doesn't have it mounted, you can do so with:: | |
5af1c6f3 | 2242 | |
f80dba8d | 2243 | # mount -t cgroup -o blkio none /cgroup |
5af1c6f3 | 2244 | |
f80dba8d | 2245 | .. option:: cgroup_weight=int |
5af1c6f3 | 2246 | |
f80dba8d MT |
2247 | Set the weight of the cgroup to this value. See the documentation that comes |
2248 | with the kernel, allowed values are in the range of 100..1000. | |
a086c257 | 2249 | |
f80dba8d | 2250 | .. option:: cgroup_nodelete=bool |
8c07860d | 2251 | |
f80dba8d MT |
2252 | Normally fio will delete the cgroups it has created after the job |
2253 | completion. To override this behavior and to leave cgroups around after the | |
2254 | job completion, set ``cgroup_nodelete=1``. This can be useful if one wants | |
2255 | to inspect various cgroup files after job completion. Default: false. | |
8c07860d | 2256 | |
f80dba8d | 2257 | .. option:: flow_id=int |
8c07860d | 2258 | |
f80dba8d MT |
2259 | The ID of the flow. If not specified, it defaults to being a global |
2260 | flow. See :option:`flow`. | |
1907dbc6 | 2261 | |
f80dba8d | 2262 | .. option:: flow=int |
71bfa161 | 2263 | |
f80dba8d MT |
2264 | Weight in token-based flow control. If this value is used, then there is a |
2265 | 'flow counter' which is used to regulate the proportion of activity between | |
2266 | two or more jobs. Fio attempts to keep this flow counter near zero. The | |
2267 | ``flow`` parameter stands for how much should be added or subtracted to the | |
2268 | flow counter on each iteration of the main I/O loop. That is, if one job has | |
2269 | ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8 | |
2270 | ratio in how much one runs vs the other. | |
71bfa161 | 2271 | |
f80dba8d | 2272 | .. option:: flow_watermark=int |
a31041ea | 2273 | |
f80dba8d MT |
2274 | The maximum value that the absolute value of the flow counter is allowed to |
2275 | reach before the job must wait for a lower value of the counter. | |
82407585 | 2276 | |
f80dba8d | 2277 | .. option:: flow_sleep=int |
82407585 | 2278 | |
f80dba8d MT |
2279 | The period of time, in microseconds, to wait after the flow watermark has |
2280 | been exceeded before retrying operations. | |
82407585 | 2281 | |
f80dba8d | 2282 | .. option:: stonewall, wait_for_previous |
82407585 | 2283 | |
f80dba8d MT |
2284 | Wait for preceding jobs in the job file to exit, before starting this |
2285 | one. Can be used to insert serialization points in the job file. A stone | |
2286 | wall also implies starting a new reporting group, see | |
2287 | :option:`group_reporting`. | |
2288 | ||
2289 | .. option:: exitall | |
2290 | ||
2291 | When one job finishes, terminate the rest. The default is to wait for each | |
2292 | job to finish, sometimes that is not the desired action. | |
2293 | ||
2294 | .. option:: exec_prerun=str | |
2295 | ||
2296 | Before running this job, issue the command specified through | |
2297 | :manpage:`system(3)`. Output is redirected in a file called | |
2298 | :file:`jobname.prerun.txt`. | |
2299 | ||
2300 | .. option:: exec_postrun=str | |
2301 | ||
2302 | After the job completes, issue the command specified though | |
2303 | :manpage:`system(3)`. Output is redirected in a file called | |
2304 | :file:`jobname.postrun.txt`. | |
2305 | ||
2306 | .. option:: uid=int | |
2307 | ||
2308 | Instead of running as the invoking user, set the user ID to this value | |
2309 | before the thread/process does any work. | |
2310 | ||
2311 | .. option:: gid=int | |
2312 | ||
2313 | Set group ID, see :option:`uid`. | |
2314 | ||
2315 | ||
2316 | Verification | |
2317 | ~~~~~~~~~~~~ | |
2318 | ||
2319 | .. option:: verify_only | |
2320 | ||
2321 | Do not perform specified workload, only verify data still matches previous | |
2322 | invocation of this workload. This option allows one to check data multiple | |
2323 | times at a later date without overwriting it. This option makes sense only | |
2324 | for workloads that write data, and does not support workloads with the | |
2325 | :option:`time_based` option set. | |
2326 | ||
2327 | .. option:: do_verify=bool | |
2328 | ||
2329 | Run the verify phase after a write phase. Only valid if :option:`verify` is | |
2330 | set. Default: true. | |
2331 | ||
2332 | .. option:: verify=str | |
2333 | ||
2334 | If writing to a file, fio can verify the file contents after each iteration | |
2335 | of the job. Each verification method also implies verification of special | |
2336 | header, which is written to the beginning of each block. This header also | |
2337 | includes meta information, like offset of the block, block number, timestamp | |
2338 | when block was written, etc. :option:`verify` can be combined with | |
2339 | :option:`verify_pattern` option. The allowed values are: | |
2340 | ||
2341 | **md5** | |
2342 | Use an md5 sum of the data area and store it in the header of | |
2343 | each block. | |
2344 | ||
2345 | **crc64** | |
2346 | Use an experimental crc64 sum of the data area and store it in the | |
2347 | header of each block. | |
2348 | ||
2349 | **crc32c** | |
2350 | Use a crc32c sum of the data area and store it in the header of each | |
2351 | block. | |
2352 | ||
2353 | **crc32c-intel** | |
2354 | Use hardware assisted crc32c calculation provided on SSE4.2 enabled | |
2355 | processors. Falls back to regular software crc32c, if not supported | |
2356 | by the system. | |
2357 | ||
2358 | **crc32** | |
2359 | Use a crc32 sum of the data area and store it in the header of each | |
2360 | block. | |
2361 | ||
2362 | **crc16** | |
2363 | Use a crc16 sum of the data area and store it in the header of each | |
2364 | block. | |
2365 | ||
2366 | **crc7** | |
2367 | Use a crc7 sum of the data area and store it in the header of each | |
2368 | block. | |
2369 | ||
2370 | **xxhash** | |
2371 | Use xxhash as the checksum function. Generally the fastest software | |
2372 | checksum that fio supports. | |
2373 | ||
2374 | **sha512** | |
2375 | Use sha512 as the checksum function. | |
2376 | ||
2377 | **sha256** | |
2378 | Use sha256 as the checksum function. | |
2379 | ||
2380 | **sha1** | |
2381 | Use optimized sha1 as the checksum function. | |
82407585 | 2382 | |
ae3a5acc JA |
2383 | **sha3-224** |
2384 | Use optimized sha3-224 as the checksum function. | |
2385 | ||
2386 | **sha3-256** | |
2387 | Use optimized sha3-256 as the checksum function. | |
2388 | ||
2389 | **sha3-384** | |
2390 | Use optimized sha3-384 as the checksum function. | |
2391 | ||
2392 | **sha3-512** | |
2393 | Use optimized sha3-512 as the checksum function. | |
2394 | ||
f80dba8d MT |
2395 | **meta** |
2396 | This option is deprecated, since now meta information is included in | |
2397 | generic verification header and meta verification happens by | |
2398 | default. For detailed information see the description of the | |
2399 | :option:`verify` setting. This option is kept because of | |
2400 | compatibility's sake with old configurations. Do not use it. | |
2401 | ||
2402 | **pattern** | |
2403 | Verify a strict pattern. Normally fio includes a header with some | |
2404 | basic information and checksumming, but if this option is set, only | |
2405 | the specific pattern set with :option:`verify_pattern` is verified. | |
2406 | ||
2407 | **null** | |
2408 | Only pretend to verify. Useful for testing internals with | |
2409 | :option:`ioengine` `=null`, not for much else. | |
2410 | ||
2411 | This option can be used for repeated burn-in tests of a system to make sure | |
2412 | that the written data is also correctly read back. If the data direction | |
2413 | given is a read or random read, fio will assume that it should verify a | |
2414 | previously written file. If the data direction includes any form of write, | |
2415 | the verify will be of the newly written data. | |
2416 | ||
2417 | .. option:: verifysort=bool | |
2418 | ||
2419 | If true, fio will sort written verify blocks when it deems it faster to read | |
2420 | them back in a sorted manner. This is often the case when overwriting an | |
2421 | existing file, since the blocks are already laid out in the file system. You | |
2422 | can ignore this option unless doing huge amounts of really fast I/O where | |
2423 | the red-black tree sorting CPU time becomes significant. Default: true. | |
2424 | ||
2425 | .. option:: verifysort_nr=int | |
2426 | ||
2427 | Pre-load and sort verify blocks for a read workload. | |
2428 | ||
2429 | .. option:: verify_offset=int | |
2430 | ||
2431 | Swap the verification header with data somewhere else in the block before | |
2432 | writing. It is swapped back before verifying. | |
2433 | ||
2434 | .. option:: verify_interval=int | |
2435 | ||
2436 | Write the verification header at a finer granularity than the | |
2437 | :option:`blocksize`. It will be written for chunks the size of | |
2438 | ``verify_interval``. :option:`blocksize` should divide this evenly. | |
2439 | ||
2440 | .. option:: verify_pattern=str | |
2441 | ||
2442 | If set, fio will fill the I/O buffers with this pattern. Fio defaults to | |
2443 | filling with totally random bytes, but sometimes it's interesting to fill | |
2444 | with a known pattern for I/O verification purposes. Depending on the width | |
2445 | of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time(it can | |
2446 | be either a decimal or a hex number). The ``verify_pattern`` if larger than | |
2447 | a 32-bit quantity has to be a hex number that starts with either "0x" or | |
2448 | "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o | |
2449 | format, which means that for each block offset will be written and then | |
2450 | verified back, e.g.:: | |
61b9861d RP |
2451 | |
2452 | verify_pattern=%o | |
2453 | ||
f80dba8d MT |
2454 | Or use combination of everything:: |
2455 | ||
61b9861d | 2456 | verify_pattern=0xff%o"abcd"-12 |
e28218f3 | 2457 | |
f80dba8d MT |
2458 | .. option:: verify_fatal=bool |
2459 | ||
2460 | Normally fio will keep checking the entire contents before quitting on a | |
2461 | block verification failure. If this option is set, fio will exit the job on | |
2462 | the first observed failure. Default: false. | |
2463 | ||
2464 | .. option:: verify_dump=bool | |
2465 | ||
2466 | If set, dump the contents of both the original data block and the data block | |
2467 | we read off disk to files. This allows later analysis to inspect just what | |
2468 | kind of data corruption occurred. Off by default. | |
2469 | ||
2470 | .. option:: verify_async=int | |
2471 | ||
2472 | Fio will normally verify I/O inline from the submitting thread. This option | |
2473 | takes an integer describing how many async offload threads to create for I/O | |
2474 | verification instead, causing fio to offload the duty of verifying I/O | |
2475 | contents to one or more separate threads. If using this offload option, even | |
2476 | sync I/O engines can benefit from using an :option:`iodepth` setting higher | |
2477 | than 1, as it allows them to have I/O in flight while verifies are running. | |
d7e6ea1c | 2478 | Defaults to 0 async threads, i.e. verification is not asynchronous. |
f80dba8d MT |
2479 | |
2480 | .. option:: verify_async_cpus=str | |
2481 | ||
2482 | Tell fio to set the given CPU affinity on the async I/O verification | |
2483 | threads. See :option:`cpus_allowed` for the format used. | |
2484 | ||
2485 | .. option:: verify_backlog=int | |
2486 | ||
2487 | Fio will normally verify the written contents of a job that utilizes verify | |
2488 | once that job has completed. In other words, everything is written then | |
2489 | everything is read back and verified. You may want to verify continually | |
2490 | instead for a variety of reasons. Fio stores the meta data associated with | |
2491 | an I/O block in memory, so for large verify workloads, quite a bit of memory | |
2492 | would be used up holding this meta data. If this option is enabled, fio will | |
2493 | write only N blocks before verifying these blocks. | |
2494 | ||
2495 | .. option:: verify_backlog_batch=int | |
2496 | ||
2497 | Control how many blocks fio will verify if :option:`verify_backlog` is | |
2498 | set. If not set, will default to the value of :option:`verify_backlog` | |
2499 | (meaning the entire queue is read back and verified). If | |
2500 | ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all | |
2501 | blocks will be verified, if ``verify_backlog_batch`` is larger than | |
2502 | :option:`verify_backlog`, some blocks will be verified more than once. | |
2503 | ||
2504 | .. option:: verify_state_save=bool | |
2505 | ||
2506 | When a job exits during the write phase of a verify workload, save its | |
2507 | current state. This allows fio to replay up until that point, if the verify | |
2508 | state is loaded for the verify read phase. The format of the filename is, | |
2509 | roughly:: | |
2510 | ||
2511 | <type>-<jobname>-<jobindex>-verify.state. | |
2512 | ||
2513 | <type> is "local" for a local run, "sock" for a client/server socket | |
2514 | connection, and "ip" (192.168.0.1, for instance) for a networked | |
d7e6ea1c | 2515 | client/server connection. Defaults to true. |
f80dba8d MT |
2516 | |
2517 | .. option:: verify_state_load=bool | |
2518 | ||
2519 | If a verify termination trigger was used, fio stores the current write state | |
2520 | of each thread. This can be used at verification time so that fio knows how | |
2521 | far it should verify. Without this information, fio will run a full | |
2522 | verification pass, according to the settings in the job file used. | |
2523 | ||
2524 | .. option:: trim_percentage=int | |
2525 | ||
2526 | Number of verify blocks to discard/trim. | |
2527 | ||
2528 | .. option:: trim_verify_zero=bool | |
2529 | ||
2530 | Verify that trim/discarded blocks are returned as zeroes. | |
2531 | ||
2532 | .. option:: trim_backlog=int | |
2533 | ||
2534 | Verify that trim/discarded blocks are returned as zeroes. | |
2535 | ||
2536 | .. option:: trim_backlog_batch=int | |
2537 | ||
2538 | Trim this number of I/O blocks. | |
2539 | ||
2540 | .. option:: experimental_verify=bool | |
2541 | ||
2542 | Enable experimental verification. | |
2543 | ||
2544 | ||
2545 | Steady state | |
2546 | ~~~~~~~~~~~~ | |
2547 | ||
2548 | .. option:: steadystate=str:float, ss=str:float | |
2549 | ||
2550 | Define the criterion and limit for assessing steady state performance. The | |
2551 | first parameter designates the criterion whereas the second parameter sets | |
2552 | the threshold. When the criterion falls below the threshold for the | |
2553 | specified duration, the job will stop. For example, `iops_slope:0.1%` will | |
2554 | direct fio to terminate the job when the least squares regression slope | |
2555 | falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled | |
2556 | this will apply to all jobs in the group. Below is the list of available | |
2557 | steady state assessment criteria. All assessments are carried out using only | |
2558 | data from the rolling collection window. Threshold limits can be expressed | |
2559 | as a fixed value or as a percentage of the mean in the collection window. | |
2560 | ||
2561 | **iops** | |
2562 | Collect IOPS data. Stop the job if all individual IOPS measurements | |
2563 | are within the specified limit of the mean IOPS (e.g., ``iops:2`` | |
2564 | means that all individual IOPS values must be within 2 of the mean, | |
2565 | whereas ``iops:0.2%`` means that all individual IOPS values must be | |
2566 | within 0.2% of the mean IOPS to terminate the job). | |
2567 | ||
2568 | **iops_slope** | |
2569 | Collect IOPS data and calculate the least squares regression | |
2570 | slope. Stop the job if the slope falls below the specified limit. | |
2571 | ||
2572 | **bw** | |
2573 | Collect bandwidth data. Stop the job if all individual bandwidth | |
2574 | measurements are within the specified limit of the mean bandwidth. | |
2575 | ||
2576 | **bw_slope** | |
2577 | Collect bandwidth data and calculate the least squares regression | |
2578 | slope. Stop the job if the slope falls below the specified limit. | |
2579 | ||
2580 | .. option:: steadystate_duration=time, ss_dur=time | |
2581 | ||
2582 | A rolling window of this duration will be used to judge whether steady state | |
2583 | has been reached. Data will be collected once per second. The default is 0 | |
f75ede1d SW |
2584 | which disables steady state detection. When the unit is omitted, the |
2585 | value is given in seconds. | |
f80dba8d MT |
2586 | |
2587 | .. option:: steadystate_ramp_time=time, ss_ramp=time | |
2588 | ||
2589 | Allow the job to run for the specified duration before beginning data | |
2590 | collection for checking the steady state job termination criterion. The | |
f75ede1d | 2591 | default is 0. When the unit is omitted, the value is given in seconds. |
f80dba8d MT |
2592 | |
2593 | ||
2594 | Measurements and reporting | |
2595 | ~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
2596 | ||
2597 | .. option:: per_job_logs=bool | |
2598 | ||
2599 | If set, this generates bw/clat/iops log with per file private filenames. If | |
2600 | not set, jobs with identical names will share the log filename. Default: | |
2601 | true. | |
2602 | ||
2603 | .. option:: group_reporting | |
2604 | ||
2605 | It may sometimes be interesting to display statistics for groups of jobs as | |
2606 | a whole instead of for each individual job. This is especially true if | |
2607 | :option:`numjobs` is used; looking at individual thread/process output | |
2608 | quickly becomes unwieldy. To see the final report per-group instead of | |
2609 | per-job, use :option:`group_reporting`. Jobs in a file will be part of the | |
2610 | same reporting group, unless if separated by a :option:`stonewall`, or by | |
2611 | using :option:`new_group`. | |
2612 | ||
2613 | .. option:: new_group | |
2614 | ||
2615 | Start a new reporting group. See: :option:`group_reporting`. If not given, | |
2616 | all jobs in a file will be part of the same reporting group, unless | |
2617 | separated by a :option:`stonewall`. | |
2618 | ||
8243be59 JA |
2619 | .. option:: stats |
2620 | ||
2621 | By default, fio collects and shows final output results for all jobs | |
2622 | that run. If this option is set to 0, then fio will ignore it in | |
2623 | the final stat output. | |
2624 | ||
f80dba8d MT |
2625 | .. option:: write_bw_log=str |
2626 | ||
2627 | If given, write a bandwidth log for this job. Can be used to store data of | |
2628 | the bandwidth of the jobs in their lifetime. The included | |
2629 | :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these | |
2630 | text files into nice graphs. See :option:`write_lat_log` for behaviour of | |
2631 | given filename. For this option, the postfix is :file:`_bw.x.log`, where `x` | |
2632 | is the index of the job (`1..N`, where `N` is the number of jobs). If | |
2633 | :option:`per_job_logs` is false, then the filename will not include the job | |
2634 | index. See `Log File Formats`_. | |
2635 | ||
2636 | .. option:: write_lat_log=str | |
2637 | ||
2638 | Same as :option:`write_bw_log`, except that this option stores I/O | |
2639 | submission, completion, and total latencies instead. If no filename is given | |
2640 | with this option, the default filename of :file:`jobname_type.log` is | |
2641 | used. Even if the filename is given, fio will still append the type of | |
2642 | log. So if one specifies:: | |
e3cedca7 JA |
2643 | |
2644 | write_lat_log=foo | |
2645 | ||
f80dba8d MT |
2646 | The actual log names will be :file:`foo_slat.x.log`, :file:`foo_clat.x.log`, |
2647 | and :file:`foo_lat.x.log`, where `x` is the index of the job (1..N, where N | |
2648 | is the number of jobs). This helps :command:`fio_generate_plot` find the | |
2649 | logs automatically. If :option:`per_job_logs` is false, then the filename | |
2650 | will not include the job index. See `Log File Formats`_. | |
be4ecfdf | 2651 | |
f80dba8d | 2652 | .. option:: write_hist_log=str |
06842027 | 2653 | |
f80dba8d MT |
2654 | Same as :option:`write_lat_log`, but writes I/O completion latency |
2655 | histograms. If no filename is given with this option, the default filename | |
2656 | of :file:`jobname_clat_hist.x.log` is used, where `x` is the index of the | |
2657 | job (1..N, where `N` is the number of jobs). Even if the filename is given, | |
2658 | fio will still append the type of log. If :option:`per_job_logs` is false, | |
2659 | then the filename will not include the job index. See `Log File Formats`_. | |
06842027 | 2660 | |
f80dba8d | 2661 | .. option:: write_iops_log=str |
06842027 | 2662 | |
f80dba8d MT |
2663 | Same as :option:`write_bw_log`, but writes IOPS. If no filename is given |
2664 | with this option, the default filename of :file:`jobname_type.x.log` is | |
2665 | used,where `x` is the index of the job (1..N, where `N` is the number of | |
2666 | jobs). Even if the filename is given, fio will still append the type of | |
2667 | log. If :option:`per_job_logs` is false, then the filename will not include | |
2668 | the job index. See `Log File Formats`_. | |
06842027 | 2669 | |
f80dba8d | 2670 | .. option:: log_avg_msec=int |
06842027 | 2671 | |
f80dba8d MT |
2672 | By default, fio will log an entry in the iops, latency, or bw log for every |
2673 | I/O that completes. When writing to the disk log, that can quickly grow to a | |
2674 | very large size. Setting this option makes fio average the each log entry | |
2675 | over the specified period of time, reducing the resolution of the log. See | |
2676 | :option:`log_max_value` as well. Defaults to 0, logging all entries. | |
06842027 | 2677 | |
f80dba8d | 2678 | .. option:: log_hist_msec=int |
06842027 | 2679 | |
f80dba8d MT |
2680 | Same as :option:`log_avg_msec`, but logs entries for completion latency |
2681 | histograms. Computing latency percentiles from averages of intervals using | |
c60ebc45 | 2682 | :option:`log_avg_msec` is inaccurate. Setting this option makes fio log |
f80dba8d MT |
2683 | histogram entries over the specified period of time, reducing log sizes for |
2684 | high IOPS devices while retaining percentile accuracy. See | |
2685 | :option:`log_hist_coarseness` as well. Defaults to 0, meaning histogram | |
2686 | logging is disabled. | |
06842027 | 2687 | |
f80dba8d | 2688 | .. option:: log_hist_coarseness=int |
06842027 | 2689 | |
f80dba8d MT |
2690 | Integer ranging from 0 to 6, defining the coarseness of the resolution of |
2691 | the histogram logs enabled with :option:`log_hist_msec`. For each increment | |
2692 | in coarseness, fio outputs half as many bins. Defaults to 0, for which | |
2693 | histogram logs contain 1216 latency bins. See `Log File Formats`_. | |
8b28bd41 | 2694 | |
f80dba8d | 2695 | .. option:: log_max_value=bool |
66c098b8 | 2696 | |
f80dba8d MT |
2697 | If :option:`log_avg_msec` is set, fio logs the average over that window. If |
2698 | you instead want to log the maximum value, set this option to 1. Defaults to | |
2699 | 0, meaning that averaged values are logged. | |
a696fa2a | 2700 | |
f80dba8d | 2701 | .. option:: log_offset=int |
a696fa2a | 2702 | |
f80dba8d MT |
2703 | If this is set, the iolog options will include the byte offset for the I/O |
2704 | entry as well as the other data values. | |
71bfa161 | 2705 | |
f80dba8d | 2706 | .. option:: log_compression=int |
7de87099 | 2707 | |
f80dba8d MT |
2708 | If this is set, fio will compress the I/O logs as it goes, to keep the |
2709 | memory footprint lower. When a log reaches the specified size, that chunk is | |
2710 | removed and compressed in the background. Given that I/O logs are fairly | |
2711 | highly compressible, this yields a nice memory savings for longer runs. The | |
2712 | downside is that the compression will consume some background CPU cycles, so | |
2713 | it may impact the run. This, however, is also true if the logging ends up | |
2714 | consuming most of the system memory. So pick your poison. The I/O logs are | |
2715 | saved normally at the end of a run, by decompressing the chunks and storing | |
2716 | them in the specified log file. This feature depends on the availability of | |
2717 | zlib. | |
e0b0d892 | 2718 | |
f80dba8d | 2719 | .. option:: log_compression_cpus=str |
e0b0d892 | 2720 | |
f80dba8d MT |
2721 | Define the set of CPUs that are allowed to handle online log compression for |
2722 | the I/O jobs. This can provide better isolation between performance | |
2723 | sensitive jobs, and background compression work. | |
9e684a49 | 2724 | |
f80dba8d | 2725 | .. option:: log_store_compressed=bool |
9e684a49 | 2726 | |
f80dba8d MT |
2727 | If set, fio will store the log files in a compressed format. They can be |
2728 | decompressed with fio, using the :option:`--inflate-log` command line | |
2729 | parameter. The files will be stored with a :file:`.fz` suffix. | |
9e684a49 | 2730 | |
f80dba8d | 2731 | .. option:: log_unix_epoch=bool |
9e684a49 | 2732 | |
f80dba8d MT |
2733 | If set, fio will log Unix timestamps to the log files produced by enabling |
2734 | write_type_log for each log type, instead of the default zero-based | |
2735 | timestamps. | |
2736 | ||
2737 | .. option:: block_error_percentiles=bool | |
2738 | ||
2739 | If set, record errors in trim block-sized units from writes and trims and | |
2740 | output a histogram of how many trims it took to get to errors, and what kind | |
2741 | of error was encountered. | |
2742 | ||
2743 | .. option:: bwavgtime=int | |
2744 | ||
2745 | Average the calculated bandwidth over the given time. Value is specified in | |
2746 | milliseconds. If the job also does bandwidth logging through | |
2747 | :option:`write_bw_log`, then the minimum of this option and | |
2748 | :option:`log_avg_msec` will be used. Default: 500ms. | |
2749 | ||
2750 | .. option:: iopsavgtime=int | |
2751 | ||
2752 | Average the calculated IOPS over the given time. Value is specified in | |
2753 | milliseconds. If the job also does IOPS logging through | |
2754 | :option:`write_iops_log`, then the minimum of this option and | |
2755 | :option:`log_avg_msec` will be used. Default: 500ms. | |
2756 | ||
2757 | .. option:: disk_util=bool | |
2758 | ||
2759 | Generate disk utilization statistics, if the platform supports it. | |
2760 | Default: true. | |
2761 | ||
2762 | .. option:: disable_lat=bool | |
2763 | ||
2764 | Disable measurements of total latency numbers. Useful only for cutting back | |
2765 | the number of calls to :manpage:`gettimeofday(2)`, as that does impact | |
2766 | performance at really high IOPS rates. Note that to really get rid of a | |
2767 | large amount of these calls, this option must be used with | |
f75ede1d | 2768 | :option:`disable_slat` and :option:`disable_bw_measurement` as well. |
f80dba8d MT |
2769 | |
2770 | .. option:: disable_clat=bool | |
2771 | ||
2772 | Disable measurements of completion latency numbers. See | |
2773 | :option:`disable_lat`. | |
2774 | ||
2775 | .. option:: disable_slat=bool | |
2776 | ||
2777 | Disable measurements of submission latency numbers. See | |
2778 | :option:`disable_slat`. | |
2779 | ||
f75ede1d | 2780 | .. option:: disable_bw_measurement=bool, disable_bw=bool |
f80dba8d MT |
2781 | |
2782 | Disable measurements of throughput/bandwidth numbers. See | |
2783 | :option:`disable_lat`. | |
2784 | ||
2785 | .. option:: clat_percentiles=bool | |
2786 | ||
2787 | Enable the reporting of percentiles of completion latencies. | |
2788 | ||
2789 | .. option:: percentile_list=float_list | |
2790 | ||
2791 | Overwrite the default list of percentiles for completion latencies and the | |
2792 | block error histogram. Each number is a floating number in the range | |
2793 | (0,100], and the maximum length of the list is 20. Use ``:`` to separate the | |
2794 | numbers, and list the numbers in ascending order. For example, | |
2795 | ``--percentile_list=99.5:99.9`` will cause fio to report the values of | |
2796 | completion latency below which 99.5% and 99.9% of the observed latencies | |
2797 | fell, respectively. | |
2798 | ||
2799 | ||
2800 | Error handling | |
2801 | ~~~~~~~~~~~~~~ | |
2802 | ||
2803 | .. option:: exitall_on_error | |
2804 | ||
2805 | When one job finishes in error, terminate the rest. The default is to wait | |
2806 | for each job to finish. | |
2807 | ||
2808 | .. option:: continue_on_error=str | |
2809 | ||
2810 | Normally fio will exit the job on the first observed failure. If this option | |
2811 | is set, fio will continue the job when there is a 'non-fatal error' (EIO or | |
2812 | EILSEQ) until the runtime is exceeded or the I/O size specified is | |
2813 | completed. If this option is used, there are two more stats that are | |
2814 | appended, the total error count and the first error. The error field given | |
2815 | in the stats is the first error that was hit during the run. | |
2816 | ||
2817 | The allowed values are: | |
2818 | ||
2819 | **none** | |
2820 | Exit on any I/O or verify errors. | |
2821 | ||
2822 | **read** | |
2823 | Continue on read errors, exit on all others. | |
2824 | ||
2825 | **write** | |
2826 | Continue on write errors, exit on all others. | |
2827 | ||
2828 | **io** | |
2829 | Continue on any I/O error, exit on all others. | |
2830 | ||
2831 | **verify** | |
2832 | Continue on verify errors, exit on all others. | |
2833 | ||
2834 | **all** | |
2835 | Continue on all errors. | |
2836 | ||
2837 | **0** | |
2838 | Backward-compatible alias for 'none'. | |
2839 | ||
2840 | **1** | |
2841 | Backward-compatible alias for 'all'. | |
2842 | ||
2843 | .. option:: ignore_error=str | |
2844 | ||
2845 | Sometimes you want to ignore some errors during test in that case you can | |
a35ef7cb TK |
2846 | specify error list for each error type, instead of only being able to |
2847 | ignore the default 'non-fatal error' using :option:`continue_on_error`. | |
f80dba8d MT |
2848 | ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for |
2849 | given error type is separated with ':'. Error may be symbol ('ENOSPC', | |
2850 | 'ENOMEM') or integer. Example:: | |
2851 | ||
2852 | ignore_error=EAGAIN,ENOSPC:122 | |
2853 | ||
2854 | This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from | |
a35ef7cb TK |
2855 | WRITE. This option works by overriding :option:`continue_on_error` with |
2856 | the list of errors for each error type if any. | |
f80dba8d MT |
2857 | |
2858 | .. option:: error_dump=bool | |
2859 | ||
2860 | If set dump every error even if it is non fatal, true by default. If | |
2861 | disabled only fatal error will be dumped. | |
2862 | ||
f75ede1d SW |
2863 | Running predefined workloads |
2864 | ---------------------------- | |
2865 | ||
2866 | Fio includes predefined profiles that mimic the I/O workloads generated by | |
2867 | other tools. | |
2868 | ||
2869 | .. option:: profile=str | |
2870 | ||
2871 | The predefined workload to run. Current profiles are: | |
2872 | ||
2873 | **tiobench** | |
2874 | Threaded I/O bench (tiotest/tiobench) like workload. | |
2875 | ||
2876 | **act** | |
2877 | Aerospike Certification Tool (ACT) like workload. | |
2878 | ||
2879 | To view a profile's additional options use :option:`--cmdhelp` after specifying | |
2880 | the profile. For example:: | |
2881 | ||
2882 | $ fio --profile=act --cmdhelp | |
2883 | ||
2884 | Act profile options | |
2885 | ~~~~~~~~~~~~~~~~~~~ | |
2886 | ||
2887 | .. option:: device-names=str | |
2888 | :noindex: | |
2889 | ||
2890 | Devices to use. | |
2891 | ||
2892 | .. option:: load=int | |
2893 | :noindex: | |
2894 | ||
2895 | ACT load multiplier. Default: 1. | |
2896 | ||
2897 | .. option:: test-duration=time | |
2898 | :noindex: | |
2899 | ||
2900 | How long the entire test takes to run. Default: 24h. | |
2901 | ||
2902 | .. option:: threads-per-queue=int | |
2903 | :noindex: | |
2904 | ||
2905 | Number of read IO threads per device. Default: 8. | |
2906 | ||
2907 | .. option:: read-req-num-512-blocks=int | |
2908 | :noindex: | |
2909 | ||
2910 | Number of 512B blocks to read at the time. Default: 3. | |
2911 | ||
2912 | .. option:: large-block-op-kbytes=int | |
2913 | :noindex: | |
2914 | ||
2915 | Size of large block ops in KiB (writes). Default: 131072. | |
2916 | ||
2917 | .. option:: prep | |
2918 | :noindex: | |
2919 | ||
2920 | Set to run ACT prep phase. | |
2921 | ||
2922 | Tiobench profile options | |
2923 | ~~~~~~~~~~~~~~~~~~~~~~~~ | |
2924 | ||
2925 | .. option:: size=str | |
2926 | :noindex: | |
2927 | ||
2928 | Size in MiB | |
2929 | ||
2930 | .. option:: block=int | |
2931 | :noindex: | |
2932 | ||
2933 | Block size in bytes. Default: 4096. | |
2934 | ||
2935 | .. option:: numruns=int | |
2936 | :noindex: | |
2937 | ||
2938 | Number of runs. | |
2939 | ||
2940 | .. option:: dir=str | |
2941 | :noindex: | |
2942 | ||
2943 | Test directory. | |
2944 | ||
2945 | .. option:: threads=int | |
2946 | :noindex: | |
2947 | ||
2948 | Number of threads. | |
f80dba8d MT |
2949 | |
2950 | Interpreting the output | |
2951 | ----------------------- | |
2952 | ||
2953 | Fio spits out a lot of output. While running, fio will display the status of the | |
2954 | jobs created. An example of that would be:: | |
2955 | ||
9d25d068 | 2956 | Jobs: 1 (f=1): [_(1),M(1)][24.8%][r=20.5MiB/s,w=23.5MiB/s][r=82,w=94 IOPS][eta 01m:31s] |
f80dba8d MT |
2957 | |
2958 | The characters inside the square brackets denote the current status of each | |
2959 | thread. The possible values (in typical life cycle order) are: | |
2960 | ||
2961 | +------+-----+-----------------------------------------------------------+ | |
2962 | | Idle | Run | | | |
2963 | +======+=====+===========================================================+ | |
2964 | | P | | Thread setup, but not started. | | |
2965 | +------+-----+-----------------------------------------------------------+ | |
2966 | | C | | Thread created. | | |
2967 | +------+-----+-----------------------------------------------------------+ | |
2968 | | I | | Thread initialized, waiting or generating necessary data. | | |
2969 | +------+-----+-----------------------------------------------------------+ | |
2970 | | | p | Thread running pre-reading file(s). | | |
2971 | +------+-----+-----------------------------------------------------------+ | |
2972 | | | R | Running, doing sequential reads. | | |
2973 | +------+-----+-----------------------------------------------------------+ | |
2974 | | | r | Running, doing random reads. | | |
2975 | +------+-----+-----------------------------------------------------------+ | |
2976 | | | W | Running, doing sequential writes. | | |
2977 | +------+-----+-----------------------------------------------------------+ | |
2978 | | | w | Running, doing random writes. | | |
2979 | +------+-----+-----------------------------------------------------------+ | |
2980 | | | M | Running, doing mixed sequential reads/writes. | | |
2981 | +------+-----+-----------------------------------------------------------+ | |
2982 | | | m | Running, doing mixed random reads/writes. | | |
2983 | +------+-----+-----------------------------------------------------------+ | |
2984 | | | F | Running, currently waiting for :manpage:`fsync(2)` | | |
2985 | +------+-----+-----------------------------------------------------------+ | |
2986 | | | V | Running, doing verification of written data. | | |
2987 | +------+-----+-----------------------------------------------------------+ | |
2988 | | E | | Thread exited, not reaped by main thread yet. | | |
2989 | +------+-----+-----------------------------------------------------------+ | |
2990 | | _ | | Thread reaped, or | | |
2991 | +------+-----+-----------------------------------------------------------+ | |
2992 | | X | | Thread reaped, exited with an error. | | |
2993 | +------+-----+-----------------------------------------------------------+ | |
2994 | | K | | Thread reaped, exited due to signal. | | |
2995 | +------+-----+-----------------------------------------------------------+ | |
2996 | ||
2997 | Fio will condense the thread string as not to take up more space on the command | |
2998 | line as is needed. For instance, if you have 10 readers and 10 writers running, | |
2999 | the output would look like this:: | |
3000 | ||
9d25d068 | 3001 | Jobs: 20 (f=20): [R(10),W(10)][4.0%][r=20.5MiB/s,w=23.5MiB/s][r=82,w=94 IOPS][eta 57m:36s] |
f80dba8d MT |
3002 | |
3003 | Fio will still maintain the ordering, though. So the above means that jobs 1..10 | |
3004 | are readers, and 11..20 are writers. | |
3005 | ||
3006 | The other values are fairly self explanatory -- number of threads currently | |
9d25d068 SW |
3007 | running and doing I/O, the number of currently open files (f=), the rate of I/O |
3008 | since last check (read speed listed first, then write speed and optionally trim | |
3009 | speed), and the estimated completion percentage and time for the current | |
f80dba8d MT |
3010 | running group. It's impossible to estimate runtime of the following groups (if |
3011 | any). Note that the string is displayed in order, so it's possible to tell which | |
3012 | of the jobs are currently doing what. The first character is the first job | |
3013 | defined in the job file, and so forth. | |
3014 | ||
3015 | When fio is done (or interrupted by :kbd:`ctrl-c`), it will show the data for | |
3016 | each thread, group of threads, and disks in that order. For each data direction, | |
3017 | the output looks like:: | |
3018 | ||
3019 | Client1 (g=0): err= 0: | |
3020 | write: io= 32MiB, bw= 666KiB/s, iops=89 , runt= 50320msec | |
3021 | slat (msec): min= 0, max= 136, avg= 0.03, stdev= 1.92 | |
3022 | clat (msec): min= 0, max= 631, avg=48.50, stdev=86.82 | |
3023 | bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, stdev=681.68 | |
3024 | cpu : usr=1.49%, sys=0.25%, ctx=7969, majf=0, minf=17 | |
3025 | IO depths : 1=0.1%, 2=0.3%, 4=0.5%, 8=99.0%, 16=0.0%, 32=0.0%, >32=0.0% | |
3026 | submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% | |
3027 | complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% | |
3028 | issued r/w: total=0/32768, short=0/0 | |
3029 | lat (msec): 2=1.6%, 4=0.0%, 10=3.2%, 20=12.8%, 50=38.4%, 100=24.8%, | |
3030 | lat (msec): 250=15.2%, 500=0.0%, 750=0.0%, 1000=0.0%, >=2048=0.0% | |
71bfa161 JA |
3031 | |
3032 | The client number is printed, along with the group id and error of that | |
f80dba8d MT |
3033 | thread. Below is the I/O statistics, here for writes. In the order listed, they |
3034 | denote: | |
3035 | ||
3036 | **io** | |
3037 | Number of megabytes I/O performed. | |
3038 | ||
3039 | **bw** | |
3040 | Average bandwidth rate. | |
3041 | ||
3042 | **iops** | |
c60ebc45 | 3043 | Average I/Os performed per second. |
f80dba8d MT |
3044 | |
3045 | **runt** | |
3046 | The runtime of that thread. | |
3047 | ||
3048 | **slat** | |
3049 | Submission latency (avg being the average, stdev being the standard | |
3050 | deviation). This is the time it took to submit the I/O. For sync I/O, | |
3051 | the slat is really the completion latency, since queue/complete is one | |
3052 | operation there. This value can be in milliseconds or microseconds, fio | |
3053 | will choose the most appropriate base and print that. In the example | |
3054 | above, milliseconds is the best scale. Note: in :option:`--minimal` mode | |
0d237712 | 3055 | latencies are always expressed in microseconds. |
f80dba8d MT |
3056 | |
3057 | **clat** | |
3058 | Completion latency. Same names as slat, this denotes the time from | |
3059 | submission to completion of the I/O pieces. For sync I/O, clat will | |
3060 | usually be equal (or very close) to 0, as the time from submit to | |
3061 | complete is basically just CPU time (I/O has already been done, see slat | |
3062 | explanation). | |
3063 | ||
3064 | **bw** | |
3065 | Bandwidth. Same names as the xlat stats, but also includes an | |
3066 | approximate percentage of total aggregate bandwidth this thread received | |
3067 | in this group. This last value is only really useful if the threads in | |
3068 | this group are on the same disk, since they are then competing for disk | |
3069 | access. | |
3070 | ||
3071 | **cpu** | |
3072 | CPU usage. User and system time, along with the number of context | |
3073 | switches this thread went through, usage of system and user time, and | |
3074 | finally the number of major and minor page faults. The CPU utilization | |
3075 | numbers are averages for the jobs in that reporting group, while the | |
23a8e176 | 3076 | context and fault counters are summed. |
f80dba8d MT |
3077 | |
3078 | **IO depths** | |
3079 | The distribution of I/O depths over the job life time. The numbers are | |
3080 | divided into powers of 2, so for example the 16= entries includes depths | |
3081 | up to that value but higher than the previous entry. In other words, it | |
3082 | covers the range from 16 to 31. | |
3083 | ||
3084 | **IO submit** | |
3085 | How many pieces of I/O were submitting in a single submit call. Each | |
c60ebc45 SW |
3086 | entry denotes that amount and below, until the previous entry -- e.g., |
3087 | 8=100% mean that we submitted anywhere in between 5-8 I/Os per submit | |
f80dba8d MT |
3088 | call. |
3089 | ||
3090 | **IO complete** | |
3091 | Like the above submit number, but for completions instead. | |
3092 | ||
3093 | **IO issued** | |
3094 | The number of read/write requests issued, and how many of them were | |
3095 | short. | |
3096 | ||
3097 | **IO latencies** | |
3098 | The distribution of I/O completion latencies. This is the time from when | |
3099 | I/O leaves fio and when it gets completed. The numbers follow the same | |
3100 | pattern as the I/O depths, meaning that 2=1.6% means that 1.6% of the | |
3101 | I/O completed within 2 msecs, 20=12.8% means that 12.8% of the I/O took | |
3102 | more than 10 msecs, but less than (or equal to) 20 msecs. | |
71bfa161 JA |
3103 | |
3104 | After each client has been listed, the group statistics are printed. They | |
f80dba8d | 3105 | will look like this:: |
71bfa161 | 3106 | |
f80dba8d MT |
3107 | Run status group 0 (all jobs): |
3108 | READ: io=64MB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec | |
3109 | WRITE: io=64MB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec | |
71bfa161 JA |
3110 | |
3111 | For each data direction, it prints: | |
3112 | ||
f80dba8d MT |
3113 | **io** |
3114 | Number of megabytes I/O performed. | |
3115 | **aggrb** | |
3116 | Aggregate bandwidth of threads in this group. | |
3117 | **minb** | |
3118 | The minimum average bandwidth a thread saw. | |
3119 | **maxb** | |
3120 | The maximum average bandwidth a thread saw. | |
3121 | **mint** | |
3122 | The smallest runtime of the threads in that group. | |
3123 | **maxt** | |
3124 | The longest runtime of the threads in that group. | |
71bfa161 | 3125 | |
f80dba8d | 3126 | And finally, the disk statistics are printed. They will look like this:: |
71bfa161 | 3127 | |
f80dba8d MT |
3128 | Disk stats (read/write): |
3129 | sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00% | |
71bfa161 JA |
3130 | |
3131 | Each value is printed for both reads and writes, with reads first. The | |
3132 | numbers denote: | |
3133 | ||
f80dba8d | 3134 | **ios** |
c60ebc45 | 3135 | Number of I/Os performed by all groups. |
f80dba8d MT |
3136 | **merge** |
3137 | Number of merges I/O the I/O scheduler. | |
3138 | **ticks** | |
3139 | Number of ticks we kept the disk busy. | |
3140 | **io_queue** | |
3141 | Total time spent in the disk queue. | |
3142 | **util** | |
3143 | The disk utilization. A value of 100% means we kept the disk | |
71bfa161 JA |
3144 | busy constantly, 50% would be a disk idling half of the time. |
3145 | ||
f80dba8d MT |
3146 | It is also possible to get fio to dump the current output while it is running, |
3147 | without terminating the job. To do that, send fio the **USR1** signal. You can | |
3148 | also get regularly timed dumps by using the :option:`--status-interval` | |
3149 | parameter, or by creating a file in :file:`/tmp` named | |
3150 | :file:`fio-dump-status`. If fio sees this file, it will unlink it and dump the | |
3151 | current output status. | |
8423bd11 | 3152 | |
71bfa161 | 3153 | |
f80dba8d MT |
3154 | Terse output |
3155 | ------------ | |
71bfa161 | 3156 | |
f80dba8d MT |
3157 | For scripted usage where you typically want to generate tables or graphs of the |
3158 | results, fio can output the results in a semicolon separated format. The format | |
3159 | is one long line of values, such as:: | |
71bfa161 | 3160 | |
f80dba8d MT |
3161 | 2;card0;0;0;7139336;121836;60004;1;10109;27.932460;116.933948;220;126861;3495.446807;1085.368601;226;126864;3523.635629;1089.012448;24063;99944;50.275485%;59818.274627;5540.657370;7155060;122104;60004;1;8338;29.086342;117.839068;388;128077;5032.488518;1234.785715;391;128085;5061.839412;1236.909129;23436;100928;50.287926%;59964.832030;5644.844189;14.595833%;19.394167%;123706;0;7313;0.1%;0.1%;0.1%;0.1%;0.1%;0.1%;100.0%;0.00%;0.00%;0.00%;0.00%;0.00%;0.00%;0.01%;0.02%;0.05%;0.16%;6.04%;40.40%;52.68%;0.64%;0.01%;0.00%;0.01%;0.00%;0.00%;0.00%;0.00%;0.00% |
3162 | A description of this job goes here. | |
562c2d2f DN |
3163 | |
3164 | The job description (if provided) follows on a second line. | |
71bfa161 | 3165 | |
f80dba8d MT |
3166 | To enable terse output, use the :option:`--minimal` command line option. The |
3167 | first value is the version of the terse output format. If the output has to be | |
3168 | changed for some reason, this number will be incremented by 1 to signify that | |
3169 | change. | |
6820cb3b | 3170 | |
71bfa161 JA |
3171 | Split up, the format is as follows: |
3172 | ||
f80dba8d MT |
3173 | :: |
3174 | ||
3175 | terse version, fio version, jobname, groupid, error | |
3176 | ||
3177 | READ status:: | |
3178 | ||
3179 | Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec) | |
3180 | Submission latency: min, max, mean, stdev (usec) | |
3181 | Completion latency: min, max, mean, stdev (usec) | |
3182 | Completion latency percentiles: 20 fields (see below) | |
3183 | Total latency: min, max, mean, stdev (usec) | |
3184 | Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev | |
3185 | ||
3186 | WRITE status: | |
3187 | ||
3188 | :: | |
3189 | ||
3190 | Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec) | |
3191 | Submission latency: min, max, mean, stdev (usec) | |
3192 | Completion latency: min, max, mean, stdev(usec) | |
3193 | Completion latency percentiles: 20 fields (see below) | |
3194 | Total latency: min, max, mean, stdev (usec) | |
3195 | Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev | |
3196 | ||
3197 | CPU usage:: | |
3198 | ||
3199 | user, system, context switches, major faults, minor faults | |
3200 | ||
3201 | I/O depths:: | |
3202 | ||
3203 | <=1, 2, 4, 8, 16, 32, >=64 | |
3204 | ||
3205 | I/O latencies microseconds:: | |
3206 | ||
3207 | <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000 | |
3208 | ||
3209 | I/O latencies milliseconds:: | |
3210 | ||
3211 | <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000 | |
3212 | ||
3213 | Disk utilization:: | |
3214 | ||
3215 | Disk name, Read ios, write ios, | |
3216 | Read merges, write merges, | |
3217 | Read ticks, write ticks, | |
3218 | Time spent in queue, disk utilization percentage | |
3219 | ||
3220 | Additional Info (dependent on continue_on_error, default off):: | |
3221 | ||
3222 | total # errors, first error code | |
3223 | ||
3224 | Additional Info (dependent on description being set):: | |
3225 | ||
3226 | Text description | |
3227 | ||
3228 | Completion latency percentiles can be a grouping of up to 20 sets, so for the | |
3229 | terse output fio writes all of them. Each field will look like this:: | |
1db92cb6 JA |
3230 | |
3231 | 1.00%=6112 | |
3232 | ||
f80dba8d | 3233 | which is the Xth percentile, and the `usec` latency associated with it. |
1db92cb6 | 3234 | |
f80dba8d MT |
3235 | For disk utilization, all disks used by fio are shown. So for each disk there |
3236 | will be a disk utilization section. | |
f2f788dd | 3237 | |
2fc26c3d IC |
3238 | Below is a single line containing short names for each of the fields in the |
3239 | minimal output v3, separated by semicolons: | |
3240 | ||
3241 | terse_version_3;fio_version;jobname;groupid;error;read_kb;read_bandwidth;read_iops;read_runtime_ms;read_slat_min;read_slat_max;read_slat_mean;read_slat_dev;read_clat_max;read_clat_min;read_clat_mean;read_clat_dev;read_clat_pct01;read_clat_pct02;read_clat_pct03;read_clat_pct04;read_clat_pct05;read_clat_pct06;read_clat_pct07;read_clat_pct08;read_clat_pct09;read_clat_pct10;read_clat_pct11;read_clat_pct12;read_clat_pct13;read_clat_pct14;read_clat_pct15;read_clat_pct16;read_clat_pct17;read_clat_pct18;read_clat_pct19;read_clat_pct20;read_tlat_min;read_lat_max;read_lat_mean;read_lat_dev;read_bw_min;read_bw_max;read_bw_agg_pct;read_bw_mean;read_bw_dev;write_kb;write_bandwidth;write_iops;write_runtime_ms;write_slat_min;write_slat_max;write_slat_mean;write_slat_dev;write_clat_max;write_clat_min;write_clat_mean;write_clat_dev;write_clat_pct01;write_clat_pct02;write_clat_pct03;write_clat_pct04;write_clat_pct05;write_clat_pct06;write_clat_pct07;write_clat_pct08;write_clat_pct09;write_clat_pct10;write_clat_pct11;write_clat_pct12;write_clat_pct13;write_clat_pct14;write_clat_pct15;write_clat_pct16;write_clat_pct17;write_clat_pct18;write_clat_pct19;write_clat_pct20;write_tlat_min;write_lat_max;write_lat_mean;write_lat_dev;write_bw_min;write_bw_max;write_bw_agg_pct;write_bw_mean;write_bw_dev;cpu_user;cpu_sys;cpu_csw;cpu_mjf;pu_minf;iodepth_1;iodepth_2;iodepth_4;iodepth_8;iodepth_16;iodepth_32;iodepth_64;lat_2us;lat_4us;lat_10us;lat_20us;lat_50us;lat_100us;lat_250us;lat_500us;lat_750us;lat_1000us;lat_2ms;lat_4ms;lat_10ms;lat_20ms;lat_50ms;lat_100ms;lat_250ms;lat_500ms;lat_750ms;lat_1000ms;lat_2000ms;lat_over_2000ms;disk_name;disk_read_iops;disk_write_iops;disk_read_merges;disk_write_merges;disk_read_ticks;write_ticks;disk_queue_time;disk_util | |
3242 | ||
25c8b9d7 | 3243 | |
f80dba8d MT |
3244 | Trace file format |
3245 | ----------------- | |
3246 | ||
3247 | There are two trace file format that you can encounter. The older (v1) format is | |
3248 | unsupported since version 1.20-rc3 (March 2008). It will still be described | |
25c8b9d7 PD |
3249 | below in case that you get an old trace and want to understand it. |
3250 | ||
3251 | In any case the trace is a simple text file with a single action per line. | |
3252 | ||
3253 | ||
f80dba8d MT |
3254 | Trace file format v1 |
3255 | ~~~~~~~~~~~~~~~~~~~~ | |
3256 | ||
3257 | Each line represents a single I/O action in the following format:: | |
3258 | ||
3259 | rw, offset, length | |
25c8b9d7 | 3260 | |
f80dba8d | 3261 | where `rw=0/1` for read/write, and the offset and length entries being in bytes. |
25c8b9d7 | 3262 | |
f80dba8d | 3263 | This format is not supported in fio versions => 1.20-rc3. |
25c8b9d7 | 3264 | |
25c8b9d7 | 3265 | |
f80dba8d MT |
3266 | Trace file format v2 |
3267 | ~~~~~~~~~~~~~~~~~~~~ | |
25c8b9d7 | 3268 | |
f80dba8d MT |
3269 | The second version of the trace file format was added in fio version 1.17. It |
3270 | allows to access more then one file per trace and has a bigger set of possible | |
3271 | file actions. | |
25c8b9d7 | 3272 | |
f80dba8d | 3273 | The first line of the trace file has to be:: |
25c8b9d7 | 3274 | |
f80dba8d | 3275 | fio version 2 iolog |
25c8b9d7 PD |
3276 | |
3277 | Following this can be lines in two different formats, which are described below. | |
3278 | ||
f80dba8d | 3279 | The file management format:: |
25c8b9d7 | 3280 | |
f80dba8d | 3281 | filename action |
25c8b9d7 PD |
3282 | |
3283 | The filename is given as an absolute path. The action can be one of these: | |
3284 | ||
f80dba8d MT |
3285 | **add** |
3286 | Add the given filename to the trace. | |
3287 | **open** | |
3288 | Open the file with the given filename. The filename has to have | |
3289 | been added with the **add** action before. | |
3290 | **close** | |
3291 | Close the file with the given filename. The file has to have been | |
3292 | opened before. | |
3293 | ||
3294 | ||
3295 | The file I/O action format:: | |
3296 | ||
3297 | filename action offset length | |
3298 | ||
3299 | The `filename` is given as an absolute path, and has to have been added and | |
3300 | opened before it can be used with this format. The `offset` and `length` are | |
3301 | given in bytes. The `action` can be one of these: | |
3302 | ||
3303 | **wait** | |
3304 | Wait for `offset` microseconds. Everything below 100 is discarded. | |
3305 | The time is relative to the previous `wait` statement. | |
3306 | **read** | |
3307 | Read `length` bytes beginning from `offset`. | |
3308 | **write** | |
3309 | Write `length` bytes beginning from `offset`. | |
3310 | **sync** | |
3311 | :manpage:`fsync(2)` the file. | |
3312 | **datasync** | |
3313 | :manpage:`fdatasync(2)` the file. | |
3314 | **trim** | |
3315 | Trim the given file from the given `offset` for `length` bytes. | |
3316 | ||
3317 | CPU idleness profiling | |
3318 | ---------------------- | |
3319 | ||
3320 | In some cases, we want to understand CPU overhead in a test. For example, we | |
3321 | test patches for the specific goodness of whether they reduce CPU usage. | |
3322 | Fio implements a balloon approach to create a thread per CPU that runs at idle | |
3323 | priority, meaning that it only runs when nobody else needs the cpu. | |
3324 | By measuring the amount of work completed by the thread, idleness of each CPU | |
3325 | can be derived accordingly. | |
3326 | ||
3327 | An unit work is defined as touching a full page of unsigned characters. Mean and | |
3328 | standard deviation of time to complete an unit work is reported in "unit work" | |
3329 | section. Options can be chosen to report detailed percpu idleness or overall | |
3330 | system idleness by aggregating percpu stats. | |
3331 | ||
3332 | ||
3333 | Verification and triggers | |
3334 | ------------------------- | |
3335 | ||
3336 | Fio is usually run in one of two ways, when data verification is done. The first | |
3337 | is a normal write job of some sort with verify enabled. When the write phase has | |
3338 | completed, fio switches to reads and verifies everything it wrote. The second | |
3339 | model is running just the write phase, and then later on running the same job | |
3340 | (but with reads instead of writes) to repeat the same I/O patterns and verify | |
3341 | the contents. Both of these methods depend on the write phase being completed, | |
3342 | as fio otherwise has no idea how much data was written. | |
3343 | ||
3344 | With verification triggers, fio supports dumping the current write state to | |
3345 | local files. Then a subsequent read verify workload can load this state and know | |
3346 | exactly where to stop. This is useful for testing cases where power is cut to a | |
3347 | server in a managed fashion, for instance. | |
99b9a85a JA |
3348 | |
3349 | A verification trigger consists of two things: | |
3350 | ||
f80dba8d MT |
3351 | 1) Storing the write state of each job. |
3352 | 2) Executing a trigger command. | |
99b9a85a | 3353 | |
f80dba8d MT |
3354 | The write state is relatively small, on the order of hundreds of bytes to single |
3355 | kilobytes. It contains information on the number of completions done, the last X | |
3356 | completions, etc. | |
99b9a85a | 3357 | |
f80dba8d MT |
3358 | A trigger is invoked either through creation ('touch') of a specified file in |
3359 | the system, or through a timeout setting. If fio is run with | |
3360 | :option:`--trigger-file` = :file:`/tmp/trigger-file`, then it will continually | |
3361 | check for the existence of :file:`/tmp/trigger-file`. When it sees this file, it | |
3362 | will fire off the trigger (thus saving state, and executing the trigger | |
99b9a85a JA |
3363 | command). |
3364 | ||
f80dba8d MT |
3365 | For client/server runs, there's both a local and remote trigger. If fio is |
3366 | running as a server backend, it will send the job states back to the client for | |
3367 | safe storage, then execute the remote trigger, if specified. If a local trigger | |
3368 | is specified, the server will still send back the write state, but the client | |
3369 | will then execute the trigger. | |
99b9a85a | 3370 | |
f80dba8d MT |
3371 | Verification trigger example |
3372 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
99b9a85a | 3373 | |
f80dba8d MT |
3374 | Lets say we want to run a powercut test on the remote machine 'server'. Our |
3375 | write workload is in :file:`write-test.fio`. We want to cut power to 'server' at | |
3376 | some point during the run, and we'll run this test from the safety or our local | |
3377 | machine, 'localbox'. On the server, we'll start the fio backend normally:: | |
99b9a85a | 3378 | |
f80dba8d | 3379 | server# fio --server |
99b9a85a | 3380 | |
f80dba8d | 3381 | and on the client, we'll fire off the workload:: |
99b9a85a | 3382 | |
f80dba8d | 3383 | localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger-remote="bash -c \"echo b > /proc/sysrq-triger\"" |
99b9a85a | 3384 | |
f80dba8d | 3385 | We set :file:`/tmp/my-trigger` as the trigger file, and we tell fio to execute:: |
99b9a85a | 3386 | |
f80dba8d | 3387 | echo b > /proc/sysrq-trigger |
99b9a85a | 3388 | |
f80dba8d MT |
3389 | on the server once it has received the trigger and sent us the write state. This |
3390 | will work, but it's not **really** cutting power to the server, it's merely | |
3391 | abruptly rebooting it. If we have a remote way of cutting power to the server | |
3392 | through IPMI or similar, we could do that through a local trigger command | |
3393 | instead. Lets assume we have a script that does IPMI reboot of a given hostname, | |
3394 | ipmi-reboot. On localbox, we could then have run fio with a local trigger | |
3395 | instead:: | |
99b9a85a | 3396 | |
f80dba8d | 3397 | localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger="ipmi-reboot server" |
99b9a85a | 3398 | |
f80dba8d MT |
3399 | For this case, fio would wait for the server to send us the write state, then |
3400 | execute ``ipmi-reboot server`` when that happened. | |
3401 | ||
3402 | Loading verify state | |
3403 | ~~~~~~~~~~~~~~~~~~~~ | |
3404 | ||
3405 | To load store write state, read verification job file must contain the | |
3406 | :option:`verify_state_load` option. If that is set, fio will load the previously | |
99b9a85a | 3407 | stored state. For a local fio run this is done by loading the files directly, |
f80dba8d MT |
3408 | and on a client/server run, the server backend will ask the client to send the |
3409 | files over and load them from there. | |
a3ae5b05 JA |
3410 | |
3411 | ||
f80dba8d MT |
3412 | Log File Formats |
3413 | ---------------- | |
a3ae5b05 JA |
3414 | |
3415 | Fio supports a variety of log file formats, for logging latencies, bandwidth, | |
3416 | and IOPS. The logs share a common format, which looks like this: | |
3417 | ||
f80dba8d | 3418 | *time* (`msec`), *value*, *data direction*, *offset* |
a3ae5b05 | 3419 | |
f80dba8d | 3420 | Time for the log entry is always in milliseconds. The *value* logged depends |
a3ae5b05 JA |
3421 | on the type of log, it will be one of the following: |
3422 | ||
f80dba8d MT |
3423 | **Latency log** |
3424 | Value is latency in usecs | |
3425 | **Bandwidth log** | |
3426 | Value is in KiB/sec | |
3427 | **IOPS log** | |
3428 | Value is IOPS | |
3429 | ||
3430 | *Data direction* is one of the following: | |
3431 | ||
3432 | **0** | |
3433 | I/O is a READ | |
3434 | **1** | |
3435 | I/O is a WRITE | |
3436 | **2** | |
3437 | I/O is a TRIM | |
3438 | ||
3439 | The *offset* is the offset, in bytes, from the start of the file, for that | |
3440 | particular I/O. The logging of the offset can be toggled with | |
3441 | :option:`log_offset`. | |
3442 | ||
3443 | If windowed logging is enabled through :option:`log_avg_msec` then fio doesn't | |
c60ebc45 | 3444 | log individual I/Os. Instead of logs the average values over the specified period |
f80dba8d MT |
3445 | of time. Since 'data direction' and 'offset' are per-I/O values, they aren't |
3446 | applicable if windowed logging is enabled. If windowed logging is enabled and | |
3447 | :option:`log_max_value` is set, then fio logs maximum values in that window | |
3448 | instead of averages. | |
3449 | ||
3450 | ||
3451 | Client/server | |
3452 | ------------- | |
3453 | ||
3454 | Normally fio is invoked as a stand-alone application on the machine where the | |
3455 | I/O workload should be generated. However, the frontend and backend of fio can | |
3456 | be run separately. Ie the fio server can generate an I/O workload on the "Device | |
3457 | Under Test" while being controlled from another machine. | |
3458 | ||
3459 | Start the server on the machine which has access to the storage DUT:: | |
3460 | ||
3461 | fio --server=args | |
3462 | ||
3463 | where args defines what fio listens to. The arguments are of the form | |
3464 | ``type,hostname`` or ``IP,port``. *type* is either ``ip`` (or ip4) for TCP/IP | |
3465 | v4, ``ip6`` for TCP/IP v6, or ``sock`` for a local unix domain socket. | |
3466 | *hostname* is either a hostname or IP address, and *port* is the port to listen | |
3467 | to (only valid for TCP/IP, not a local socket). Some examples: | |
3468 | ||
3469 | 1) ``fio --server`` | |
3470 | ||
3471 | Start a fio server, listening on all interfaces on the default port (8765). | |
3472 | ||
3473 | 2) ``fio --server=ip:hostname,4444`` | |
3474 | ||
3475 | Start a fio server, listening on IP belonging to hostname and on port 4444. | |
3476 | ||
3477 | 3) ``fio --server=ip6:::1,4444`` | |
3478 | ||
3479 | Start a fio server, listening on IPv6 localhost ::1 and on port 4444. | |
3480 | ||
3481 | 4) ``fio --server=,4444`` | |
3482 | ||
3483 | Start a fio server, listening on all interfaces on port 4444. | |
3484 | ||
3485 | 5) ``fio --server=1.2.3.4`` | |
3486 | ||
3487 | Start a fio server, listening on IP 1.2.3.4 on the default port. | |
3488 | ||
3489 | 6) ``fio --server=sock:/tmp/fio.sock`` | |
3490 | ||
3491 | Start a fio server, listening on the local socket /tmp/fio.sock. | |
3492 | ||
3493 | Once a server is running, a "client" can connect to the fio server with:: | |
3494 | ||
3495 | fio <local-args> --client=<server> <remote-args> <job file(s)> | |
3496 | ||
3497 | where `local-args` are arguments for the client where it is running, `server` | |
3498 | is the connect string, and `remote-args` and `job file(s)` are sent to the | |
3499 | server. The `server` string follows the same format as it does on the server | |
3500 | side, to allow IP/hostname/socket and port strings. | |
3501 | ||
3502 | Fio can connect to multiple servers this way:: | |
3503 | ||
3504 | fio --client=<server1> <job file(s)> --client=<server2> <job file(s)> | |
3505 | ||
3506 | If the job file is located on the fio server, then you can tell the server to | |
3507 | load a local file as well. This is done by using :option:`--remote-config` :: | |
3508 | ||
3509 | fio --client=server --remote-config /path/to/file.fio | |
3510 | ||
3511 | Then fio will open this local (to the server) job file instead of being passed | |
3512 | one from the client. | |
3513 | ||
3514 | If you have many servers (example: 100 VMs/containers), you can input a pathname | |
3515 | of a file containing host IPs/names as the parameter value for the | |
3516 | :option:`--client` option. For example, here is an example :file:`host.list` | |
3517 | file containing 2 hostnames:: | |
3518 | ||
3519 | host1.your.dns.domain | |
3520 | host2.your.dns.domain | |
3521 | ||
3522 | The fio command would then be:: | |
a3ae5b05 | 3523 | |
f80dba8d | 3524 | fio --client=host.list <job file(s)> |
a3ae5b05 | 3525 | |
f80dba8d MT |
3526 | In this mode, you cannot input server-specific parameters or job files -- all |
3527 | servers receive the same job file. | |
a3ae5b05 | 3528 | |
f80dba8d MT |
3529 | In order to let ``fio --client`` runs use a shared filesystem from multiple |
3530 | hosts, ``fio --client`` now prepends the IP address of the server to the | |
3531 | filename. For example, if fio is using directory :file:`/mnt/nfs/fio` and is | |
3532 | writing filename :file:`fileio.tmp`, with a :option:`--client` `hostfile` | |
3533 | containing two hostnames ``h1`` and ``h2`` with IP addresses 192.168.10.120 and | |
3534 | 192.168.10.121, then fio will create two files:: | |
a3ae5b05 | 3535 | |
f80dba8d MT |
3536 | /mnt/nfs/fio/192.168.10.120.fileio.tmp |
3537 | /mnt/nfs/fio/192.168.10.121.fileio.tmp |