| 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 `type` of various fio actions. May be ``all`` for all types |
| 58 | or individual types separated by a comma (e.g. ``--debug=file,mem`` will |
| 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 | *steadystate* |
| 97 | Dump info related to steadystate detection. |
| 98 | *helperthread* |
| 99 | Dump info related to the helper thread. |
| 100 | *zbd* |
| 101 | Dump info related to support for zoned block devices. |
| 102 | *?* or *help* |
| 103 | Show available debug options. |
| 104 | |
| 105 | .. option:: --parse-only |
| 106 | |
| 107 | Parse options only, don't start any I/O. |
| 108 | |
| 109 | .. option:: --merge-blktrace-only |
| 110 | |
| 111 | Merge blktraces only, don't start any I/O. |
| 112 | |
| 113 | .. option:: --output=filename |
| 114 | |
| 115 | Write output to file `filename`. |
| 116 | |
| 117 | .. option:: --output-format=format |
| 118 | |
| 119 | Set the reporting `format` to `normal`, `terse`, `json`, or `json+`. Multiple |
| 120 | formats can be selected, separated by a comma. `terse` is a CSV based |
| 121 | format. `json+` is like `json`, except it adds a full dump of the latency |
| 122 | buckets. |
| 123 | |
| 124 | .. option:: --bandwidth-log |
| 125 | |
| 126 | Generate aggregate bandwidth logs. |
| 127 | |
| 128 | .. option:: --minimal |
| 129 | |
| 130 | Print statistics in a terse, semicolon-delimited format. |
| 131 | |
| 132 | .. option:: --append-terse |
| 133 | |
| 134 | Print statistics in selected mode AND terse, semicolon-delimited format. |
| 135 | **Deprecated**, use :option:`--output-format` instead to select multiple |
| 136 | formats. |
| 137 | |
| 138 | .. option:: --terse-version=version |
| 139 | |
| 140 | Set terse `version` output format (default 3, or 2 or 4 or 5). |
| 141 | |
| 142 | .. option:: --version |
| 143 | |
| 144 | Print version information and exit. |
| 145 | |
| 146 | .. option:: --help |
| 147 | |
| 148 | Print a summary of the command line options and exit. |
| 149 | |
| 150 | .. option:: --cpuclock-test |
| 151 | |
| 152 | Perform test and validation of internal CPU clock. |
| 153 | |
| 154 | .. option:: --crctest=[test] |
| 155 | |
| 156 | Test the speed of the built-in checksumming functions. If no argument is |
| 157 | given, all of them are tested. Alternatively, a comma separated list can |
| 158 | be passed, in which case the given ones are tested. |
| 159 | |
| 160 | .. option:: --cmdhelp=command |
| 161 | |
| 162 | Print help information for `command`. May be ``all`` for all commands. |
| 163 | |
| 164 | .. option:: --enghelp=[ioengine[,command]] |
| 165 | |
| 166 | List all commands defined by `ioengine`, or print help for `command` |
| 167 | defined by `ioengine`. If no `ioengine` is given, list all |
| 168 | available ioengines. |
| 169 | |
| 170 | .. option:: --showcmd=jobfile |
| 171 | |
| 172 | Convert `jobfile` to a set of command-line options. |
| 173 | |
| 174 | .. option:: --readonly |
| 175 | |
| 176 | Turn on safety read-only checks, preventing writes and trims. The |
| 177 | ``--readonly`` option is an extra safety guard to prevent users from |
| 178 | accidentally starting a write or trim workload when that is not desired. |
| 179 | Fio will only modify the device under test if |
| 180 | `rw=write/randwrite/rw/randrw/trim/randtrim/trimwrite` is given. This |
| 181 | safety net can be used as an extra precaution. |
| 182 | |
| 183 | .. option:: --eta=when |
| 184 | |
| 185 | Specifies when real-time ETA estimate should be printed. `when` may be |
| 186 | `always`, `never` or `auto`. `auto` is the default, it prints ETA |
| 187 | when requested if the output is a TTY. `always` disregards the output |
| 188 | type, and prints ETA when requested. `never` never prints ETA. |
| 189 | |
| 190 | .. option:: --eta-interval=time |
| 191 | |
| 192 | By default, fio requests client ETA status roughly every second. With |
| 193 | this option, the interval is configurable. Fio imposes a minimum |
| 194 | allowed time to avoid flooding the console, less than 250 msec is |
| 195 | not supported. |
| 196 | |
| 197 | .. option:: --eta-newline=time |
| 198 | |
| 199 | Force a new line for every `time` period passed. When the unit is omitted, |
| 200 | the value is interpreted in seconds. |
| 201 | |
| 202 | .. option:: --status-interval=time |
| 203 | |
| 204 | Force a full status dump of cumulative (from job start) values at `time` |
| 205 | intervals. This option does *not* provide per-period measurements. So |
| 206 | values such as bandwidth are running averages. When the time unit is omitted, |
| 207 | `time` is interpreted in seconds. Note that using this option with |
| 208 | ``--output-format=json`` will yield output that technically isn't valid |
| 209 | json, since the output will be collated sets of valid json. It will need |
| 210 | to be split into valid sets of json after the run. |
| 211 | |
| 212 | .. option:: --section=name |
| 213 | |
| 214 | Only run specified section `name` in job file. Multiple sections can be specified. |
| 215 | The ``--section`` option allows one to combine related jobs into one file. |
| 216 | E.g. one job file could define light, moderate, and heavy sections. Tell |
| 217 | fio to run only the "heavy" section by giving ``--section=heavy`` |
| 218 | command line option. One can also specify the "write" operations in one |
| 219 | section and "verify" operation in another section. The ``--section`` option |
| 220 | only applies to job sections. The reserved *global* section is always |
| 221 | parsed and used. |
| 222 | |
| 223 | .. option:: --alloc-size=kb |
| 224 | |
| 225 | Allocate additional internal smalloc pools of size `kb` in KiB. The |
| 226 | ``--alloc-size`` option increases shared memory set aside for use by fio. |
| 227 | If running large jobs with randommap enabled, fio can run out of memory. |
| 228 | Smalloc is an internal allocator for shared structures from a fixed size |
| 229 | memory pool and can grow to 16 pools. The pool size defaults to 16MiB. |
| 230 | |
| 231 | NOTE: While running :file:`.fio_smalloc.*` backing store files are visible |
| 232 | in :file:`/tmp`. |
| 233 | |
| 234 | .. option:: --warnings-fatal |
| 235 | |
| 236 | All fio parser warnings are fatal, causing fio to exit with an |
| 237 | error. |
| 238 | |
| 239 | .. option:: --max-jobs=nr |
| 240 | |
| 241 | Set the maximum number of threads/processes to support to `nr`. |
| 242 | NOTE: On Linux, it may be necessary to increase the shared-memory |
| 243 | limit (:file:`/proc/sys/kernel/shmmax`) if fio runs into errors while |
| 244 | creating jobs. |
| 245 | |
| 246 | .. option:: --server=args |
| 247 | |
| 248 | Start a backend server, with `args` specifying what to listen to. |
| 249 | See `Client/Server`_ section. |
| 250 | |
| 251 | .. option:: --daemonize=pidfile |
| 252 | |
| 253 | Background a fio server, writing the pid to the given `pidfile` file. |
| 254 | |
| 255 | .. option:: --client=hostname |
| 256 | |
| 257 | Instead of running the jobs locally, send and run them on the given `hostname` |
| 258 | or set of `hostname`\s. See `Client/Server`_ section. |
| 259 | |
| 260 | .. option:: --remote-config=file |
| 261 | |
| 262 | Tell fio server to load this local `file`. |
| 263 | |
| 264 | .. option:: --idle-prof=option |
| 265 | |
| 266 | Report CPU idleness. `option` is one of the following: |
| 267 | |
| 268 | **calibrate** |
| 269 | Run unit work calibration only and exit. |
| 270 | |
| 271 | **system** |
| 272 | Show aggregate system idleness and unit work. |
| 273 | |
| 274 | **percpu** |
| 275 | As **system** but also show per CPU idleness. |
| 276 | |
| 277 | .. option:: --inflate-log=log |
| 278 | |
| 279 | Inflate and output compressed `log`. |
| 280 | |
| 281 | .. option:: --trigger-file=file |
| 282 | |
| 283 | Execute trigger command when `file` exists. |
| 284 | |
| 285 | .. option:: --trigger-timeout=time |
| 286 | |
| 287 | Execute trigger at this `time`. |
| 288 | |
| 289 | .. option:: --trigger=command |
| 290 | |
| 291 | Set this `command` as local trigger. |
| 292 | |
| 293 | .. option:: --trigger-remote=command |
| 294 | |
| 295 | Set this `command` as remote trigger. |
| 296 | |
| 297 | .. option:: --aux-path=path |
| 298 | |
| 299 | Use the directory specified by `path` for generated state files instead |
| 300 | of the current working directory. |
| 301 | |
| 302 | Any parameters following the options will be assumed to be job files, unless |
| 303 | they match a job file parameter. Multiple job files can be listed and each job |
| 304 | file will be regarded as a separate group. Fio will :option:`stonewall` |
| 305 | execution between each group. |
| 306 | |
| 307 | |
| 308 | Job file format |
| 309 | --------------- |
| 310 | |
| 311 | As previously described, fio accepts one or more job files describing what it is |
| 312 | supposed to do. The job file format is the classic ini file, where the names |
| 313 | enclosed in [] brackets define the job name. You are free to use any ASCII name |
| 314 | you want, except *global* which has special meaning. Following the job name is |
| 315 | a sequence of zero or more parameters, one per line, that define the behavior of |
| 316 | the job. If the first character in a line is a ';' or a '#', the entire line is |
| 317 | discarded as a comment. |
| 318 | |
| 319 | A *global* section sets defaults for the jobs described in that file. A job may |
| 320 | override a *global* section parameter, and a job file may even have several |
| 321 | *global* sections if so desired. A job is only affected by a *global* section |
| 322 | residing above it. |
| 323 | |
| 324 | The :option:`--cmdhelp` option also lists all options. If used with a `command` |
| 325 | argument, :option:`--cmdhelp` will detail the given `command`. |
| 326 | |
| 327 | See the `examples/` directory for inspiration on how to write job files. Note |
| 328 | the copyright and license requirements currently apply to `examples/` files. |
| 329 | |
| 330 | So let's look at a really simple job file that defines two processes, each |
| 331 | randomly reading from a 128MiB file: |
| 332 | |
| 333 | .. code-block:: ini |
| 334 | |
| 335 | ; -- start job file -- |
| 336 | [global] |
| 337 | rw=randread |
| 338 | size=128m |
| 339 | |
| 340 | [job1] |
| 341 | |
| 342 | [job2] |
| 343 | |
| 344 | ; -- end job file -- |
| 345 | |
| 346 | As you can see, the job file sections themselves are empty as all the described |
| 347 | parameters are shared. As no :option:`filename` option is given, fio makes up a |
| 348 | `filename` for each of the jobs as it sees fit. On the command line, this job |
| 349 | would look as follows:: |
| 350 | |
| 351 | $ fio --name=global --rw=randread --size=128m --name=job1 --name=job2 |
| 352 | |
| 353 | |
| 354 | Let's look at an example that has a number of processes writing randomly to |
| 355 | files: |
| 356 | |
| 357 | .. code-block:: ini |
| 358 | |
| 359 | ; -- start job file -- |
| 360 | [random-writers] |
| 361 | ioengine=libaio |
| 362 | iodepth=4 |
| 363 | rw=randwrite |
| 364 | bs=32k |
| 365 | direct=0 |
| 366 | size=64m |
| 367 | numjobs=4 |
| 368 | ; -- end job file -- |
| 369 | |
| 370 | Here we have no *global* section, as we only have one job defined anyway. We |
| 371 | want to use async I/O here, with a depth of 4 for each file. We also increased |
| 372 | the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical |
| 373 | jobs. The result is 4 processes each randomly writing to their own 64MiB |
| 374 | file. Instead of using the above job file, you could have given the parameters |
| 375 | on the command line. For this case, you would specify:: |
| 376 | |
| 377 | $ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4 |
| 378 | |
| 379 | When fio is utilized as a basis of any reasonably large test suite, it might be |
| 380 | desirable to share a set of standardized settings across multiple job files. |
| 381 | Instead of copy/pasting such settings, any section may pull in an external |
| 382 | :file:`filename.fio` file with *include filename* directive, as in the following |
| 383 | example:: |
| 384 | |
| 385 | ; -- start job file including.fio -- |
| 386 | [global] |
| 387 | filename=/tmp/test |
| 388 | filesize=1m |
| 389 | include glob-include.fio |
| 390 | |
| 391 | [test] |
| 392 | rw=randread |
| 393 | bs=4k |
| 394 | time_based=1 |
| 395 | runtime=10 |
| 396 | include test-include.fio |
| 397 | ; -- end job file including.fio -- |
| 398 | |
| 399 | .. code-block:: ini |
| 400 | |
| 401 | ; -- start job file glob-include.fio -- |
| 402 | thread=1 |
| 403 | group_reporting=1 |
| 404 | ; -- end job file glob-include.fio -- |
| 405 | |
| 406 | .. code-block:: ini |
| 407 | |
| 408 | ; -- start job file test-include.fio -- |
| 409 | ioengine=libaio |
| 410 | iodepth=4 |
| 411 | ; -- end job file test-include.fio -- |
| 412 | |
| 413 | Settings pulled into a section apply to that section only (except *global* |
| 414 | section). Include directives may be nested in that any included file may contain |
| 415 | further include directive(s). Include files may not contain [] sections. |
| 416 | |
| 417 | |
| 418 | Environment variables |
| 419 | ~~~~~~~~~~~~~~~~~~~~~ |
| 420 | |
| 421 | Fio also supports environment variable expansion in job files. Any sub-string of |
| 422 | the form ``${VARNAME}`` as part of an option value (in other words, on the right |
| 423 | of the '='), will be expanded to the value of the environment variable called |
| 424 | `VARNAME`. If no such environment variable is defined, or `VARNAME` is the |
| 425 | empty string, the empty string will be substituted. |
| 426 | |
| 427 | As an example, let's look at a sample fio invocation and job file:: |
| 428 | |
| 429 | $ SIZE=64m NUMJOBS=4 fio jobfile.fio |
| 430 | |
| 431 | .. code-block:: ini |
| 432 | |
| 433 | ; -- start job file -- |
| 434 | [random-writers] |
| 435 | rw=randwrite |
| 436 | size=${SIZE} |
| 437 | numjobs=${NUMJOBS} |
| 438 | ; -- end job file -- |
| 439 | |
| 440 | This will expand to the following equivalent job file at runtime: |
| 441 | |
| 442 | .. code-block:: ini |
| 443 | |
| 444 | ; -- start job file -- |
| 445 | [random-writers] |
| 446 | rw=randwrite |
| 447 | size=64m |
| 448 | numjobs=4 |
| 449 | ; -- end job file -- |
| 450 | |
| 451 | Fio ships with a few example job files, you can also look there for inspiration. |
| 452 | |
| 453 | Reserved keywords |
| 454 | ~~~~~~~~~~~~~~~~~ |
| 455 | |
| 456 | Additionally, fio has a set of reserved keywords that will be replaced |
| 457 | internally with the appropriate value. Those keywords are: |
| 458 | |
| 459 | **$pagesize** |
| 460 | |
| 461 | The architecture page size of the running system. |
| 462 | |
| 463 | **$mb_memory** |
| 464 | |
| 465 | Megabytes of total memory in the system. |
| 466 | |
| 467 | **$ncpus** |
| 468 | |
| 469 | Number of online available CPUs. |
| 470 | |
| 471 | These can be used on the command line or in the job file, and will be |
| 472 | automatically substituted with the current system values when the job is |
| 473 | run. Simple math is also supported on these keywords, so you can perform actions |
| 474 | like:: |
| 475 | |
| 476 | size=8*$mb_memory |
| 477 | |
| 478 | and get that properly expanded to 8 times the size of memory in the machine. |
| 479 | |
| 480 | |
| 481 | Job file parameters |
| 482 | ------------------- |
| 483 | |
| 484 | This section describes in details each parameter associated with a job. Some |
| 485 | parameters take an option of a given type, such as an integer or a |
| 486 | string. Anywhere a numeric value is required, an arithmetic expression may be |
| 487 | used, provided it is surrounded by parentheses. Supported operators are: |
| 488 | |
| 489 | - addition (+) |
| 490 | - subtraction (-) |
| 491 | - multiplication (*) |
| 492 | - division (/) |
| 493 | - modulus (%) |
| 494 | - exponentiation (^) |
| 495 | |
| 496 | For time values in expressions, units are microseconds by default. This is |
| 497 | different than for time values not in expressions (not enclosed in |
| 498 | parentheses). The following types are used: |
| 499 | |
| 500 | |
| 501 | Parameter types |
| 502 | ~~~~~~~~~~~~~~~ |
| 503 | |
| 504 | **str** |
| 505 | String: A sequence of alphanumeric characters. |
| 506 | |
| 507 | **time** |
| 508 | Integer with possible time suffix. Without a unit value is interpreted as |
| 509 | seconds unless otherwise specified. Accepts a suffix of 'd' for days, 'h' for |
| 510 | hours, 'm' for minutes, 's' for seconds, 'ms' (or 'msec') for milliseconds and |
| 511 | 'us' (or 'usec') for microseconds. For example, use 10m for 10 minutes. |
| 512 | |
| 513 | .. _int: |
| 514 | |
| 515 | **int** |
| 516 | Integer. A whole number value, which may contain an integer prefix |
| 517 | and an integer suffix: |
| 518 | |
| 519 | [*integer prefix*] **number** [*integer suffix*] |
| 520 | |
| 521 | The optional *integer prefix* specifies the number's base. The default |
| 522 | is decimal. *0x* specifies hexadecimal. |
| 523 | |
| 524 | The optional *integer suffix* specifies the number's units, and includes an |
| 525 | optional unit prefix and an optional unit. For quantities of data, the |
| 526 | default unit is bytes. For quantities of time, the default unit is seconds |
| 527 | unless otherwise specified. |
| 528 | |
| 529 | With :option:`kb_base`\=1000, fio follows international standards for unit |
| 530 | prefixes. To specify power-of-10 decimal values defined in the |
| 531 | International System of Units (SI): |
| 532 | |
| 533 | * *K* -- means kilo (K) or 1000 |
| 534 | * *M* -- means mega (M) or 1000**2 |
| 535 | * *G* -- means giga (G) or 1000**3 |
| 536 | * *T* -- means tera (T) or 1000**4 |
| 537 | * *P* -- means peta (P) or 1000**5 |
| 538 | |
| 539 | To specify power-of-2 binary values defined in IEC 80000-13: |
| 540 | |
| 541 | * *Ki* -- means kibi (Ki) or 1024 |
| 542 | * *Mi* -- means mebi (Mi) or 1024**2 |
| 543 | * *Gi* -- means gibi (Gi) or 1024**3 |
| 544 | * *Ti* -- means tebi (Ti) or 1024**4 |
| 545 | * *Pi* -- means pebi (Pi) or 1024**5 |
| 546 | |
| 547 | With :option:`kb_base`\=1024 (the default), the unit prefixes are opposite |
| 548 | from those specified in the SI and IEC 80000-13 standards to provide |
| 549 | compatibility with old scripts. For example, 4k means 4096. |
| 550 | |
| 551 | For quantities of data, an optional unit of 'B' may be included |
| 552 | (e.g., 'kB' is the same as 'k'). |
| 553 | |
| 554 | The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega, |
| 555 | not milli). 'b' and 'B' both mean byte, not bit. |
| 556 | |
| 557 | Examples with :option:`kb_base`\=1000: |
| 558 | |
| 559 | * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB |
| 560 | * *1 MiB*: 1048576, 1mi, 1024ki |
| 561 | * *1 MB*: 1000000, 1m, 1000k |
| 562 | * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi |
| 563 | * *1 TB*: 1000000000, 1t, 1000m, 1000000k |
| 564 | |
| 565 | Examples with :option:`kb_base`\=1024 (default): |
| 566 | |
| 567 | * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB |
| 568 | * *1 MiB*: 1048576, 1m, 1024k |
| 569 | * *1 MB*: 1000000, 1mi, 1000ki |
| 570 | * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m |
| 571 | * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki |
| 572 | |
| 573 | To specify times (units are not case sensitive): |
| 574 | |
| 575 | * *D* -- means days |
| 576 | * *H* -- means hours |
| 577 | * *M* -- means minutes |
| 578 | * *s* -- or sec means seconds (default) |
| 579 | * *ms* -- or *msec* means milliseconds |
| 580 | * *us* -- or *usec* means microseconds |
| 581 | |
| 582 | If the option accepts an upper and lower range, use a colon ':' or |
| 583 | minus '-' to separate such values. See :ref:`irange <irange>`. |
| 584 | If the lower value specified happens to be larger than the upper value |
| 585 | the two values are swapped. |
| 586 | |
| 587 | .. _bool: |
| 588 | |
| 589 | **bool** |
| 590 | Boolean. Usually parsed as an integer, however only defined for |
| 591 | true and false (1 and 0). |
| 592 | |
| 593 | .. _irange: |
| 594 | |
| 595 | **irange** |
| 596 | Integer range with suffix. Allows value range to be given, such as |
| 597 | 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the |
| 598 | option allows two sets of ranges, they can be specified with a ',' or '/' |
| 599 | delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`. |
| 600 | |
| 601 | **float_list** |
| 602 | A list of floating point numbers, separated by a ':' character. |
| 603 | |
| 604 | With the above in mind, here follows the complete list of fio job parameters. |
| 605 | |
| 606 | |
| 607 | Units |
| 608 | ~~~~~ |
| 609 | |
| 610 | .. option:: kb_base=int |
| 611 | |
| 612 | Select the interpretation of unit prefixes in input parameters. |
| 613 | |
| 614 | **1000** |
| 615 | Inputs comply with IEC 80000-13 and the International |
| 616 | System of Units (SI). Use: |
| 617 | |
| 618 | - power-of-2 values with IEC prefixes (e.g., KiB) |
| 619 | - power-of-10 values with SI prefixes (e.g., kB) |
| 620 | |
| 621 | **1024** |
| 622 | Compatibility mode (default). To avoid breaking old scripts: |
| 623 | |
| 624 | - power-of-2 values with SI prefixes |
| 625 | - power-of-10 values with IEC prefixes |
| 626 | |
| 627 | See :option:`bs` for more details on input parameters. |
| 628 | |
| 629 | Outputs always use correct prefixes. Most outputs include both |
| 630 | side-by-side, like:: |
| 631 | |
| 632 | bw=2383.3kB/s (2327.4KiB/s) |
| 633 | |
| 634 | If only one value is reported, then kb_base selects the one to use: |
| 635 | |
| 636 | **1000** -- SI prefixes |
| 637 | |
| 638 | **1024** -- IEC prefixes |
| 639 | |
| 640 | .. option:: unit_base=int |
| 641 | |
| 642 | Base unit for reporting. Allowed values are: |
| 643 | |
| 644 | **0** |
| 645 | Use auto-detection (default). |
| 646 | **8** |
| 647 | Byte based. |
| 648 | **1** |
| 649 | Bit based. |
| 650 | |
| 651 | |
| 652 | Job description |
| 653 | ~~~~~~~~~~~~~~~ |
| 654 | |
| 655 | .. option:: name=str |
| 656 | |
| 657 | ASCII name of the job. This may be used to override the name printed by fio |
| 658 | for this job. Otherwise the job name is used. On the command line this |
| 659 | parameter has the special purpose of also signaling the start of a new job. |
| 660 | |
| 661 | .. option:: description=str |
| 662 | |
| 663 | Text description of the job. Doesn't do anything except dump this text |
| 664 | description when this job is run. It's not parsed. |
| 665 | |
| 666 | .. option:: loops=int |
| 667 | |
| 668 | Run the specified number of iterations of this job. Used to repeat the same |
| 669 | workload a given number of times. Defaults to 1. |
| 670 | |
| 671 | .. option:: numjobs=int |
| 672 | |
| 673 | Create the specified number of clones of this job. Each clone of job |
| 674 | is spawned as an independent thread or process. May be used to setup a |
| 675 | larger number of threads/processes doing the same thing. Each thread is |
| 676 | reported separately; to see statistics for all clones as a whole, use |
| 677 | :option:`group_reporting` in conjunction with :option:`new_group`. |
| 678 | See :option:`--max-jobs`. Default: 1. |
| 679 | |
| 680 | |
| 681 | Time related parameters |
| 682 | ~~~~~~~~~~~~~~~~~~~~~~~ |
| 683 | |
| 684 | .. option:: runtime=time |
| 685 | |
| 686 | Tell fio to terminate processing after the specified period of time. It |
| 687 | can be quite hard to determine for how long a specified job will run, so |
| 688 | this parameter is handy to cap the total runtime to a given time. When |
| 689 | the unit is omitted, the value is interpreted in seconds. |
| 690 | |
| 691 | .. option:: time_based |
| 692 | |
| 693 | If set, fio will run for the duration of the :option:`runtime` specified |
| 694 | even if the file(s) are completely read or written. It will simply loop over |
| 695 | the same workload as many times as the :option:`runtime` allows. |
| 696 | |
| 697 | .. option:: startdelay=irange(time) |
| 698 | |
| 699 | Delay the start of job for the specified amount of time. Can be a single |
| 700 | value or a range. When given as a range, each thread will choose a value |
| 701 | randomly from within the range. Value is in seconds if a unit is omitted. |
| 702 | |
| 703 | .. option:: ramp_time=time |
| 704 | |
| 705 | If set, fio will run the specified workload for this amount of time before |
| 706 | logging any performance numbers. Useful for letting performance settle |
| 707 | before logging results, thus minimizing the runtime required for stable |
| 708 | results. Note that the ``ramp_time`` is considered lead in time for a job, |
| 709 | thus it will increase the total runtime if a special timeout or |
| 710 | :option:`runtime` is specified. When the unit is omitted, the value is |
| 711 | given in seconds. |
| 712 | |
| 713 | .. option:: clocksource=str |
| 714 | |
| 715 | Use the given clocksource as the base of timing. The supported options are: |
| 716 | |
| 717 | **gettimeofday** |
| 718 | :manpage:`gettimeofday(2)` |
| 719 | |
| 720 | **clock_gettime** |
| 721 | :manpage:`clock_gettime(2)` |
| 722 | |
| 723 | **cpu** |
| 724 | Internal CPU clock source |
| 725 | |
| 726 | cpu is the preferred clocksource if it is reliable, as it is very fast (and |
| 727 | fio is heavy on time calls). Fio will automatically use this clocksource if |
| 728 | it's supported and considered reliable on the system it is running on, |
| 729 | unless another clocksource is specifically set. For x86/x86-64 CPUs, this |
| 730 | means supporting TSC Invariant. |
| 731 | |
| 732 | .. option:: gtod_reduce=bool |
| 733 | |
| 734 | Enable all of the :manpage:`gettimeofday(2)` reducing options |
| 735 | (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus |
| 736 | reduce precision of the timeout somewhat to really shrink the |
| 737 | :manpage:`gettimeofday(2)` call count. With this option enabled, we only do |
| 738 | about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all |
| 739 | time keeping was enabled. |
| 740 | |
| 741 | .. option:: gtod_cpu=int |
| 742 | |
| 743 | Sometimes it's cheaper to dedicate a single thread of execution to just |
| 744 | getting the current time. Fio (and databases, for instance) are very |
| 745 | intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set |
| 746 | one CPU aside for doing nothing but logging current time to a shared memory |
| 747 | location. Then the other threads/processes that run I/O workloads need only |
| 748 | copy that segment, instead of entering the kernel with a |
| 749 | :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time |
| 750 | calls will be excluded from other uses. Fio will manually clear it from the |
| 751 | CPU mask of other jobs. |
| 752 | |
| 753 | |
| 754 | Target file/device |
| 755 | ~~~~~~~~~~~~~~~~~~ |
| 756 | |
| 757 | .. option:: directory=str |
| 758 | |
| 759 | Prefix filenames with this directory. Used to place files in a different |
| 760 | location than :file:`./`. You can specify a number of directories by |
| 761 | separating the names with a ':' character. These directories will be |
| 762 | assigned equally distributed to job clones created by :option:`numjobs` as |
| 763 | long as they are using generated filenames. If specific `filename(s)` are |
| 764 | set fio will use the first listed directory, and thereby matching the |
| 765 | `filename` semantic (which generates a file for each clone if not |
| 766 | specified, but lets all clones use the same file if set). |
| 767 | |
| 768 | See the :option:`filename` option for information on how to escape "``:``" |
| 769 | characters within the directory path itself. |
| 770 | |
| 771 | Note: To control the directory fio will use for internal state files |
| 772 | use :option:`--aux-path`. |
| 773 | |
| 774 | .. option:: filename=str |
| 775 | |
| 776 | Fio normally makes up a `filename` based on the job name, thread number, and |
| 777 | file number (see :option:`filename_format`). If you want to share files |
| 778 | between threads in a job or several |
| 779 | jobs with fixed file paths, specify a `filename` for each of them to override |
| 780 | the default. If the ioengine is file based, you can specify a number of files |
| 781 | by separating the names with a ':' colon. So if you wanted a job to open |
| 782 | :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use |
| 783 | ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is |
| 784 | specified, :option:`nrfiles` is ignored. The size of regular files specified |
| 785 | by this option will be :option:`size` divided by number of files unless an |
| 786 | explicit size is specified by :option:`filesize`. |
| 787 | |
| 788 | Each colon in the wanted path must be escaped with a ``\`` |
| 789 | character. For instance, if the path is :file:`/dev/dsk/foo@3,0:c` then you |
| 790 | would use ``filename=/dev/dsk/foo@3,0\:c`` and if the path is |
| 791 | :file:`F:\\filename` then you would use ``filename=F\:\filename``. |
| 792 | |
| 793 | On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for |
| 794 | the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc. |
| 795 | Note: Windows and FreeBSD prevent write access to areas |
| 796 | of the disk containing in-use data (e.g. filesystems). |
| 797 | |
| 798 | The filename "`-`" is a reserved name, meaning *stdin* or *stdout*. Which |
| 799 | of the two depends on the read/write direction set. |
| 800 | |
| 801 | .. option:: filename_format=str |
| 802 | |
| 803 | If sharing multiple files between jobs, it is usually necessary to have fio |
| 804 | generate the exact names that you want. By default, fio will name a file |
| 805 | based on the default file format specification of |
| 806 | :file:`jobname.jobnumber.filenumber`. With this option, that can be |
| 807 | customized. Fio will recognize and replace the following keywords in this |
| 808 | string: |
| 809 | |
| 810 | **$jobname** |
| 811 | The name of the worker thread or process. |
| 812 | **$jobnum** |
| 813 | The incremental number of the worker thread or process. |
| 814 | **$filenum** |
| 815 | The incremental number of the file for that worker thread or |
| 816 | process. |
| 817 | |
| 818 | To have dependent jobs share a set of files, this option can be set to have |
| 819 | fio generate filenames that are shared between the two. For instance, if |
| 820 | :file:`testfiles.$filenum` is specified, file number 4 for any job will be |
| 821 | named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum` |
| 822 | will be used if no other format specifier is given. |
| 823 | |
| 824 | If you specify a path then the directories will be created up to the |
| 825 | main directory for the file. So for example if you specify |
| 826 | ``filename_format=a/b/c/$jobnum`` then the directories a/b/c will be |
| 827 | created before the file setup part of the job. If you specify |
| 828 | :option:`directory` then the path will be relative that directory, |
| 829 | otherwise it is treated as the absolute path. |
| 830 | |
| 831 | .. option:: unique_filename=bool |
| 832 | |
| 833 | To avoid collisions between networked clients, fio defaults to prefixing any |
| 834 | generated filenames (with a directory specified) with the source of the |
| 835 | client connecting. To disable this behavior, set this option to 0. |
| 836 | |
| 837 | .. option:: opendir=str |
| 838 | |
| 839 | Recursively open any files below directory `str`. |
| 840 | |
| 841 | .. option:: lockfile=str |
| 842 | |
| 843 | Fio defaults to not locking any files before it does I/O to them. If a file |
| 844 | or file descriptor is shared, fio can serialize I/O to that file to make the |
| 845 | end result consistent. This is usual for emulating real workloads that share |
| 846 | files. The lock modes are: |
| 847 | |
| 848 | **none** |
| 849 | No locking. The default. |
| 850 | **exclusive** |
| 851 | Only one thread or process may do I/O at a time, excluding all |
| 852 | others. |
| 853 | **readwrite** |
| 854 | Read-write locking on the file. Many readers may |
| 855 | access the file at the same time, but writes get exclusive access. |
| 856 | |
| 857 | .. option:: nrfiles=int |
| 858 | |
| 859 | Number of files to use for this job. Defaults to 1. The size of files |
| 860 | will be :option:`size` divided by this unless explicit size is specified by |
| 861 | :option:`filesize`. Files are created for each thread separately, and each |
| 862 | file will have a file number within its name by default, as explained in |
| 863 | :option:`filename` section. |
| 864 | |
| 865 | |
| 866 | .. option:: openfiles=int |
| 867 | |
| 868 | Number of files to keep open at the same time. Defaults to the same as |
| 869 | :option:`nrfiles`, can be set smaller to limit the number simultaneous |
| 870 | opens. |
| 871 | |
| 872 | .. option:: file_service_type=str |
| 873 | |
| 874 | Defines how fio decides which file from a job to service next. The following |
| 875 | types are defined: |
| 876 | |
| 877 | **random** |
| 878 | Choose a file at random. |
| 879 | |
| 880 | **roundrobin** |
| 881 | Round robin over opened files. This is the default. |
| 882 | |
| 883 | **sequential** |
| 884 | Finish one file before moving on to the next. Multiple files can |
| 885 | still be open depending on :option:`openfiles`. |
| 886 | |
| 887 | **zipf** |
| 888 | Use a *Zipf* distribution to decide what file to access. |
| 889 | |
| 890 | **pareto** |
| 891 | Use a *Pareto* distribution to decide what file to access. |
| 892 | |
| 893 | **normal** |
| 894 | Use a *Gaussian* (normal) distribution to decide what file to |
| 895 | access. |
| 896 | |
| 897 | **gauss** |
| 898 | Alias for normal. |
| 899 | |
| 900 | For *random*, *roundrobin*, and *sequential*, a postfix can be appended to |
| 901 | tell fio how many I/Os to issue before switching to a new file. For example, |
| 902 | specifying ``file_service_type=random:8`` would cause fio to issue |
| 903 | 8 I/Os before selecting a new file at random. For the non-uniform |
| 904 | distributions, a floating point postfix can be given to influence how the |
| 905 | distribution is skewed. See :option:`random_distribution` for a description |
| 906 | of how that would work. |
| 907 | |
| 908 | .. option:: ioscheduler=str |
| 909 | |
| 910 | Attempt to switch the device hosting the file to the specified I/O scheduler |
| 911 | before running. |
| 912 | |
| 913 | .. option:: create_serialize=bool |
| 914 | |
| 915 | If true, serialize the file creation for the jobs. This may be handy to |
| 916 | avoid interleaving of data files, which may greatly depend on the filesystem |
| 917 | used and even the number of processors in the system. Default: true. |
| 918 | |
| 919 | .. option:: create_fsync=bool |
| 920 | |
| 921 | :manpage:`fsync(2)` the data file after creation. This is the default. |
| 922 | |
| 923 | .. option:: create_on_open=bool |
| 924 | |
| 925 | If true, don't pre-create files but allow the job's open() to create a file |
| 926 | when it's time to do I/O. Default: false -- pre-create all necessary files |
| 927 | when the job starts. |
| 928 | |
| 929 | .. option:: create_only=bool |
| 930 | |
| 931 | If true, fio will only run the setup phase of the job. If files need to be |
| 932 | laid out or updated on disk, only that will be done -- the actual job contents |
| 933 | are not executed. Default: false. |
| 934 | |
| 935 | .. option:: allow_file_create=bool |
| 936 | |
| 937 | If true, fio is permitted to create files as part of its workload. If this |
| 938 | option is false, then fio will error out if |
| 939 | the files it needs to use don't already exist. Default: true. |
| 940 | |
| 941 | .. option:: allow_mounted_write=bool |
| 942 | |
| 943 | If this isn't set, fio will abort jobs that are destructive (e.g. that write) |
| 944 | to what appears to be a mounted device or partition. This should help catch |
| 945 | creating inadvertently destructive tests, not realizing that the test will |
| 946 | destroy data on the mounted file system. Note that some platforms don't allow |
| 947 | writing against a mounted device regardless of this option. Default: false. |
| 948 | |
| 949 | .. option:: pre_read=bool |
| 950 | |
| 951 | If this is given, files will be pre-read into memory before starting the |
| 952 | given I/O operation. This will also clear the :option:`invalidate` flag, |
| 953 | since it is pointless to pre-read and then drop the cache. This will only |
| 954 | work for I/O engines that are seek-able, since they allow you to read the |
| 955 | same data multiple times. Thus it will not work on non-seekable I/O engines |
| 956 | (e.g. network, splice). Default: false. |
| 957 | |
| 958 | .. option:: unlink=bool |
| 959 | |
| 960 | Unlink the job files when done. Not the default, as repeated runs of that |
| 961 | job would then waste time recreating the file set again and again. Default: |
| 962 | false. |
| 963 | |
| 964 | .. option:: unlink_each_loop=bool |
| 965 | |
| 966 | Unlink job files after each iteration or loop. Default: false. |
| 967 | |
| 968 | .. option:: zonemode=str |
| 969 | |
| 970 | Accepted values are: |
| 971 | |
| 972 | **none** |
| 973 | The :option:`zonerange`, :option:`zonesize`, |
| 974 | :option `zonecapacity` and option:`zoneskip` |
| 975 | parameters are ignored. |
| 976 | **strided** |
| 977 | I/O happens in a single zone until |
| 978 | :option:`zonesize` bytes have been transferred. |
| 979 | After that number of bytes has been |
| 980 | transferred processing of the next zone |
| 981 | starts. :option `zonecapacity` is ignored. |
| 982 | **zbd** |
| 983 | Zoned block device mode. I/O happens |
| 984 | sequentially in each zone, even if random I/O |
| 985 | has been selected. Random I/O happens across |
| 986 | all zones instead of being restricted to a |
| 987 | single zone. The :option:`zoneskip` parameter |
| 988 | is ignored. :option:`zonerange` and |
| 989 | :option:`zonesize` must be identical. |
| 990 | |
| 991 | .. option:: zonerange=int |
| 992 | |
| 993 | Size of a single zone. See also :option:`zonesize` and |
| 994 | :option:`zoneskip`. |
| 995 | |
| 996 | .. option:: zonesize=int |
| 997 | |
| 998 | For :option:`zonemode` =strided, this is the number of bytes to |
| 999 | transfer before skipping :option:`zoneskip` bytes. If this parameter |
| 1000 | is smaller than :option:`zonerange` then only a fraction of each zone |
| 1001 | with :option:`zonerange` bytes will be accessed. If this parameter is |
| 1002 | larger than :option:`zonerange` then each zone will be accessed |
| 1003 | multiple times before skipping to the next zone. |
| 1004 | |
| 1005 | For :option:`zonemode` =zbd, this is the size of a single zone. The |
| 1006 | :option:`zonerange` parameter is ignored in this mode. |
| 1007 | |
| 1008 | |
| 1009 | .. option:: zonecapacity=int |
| 1010 | |
| 1011 | For :option:`zonemode` =zbd, this defines the capacity of a single zone, |
| 1012 | which is the accessible area starting from the zone start address. |
| 1013 | This parameter only applies when using :option:`zonemode` =zbd in |
| 1014 | combination with regular block devices. If not specified it defaults to |
| 1015 | the zone size. If the target device is a zoned block device, the zone |
| 1016 | capacity is obtained from the device information and this option is |
| 1017 | ignored. |
| 1018 | |
| 1019 | .. option:: zoneskip=int |
| 1020 | |
| 1021 | For :option:`zonemode` =strided, the number of bytes to skip after |
| 1022 | :option:`zonesize` bytes of data have been transferred. This parameter |
| 1023 | must be zero for :option:`zonemode` =zbd. |
| 1024 | |
| 1025 | .. option:: read_beyond_wp=bool |
| 1026 | |
| 1027 | This parameter applies to :option:`zonemode` =zbd only. |
| 1028 | |
| 1029 | Zoned block devices are block devices that consist of multiple zones. |
| 1030 | Each zone has a type, e.g. conventional or sequential. A conventional |
| 1031 | zone can be written at any offset that is a multiple of the block |
| 1032 | size. Sequential zones must be written sequentially. The position at |
| 1033 | which a write must occur is called the write pointer. A zoned block |
| 1034 | device can be either drive managed, host managed or host aware. For |
| 1035 | host managed devices the host must ensure that writes happen |
| 1036 | sequentially. Fio recognizes host managed devices and serializes |
| 1037 | writes to sequential zones for these devices. |
| 1038 | |
| 1039 | If a read occurs in a sequential zone beyond the write pointer then |
| 1040 | the zoned block device will complete the read without reading any data |
| 1041 | from the storage medium. Since such reads lead to unrealistically high |
| 1042 | bandwidth and IOPS numbers fio only reads beyond the write pointer if |
| 1043 | explicitly told to do so. Default: false. |
| 1044 | |
| 1045 | .. option:: max_open_zones=int |
| 1046 | |
| 1047 | When running a random write test across an entire drive many more |
| 1048 | zones will be open than in a typical application workload. Hence this |
| 1049 | command line option that allows to limit the number of open zones. The |
| 1050 | number of open zones is defined as the number of zones to which write |
| 1051 | commands are issued. |
| 1052 | |
| 1053 | .. option:: zone_reset_threshold=float |
| 1054 | |
| 1055 | A number between zero and one that indicates the ratio of logical |
| 1056 | blocks with data to the total number of logical blocks in the test |
| 1057 | above which zones should be reset periodically. |
| 1058 | |
| 1059 | .. option:: zone_reset_frequency=float |
| 1060 | |
| 1061 | A number between zero and one that indicates how often a zone reset |
| 1062 | should be issued if the zone reset threshold has been exceeded. A zone |
| 1063 | reset is submitted after each (1 / zone_reset_frequency) write |
| 1064 | requests. This and the previous parameter can be used to simulate |
| 1065 | garbage collection activity. |
| 1066 | |
| 1067 | |
| 1068 | I/O type |
| 1069 | ~~~~~~~~ |
| 1070 | |
| 1071 | .. option:: direct=bool |
| 1072 | |
| 1073 | If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that |
| 1074 | OpenBSD and ZFS on Solaris don't support direct I/O. On Windows the synchronous |
| 1075 | ioengines don't support direct I/O. Default: false. |
| 1076 | |
| 1077 | .. option:: atomic=bool |
| 1078 | |
| 1079 | If value is true, attempt to use atomic direct I/O. Atomic writes are |
| 1080 | guaranteed to be stable once acknowledged by the operating system. Only |
| 1081 | Linux supports O_ATOMIC right now. |
| 1082 | |
| 1083 | .. option:: buffered=bool |
| 1084 | |
| 1085 | If value is true, use buffered I/O. This is the opposite of the |
| 1086 | :option:`direct` option. Defaults to true. |
| 1087 | |
| 1088 | .. option:: readwrite=str, rw=str |
| 1089 | |
| 1090 | Type of I/O pattern. Accepted values are: |
| 1091 | |
| 1092 | **read** |
| 1093 | Sequential reads. |
| 1094 | **write** |
| 1095 | Sequential writes. |
| 1096 | **trim** |
| 1097 | Sequential trims (Linux block devices and SCSI |
| 1098 | character devices only). |
| 1099 | **randread** |
| 1100 | Random reads. |
| 1101 | **randwrite** |
| 1102 | Random writes. |
| 1103 | **randtrim** |
| 1104 | Random trims (Linux block devices and SCSI |
| 1105 | character devices only). |
| 1106 | **rw,readwrite** |
| 1107 | Sequential mixed reads and writes. |
| 1108 | **randrw** |
| 1109 | Random mixed reads and writes. |
| 1110 | **trimwrite** |
| 1111 | Sequential trim+write sequences. Blocks will be trimmed first, |
| 1112 | then the same blocks will be written to. |
| 1113 | |
| 1114 | Fio defaults to read if the option is not specified. For the mixed I/O |
| 1115 | types, the default is to split them 50/50. For certain types of I/O the |
| 1116 | result may still be skewed a bit, since the speed may be different. |
| 1117 | |
| 1118 | It is possible to specify the number of I/Os to do before getting a new |
| 1119 | offset by appending ``:<nr>`` to the end of the string given. For a |
| 1120 | random read, it would look like ``rw=randread:8`` for passing in an offset |
| 1121 | modifier with a value of 8. If the suffix is used with a sequential I/O |
| 1122 | pattern, then the *<nr>* value specified will be **added** to the generated |
| 1123 | offset for each I/O turning sequential I/O into sequential I/O with holes. |
| 1124 | For instance, using ``rw=write:4k`` will skip 4k for every write. Also see |
| 1125 | the :option:`rw_sequencer` option. |
| 1126 | |
| 1127 | .. option:: rw_sequencer=str |
| 1128 | |
| 1129 | If an offset modifier is given by appending a number to the ``rw=<str>`` |
| 1130 | line, then this option controls how that number modifies the I/O offset |
| 1131 | being generated. Accepted values are: |
| 1132 | |
| 1133 | **sequential** |
| 1134 | Generate sequential offset. |
| 1135 | **identical** |
| 1136 | Generate the same offset. |
| 1137 | |
| 1138 | ``sequential`` is only useful for random I/O, where fio would normally |
| 1139 | generate a new random offset for every I/O. If you append e.g. 8 to randread, |
| 1140 | you would get a new random offset for every 8 I/Os. The result would be a |
| 1141 | seek for only every 8 I/Os, instead of for every I/O. Use ``rw=randread:8`` |
| 1142 | to specify that. As sequential I/O is already sequential, setting |
| 1143 | ``sequential`` for that would not result in any differences. ``identical`` |
| 1144 | behaves in a similar fashion, except it sends the same offset 8 number of |
| 1145 | times before generating a new offset. |
| 1146 | |
| 1147 | .. option:: unified_rw_reporting=bool |
| 1148 | |
| 1149 | Fio normally reports statistics on a per data direction basis, meaning that |
| 1150 | reads, writes, and trims are accounted and reported separately. If this |
| 1151 | option is set fio sums the results and report them as "mixed" instead. |
| 1152 | |
| 1153 | .. option:: randrepeat=bool |
| 1154 | |
| 1155 | Seed the random number generator used for random I/O patterns in a |
| 1156 | predictable way so the pattern is repeatable across runs. Default: true. |
| 1157 | |
| 1158 | .. option:: allrandrepeat=bool |
| 1159 | |
| 1160 | Seed all random number generators in a predictable way so results are |
| 1161 | repeatable across runs. Default: false. |
| 1162 | |
| 1163 | .. option:: randseed=int |
| 1164 | |
| 1165 | Seed the random number generators based on this seed value, to be able to |
| 1166 | control what sequence of output is being generated. If not set, the random |
| 1167 | sequence depends on the :option:`randrepeat` setting. |
| 1168 | |
| 1169 | .. option:: fallocate=str |
| 1170 | |
| 1171 | Whether pre-allocation is performed when laying down files. |
| 1172 | Accepted values are: |
| 1173 | |
| 1174 | **none** |
| 1175 | Do not pre-allocate space. |
| 1176 | |
| 1177 | **native** |
| 1178 | Use a platform's native pre-allocation call but fall back to |
| 1179 | **none** behavior if it fails/is not implemented. |
| 1180 | |
| 1181 | **posix** |
| 1182 | Pre-allocate via :manpage:`posix_fallocate(3)`. |
| 1183 | |
| 1184 | **keep** |
| 1185 | Pre-allocate via :manpage:`fallocate(2)` with |
| 1186 | FALLOC_FL_KEEP_SIZE set. |
| 1187 | |
| 1188 | **truncate** |
| 1189 | Extend file to final size via :manpage:`ftruncate(2)` |
| 1190 | instead of allocating. |
| 1191 | |
| 1192 | **0** |
| 1193 | Backward-compatible alias for **none**. |
| 1194 | |
| 1195 | **1** |
| 1196 | Backward-compatible alias for **posix**. |
| 1197 | |
| 1198 | May not be available on all supported platforms. **keep** is only available |
| 1199 | on Linux. If using ZFS on Solaris this cannot be set to **posix** |
| 1200 | because ZFS doesn't support pre-allocation. Default: **native** if any |
| 1201 | pre-allocation methods except **truncate** are available, **none** if not. |
| 1202 | |
| 1203 | Note that using **truncate** on Windows will interact surprisingly |
| 1204 | with non-sequential write patterns. When writing to a file that has |
| 1205 | been extended by setting the end-of-file information, Windows will |
| 1206 | backfill the unwritten portion of the file up to that offset with |
| 1207 | zeroes before issuing the new write. This means that a single small |
| 1208 | write to the end of an extended file will stall until the entire |
| 1209 | file has been filled with zeroes. |
| 1210 | |
| 1211 | .. option:: fadvise_hint=str |
| 1212 | |
| 1213 | Use :manpage:`posix_fadvise(2)` or :manpage:`posix_fadvise(2)` to |
| 1214 | advise the kernel on what I/O patterns are likely to be issued. |
| 1215 | Accepted values are: |
| 1216 | |
| 1217 | **0** |
| 1218 | Backwards-compatible hint for "no hint". |
| 1219 | |
| 1220 | **1** |
| 1221 | Backwards compatible hint for "advise with fio workload type". This |
| 1222 | uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL** |
| 1223 | for a sequential workload. |
| 1224 | |
| 1225 | **sequential** |
| 1226 | Advise using **FADV_SEQUENTIAL**. |
| 1227 | |
| 1228 | **random** |
| 1229 | Advise using **FADV_RANDOM**. |
| 1230 | |
| 1231 | .. option:: write_hint=str |
| 1232 | |
| 1233 | Use :manpage:`fcntl(2)` to advise the kernel what life time to expect |
| 1234 | from a write. Only supported on Linux, as of version 4.13. Accepted |
| 1235 | values are: |
| 1236 | |
| 1237 | **none** |
| 1238 | No particular life time associated with this file. |
| 1239 | |
| 1240 | **short** |
| 1241 | Data written to this file has a short life time. |
| 1242 | |
| 1243 | **medium** |
| 1244 | Data written to this file has a medium life time. |
| 1245 | |
| 1246 | **long** |
| 1247 | Data written to this file has a long life time. |
| 1248 | |
| 1249 | **extreme** |
| 1250 | Data written to this file has a very long life time. |
| 1251 | |
| 1252 | The values are all relative to each other, and no absolute meaning |
| 1253 | should be associated with them. |
| 1254 | |
| 1255 | .. option:: offset=int |
| 1256 | |
| 1257 | Start I/O at the provided offset in the file, given as either a fixed size in |
| 1258 | bytes or a percentage. If a percentage is given, the generated offset will be |
| 1259 | aligned to the minimum ``blocksize`` or to the value of ``offset_align`` if |
| 1260 | provided. Data before the given offset will not be touched. This |
| 1261 | effectively caps the file size at `real_size - offset`. Can be combined with |
| 1262 | :option:`size` to constrain the start and end range of the I/O workload. |
| 1263 | A percentage can be specified by a number between 1 and 100 followed by '%', |
| 1264 | for example, ``offset=20%`` to specify 20%. |
| 1265 | |
| 1266 | .. option:: offset_align=int |
| 1267 | |
| 1268 | If set to non-zero value, the byte offset generated by a percentage ``offset`` |
| 1269 | is aligned upwards to this value. Defaults to 0 meaning that a percentage |
| 1270 | offset is aligned to the minimum block size. |
| 1271 | |
| 1272 | .. option:: offset_increment=int |
| 1273 | |
| 1274 | If this is provided, then the real offset becomes `offset + offset_increment |
| 1275 | * thread_number`, where the thread number is a counter that starts at 0 and |
| 1276 | is incremented for each sub-job (i.e. when :option:`numjobs` option is |
| 1277 | specified). This option is useful if there are several jobs which are |
| 1278 | intended to operate on a file in parallel disjoint segments, with even |
| 1279 | spacing between the starting points. Percentages can be used for this option. |
| 1280 | If a percentage is given, the generated offset will be aligned to the minimum |
| 1281 | ``blocksize`` or to the value of ``offset_align`` if provided. |
| 1282 | |
| 1283 | .. option:: number_ios=int |
| 1284 | |
| 1285 | Fio will normally perform I/Os until it has exhausted the size of the region |
| 1286 | set by :option:`size`, or if it exhaust the allocated time (or hits an error |
| 1287 | condition). With this setting, the range/size can be set independently of |
| 1288 | the number of I/Os to perform. When fio reaches this number, it will exit |
| 1289 | normally and report status. Note that this does not extend the amount of I/O |
| 1290 | that will be done, it will only stop fio if this condition is met before |
| 1291 | other end-of-job criteria. |
| 1292 | |
| 1293 | .. option:: fsync=int |
| 1294 | |
| 1295 | If writing to a file, issue an :manpage:`fsync(2)` (or its equivalent) of |
| 1296 | the dirty data for every number of blocks given. For example, if you give 32 |
| 1297 | as a parameter, fio will sync the file after every 32 writes issued. If fio is |
| 1298 | using non-buffered I/O, we may not sync the file. The exception is the sg |
| 1299 | I/O engine, which synchronizes the disk cache anyway. Defaults to 0, which |
| 1300 | means fio does not periodically issue and wait for a sync to complete. Also |
| 1301 | see :option:`end_fsync` and :option:`fsync_on_close`. |
| 1302 | |
| 1303 | .. option:: fdatasync=int |
| 1304 | |
| 1305 | Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and |
| 1306 | not metadata blocks. In Windows, FreeBSD, DragonFlyBSD or OSX there is no |
| 1307 | :manpage:`fdatasync(2)` so this falls back to using :manpage:`fsync(2)`. |
| 1308 | Defaults to 0, which means fio does not periodically issue and wait for a |
| 1309 | data-only sync to complete. |
| 1310 | |
| 1311 | .. option:: write_barrier=int |
| 1312 | |
| 1313 | Make every `N-th` write a barrier write. |
| 1314 | |
| 1315 | .. option:: sync_file_range=str:int |
| 1316 | |
| 1317 | Use :manpage:`sync_file_range(2)` for every `int` number of write |
| 1318 | operations. Fio will track range of writes that have happened since the last |
| 1319 | :manpage:`sync_file_range(2)` call. `str` can currently be one or more of: |
| 1320 | |
| 1321 | **wait_before** |
| 1322 | SYNC_FILE_RANGE_WAIT_BEFORE |
| 1323 | **write** |
| 1324 | SYNC_FILE_RANGE_WRITE |
| 1325 | **wait_after** |
| 1326 | SYNC_FILE_RANGE_WAIT_AFTER |
| 1327 | |
| 1328 | So if you do ``sync_file_range=wait_before,write:8``, fio would use |
| 1329 | ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8 |
| 1330 | writes. Also see the :manpage:`sync_file_range(2)` man page. This option is |
| 1331 | Linux specific. |
| 1332 | |
| 1333 | .. option:: overwrite=bool |
| 1334 | |
| 1335 | If true, writes to a file will always overwrite existing data. If the file |
| 1336 | doesn't already exist, it will be created before the write phase begins. If |
| 1337 | the file exists and is large enough for the specified write phase, nothing |
| 1338 | will be done. Default: false. |
| 1339 | |
| 1340 | .. option:: end_fsync=bool |
| 1341 | |
| 1342 | If true, :manpage:`fsync(2)` file contents when a write stage has completed. |
| 1343 | Default: false. |
| 1344 | |
| 1345 | .. option:: fsync_on_close=bool |
| 1346 | |
| 1347 | If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs |
| 1348 | from :option:`end_fsync` in that it will happen on every file close, not |
| 1349 | just at the end of the job. Default: false. |
| 1350 | |
| 1351 | .. option:: rwmixread=int |
| 1352 | |
| 1353 | Percentage of a mixed workload that should be reads. Default: 50. |
| 1354 | |
| 1355 | .. option:: rwmixwrite=int |
| 1356 | |
| 1357 | Percentage of a mixed workload that should be writes. If both |
| 1358 | :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not |
| 1359 | add up to 100%, the latter of the two will be used to override the |
| 1360 | first. This may interfere with a given rate setting, if fio is asked to |
| 1361 | limit reads or writes to a certain rate. If that is the case, then the |
| 1362 | distribution may be skewed. Default: 50. |
| 1363 | |
| 1364 | .. option:: random_distribution=str:float[,str:float][,str:float] |
| 1365 | |
| 1366 | By default, fio will use a completely uniform random distribution when asked |
| 1367 | to perform random I/O. Sometimes it is useful to skew the distribution in |
| 1368 | specific ways, ensuring that some parts of the data is more hot than others. |
| 1369 | fio includes the following distribution models: |
| 1370 | |
| 1371 | **random** |
| 1372 | Uniform random distribution |
| 1373 | |
| 1374 | **zipf** |
| 1375 | Zipf distribution |
| 1376 | |
| 1377 | **pareto** |
| 1378 | Pareto distribution |
| 1379 | |
| 1380 | **normal** |
| 1381 | Normal (Gaussian) distribution |
| 1382 | |
| 1383 | **zoned** |
| 1384 | Zoned random distribution |
| 1385 | |
| 1386 | **zoned_abs** |
| 1387 | Zone absolute random distribution |
| 1388 | |
| 1389 | When using a **zipf** or **pareto** distribution, an input value is also |
| 1390 | needed to define the access pattern. For **zipf**, this is the `Zipf |
| 1391 | theta`. For **pareto**, it's the `Pareto power`. Fio includes a test |
| 1392 | program, :command:`fio-genzipf`, that can be used visualize what the given input |
| 1393 | values will yield in terms of hit rates. If you wanted to use **zipf** with |
| 1394 | a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the |
| 1395 | option. If a non-uniform model is used, fio will disable use of the random |
| 1396 | map. For the **normal** distribution, a normal (Gaussian) deviation is |
| 1397 | supplied as a value between 0 and 100. |
| 1398 | |
| 1399 | For a **zoned** distribution, fio supports specifying percentages of I/O |
| 1400 | access that should fall within what range of the file or device. For |
| 1401 | example, given a criteria of: |
| 1402 | |
| 1403 | * 60% of accesses should be to the first 10% |
| 1404 | * 30% of accesses should be to the next 20% |
| 1405 | * 8% of accesses should be to the next 30% |
| 1406 | * 2% of accesses should be to the next 40% |
| 1407 | |
| 1408 | we can define that through zoning of the random accesses. For the above |
| 1409 | example, the user would do:: |
| 1410 | |
| 1411 | random_distribution=zoned:60/10:30/20:8/30:2/40 |
| 1412 | |
| 1413 | A **zoned_abs** distribution works exactly like the **zoned**, except |
| 1414 | that it takes absolute sizes. For example, let's say you wanted to |
| 1415 | define access according to the following criteria: |
| 1416 | |
| 1417 | * 60% of accesses should be to the first 20G |
| 1418 | * 30% of accesses should be to the next 100G |
| 1419 | * 10% of accesses should be to the next 500G |
| 1420 | |
| 1421 | we can define an absolute zoning distribution with: |
| 1422 | |
| 1423 | random_distribution=zoned_abs=60/20G:30/100G:10/500g |
| 1424 | |
| 1425 | For both **zoned** and **zoned_abs**, fio supports defining up to |
| 1426 | 256 separate zones. |
| 1427 | |
| 1428 | Similarly to how :option:`bssplit` works for setting ranges and |
| 1429 | percentages of block sizes. Like :option:`bssplit`, it's possible to |
| 1430 | specify separate zones for reads, writes, and trims. If just one set |
| 1431 | is given, it'll apply to all of them. This goes for both **zoned** |
| 1432 | **zoned_abs** distributions. |
| 1433 | |
| 1434 | .. option:: percentage_random=int[,int][,int] |
| 1435 | |
| 1436 | For a random workload, set how big a percentage should be random. This |
| 1437 | defaults to 100%, in which case the workload is fully random. It can be set |
| 1438 | from anywhere from 0 to 100. Setting it to 0 would make the workload fully |
| 1439 | sequential. Any setting in between will result in a random mix of sequential |
| 1440 | and random I/O, at the given percentages. Comma-separated values may be |
| 1441 | specified for reads, writes, and trims as described in :option:`blocksize`. |
| 1442 | |
| 1443 | .. option:: norandommap |
| 1444 | |
| 1445 | Normally fio will cover every block of the file when doing random I/O. If |
| 1446 | this option is given, fio will just get a new random offset without looking |
| 1447 | at past I/O history. This means that some blocks may not be read or written, |
| 1448 | and that some blocks may be read/written more than once. If this option is |
| 1449 | used with :option:`verify` and multiple blocksizes (via :option:`bsrange`), |
| 1450 | only intact blocks are verified, i.e., partially-overwritten blocks are |
| 1451 | ignored. With an async I/O engine and an I/O depth > 1, it is possible for |
| 1452 | the same block to be overwritten, which can cause verification errors. Either |
| 1453 | do not use norandommap in this case, or also use the lfsr random generator. |
| 1454 | |
| 1455 | .. option:: softrandommap=bool |
| 1456 | |
| 1457 | See :option:`norandommap`. If fio runs with the random block map enabled and |
| 1458 | it fails to allocate the map, if this option is set it will continue without |
| 1459 | a random block map. As coverage will not be as complete as with random maps, |
| 1460 | this option is disabled by default. |
| 1461 | |
| 1462 | .. option:: random_generator=str |
| 1463 | |
| 1464 | Fio supports the following engines for generating I/O offsets for random I/O: |
| 1465 | |
| 1466 | **tausworthe** |
| 1467 | Strong 2^88 cycle random number generator. |
| 1468 | **lfsr** |
| 1469 | Linear feedback shift register generator. |
| 1470 | **tausworthe64** |
| 1471 | Strong 64-bit 2^258 cycle random number generator. |
| 1472 | |
| 1473 | **tausworthe** is a strong random number generator, but it requires tracking |
| 1474 | on the side if we want to ensure that blocks are only read or written |
| 1475 | once. **lfsr** guarantees that we never generate the same offset twice, and |
| 1476 | it's also less computationally expensive. It's not a true random generator, |
| 1477 | however, though for I/O purposes it's typically good enough. **lfsr** only |
| 1478 | works with single block sizes, not with workloads that use multiple block |
| 1479 | sizes. If used with such a workload, fio may read or write some blocks |
| 1480 | multiple times. The default value is **tausworthe**, unless the required |
| 1481 | space exceeds 2^32 blocks. If it does, then **tausworthe64** is |
| 1482 | selected automatically. |
| 1483 | |
| 1484 | |
| 1485 | Block size |
| 1486 | ~~~~~~~~~~ |
| 1487 | |
| 1488 | .. option:: blocksize=int[,int][,int], bs=int[,int][,int] |
| 1489 | |
| 1490 | The block size in bytes used for I/O units. Default: 4096. A single value |
| 1491 | applies to reads, writes, and trims. Comma-separated values may be |
| 1492 | specified for reads, writes, and trims. A value not terminated in a comma |
| 1493 | applies to subsequent types. |
| 1494 | |
| 1495 | Examples: |
| 1496 | |
| 1497 | **bs=256k** |
| 1498 | means 256k for reads, writes and trims. |
| 1499 | |
| 1500 | **bs=8k,32k** |
| 1501 | means 8k for reads, 32k for writes and trims. |
| 1502 | |
| 1503 | **bs=8k,32k,** |
| 1504 | means 8k for reads, 32k for writes, and default for trims. |
| 1505 | |
| 1506 | **bs=,8k** |
| 1507 | means default for reads, 8k for writes and trims. |
| 1508 | |
| 1509 | **bs=,8k,** |
| 1510 | means default for reads, 8k for writes, and default for trims. |
| 1511 | |
| 1512 | .. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange] |
| 1513 | |
| 1514 | A range of block sizes in bytes for I/O units. The issued I/O unit will |
| 1515 | always be a multiple of the minimum size, unless |
| 1516 | :option:`blocksize_unaligned` is set. |
| 1517 | |
| 1518 | Comma-separated ranges may be specified for reads, writes, and trims as |
| 1519 | described in :option:`blocksize`. |
| 1520 | |
| 1521 | Example: ``bsrange=1k-4k,2k-8k``. |
| 1522 | |
| 1523 | .. option:: bssplit=str[,str][,str] |
| 1524 | |
| 1525 | Sometimes you want even finer grained control of the block sizes |
| 1526 | issued, not just an even split between them. This option allows you to |
| 1527 | weight various block sizes, so that you are able to define a specific |
| 1528 | amount of block sizes issued. The format for this option is:: |
| 1529 | |
| 1530 | bssplit=blocksize/percentage:blocksize/percentage |
| 1531 | |
| 1532 | for as many block sizes as needed. So if you want to define a workload |
| 1533 | that has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would |
| 1534 | write:: |
| 1535 | |
| 1536 | bssplit=4k/10:64k/50:32k/40 |
| 1537 | |
| 1538 | Ordering does not matter. If the percentage is left blank, fio will |
| 1539 | fill in the remaining values evenly. So a bssplit option like this one:: |
| 1540 | |
| 1541 | bssplit=4k/50:1k/:32k/ |
| 1542 | |
| 1543 | would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always |
| 1544 | add up to 100, if bssplit is given a range that adds up to more, it |
| 1545 | will error out. |
| 1546 | |
| 1547 | Comma-separated values may be specified for reads, writes, and trims as |
| 1548 | described in :option:`blocksize`. |
| 1549 | |
| 1550 | If you want a workload that has 50% 2k reads and 50% 4k reads, while |
| 1551 | having 90% 4k writes and 10% 8k writes, you would specify:: |
| 1552 | |
| 1553 | bssplit=2k/50:4k/50,4k/90:8k/10 |
| 1554 | |
| 1555 | Fio supports defining up to 64 different weights for each data |
| 1556 | direction. |
| 1557 | |
| 1558 | .. option:: blocksize_unaligned, bs_unaligned |
| 1559 | |
| 1560 | If set, fio will issue I/O units with any size within |
| 1561 | :option:`blocksize_range`, not just multiples of the minimum size. This |
| 1562 | typically won't work with direct I/O, as that normally requires sector |
| 1563 | alignment. |
| 1564 | |
| 1565 | .. option:: bs_is_seq_rand=bool |
| 1566 | |
| 1567 | If this option is set, fio will use the normal read,write blocksize settings |
| 1568 | as sequential,random blocksize settings instead. Any random read or write |
| 1569 | will use the WRITE blocksize settings, and any sequential read or write will |
| 1570 | use the READ blocksize settings. |
| 1571 | |
| 1572 | .. option:: blockalign=int[,int][,int], ba=int[,int][,int] |
| 1573 | |
| 1574 | Boundary to which fio will align random I/O units. Default: |
| 1575 | :option:`blocksize`. Minimum alignment is typically 512b for using direct |
| 1576 | I/O, though it usually depends on the hardware block size. This option is |
| 1577 | mutually exclusive with using a random map for files, so it will turn off |
| 1578 | that option. Comma-separated values may be specified for reads, writes, and |
| 1579 | trims as described in :option:`blocksize`. |
| 1580 | |
| 1581 | |
| 1582 | Buffers and memory |
| 1583 | ~~~~~~~~~~~~~~~~~~ |
| 1584 | |
| 1585 | .. option:: zero_buffers |
| 1586 | |
| 1587 | Initialize buffers with all zeros. Default: fill buffers with random data. |
| 1588 | |
| 1589 | .. option:: refill_buffers |
| 1590 | |
| 1591 | If this option is given, fio will refill the I/O buffers on every |
| 1592 | submit. Only makes sense if :option:`zero_buffers` isn't specified, |
| 1593 | naturally. Defaults to being unset i.e., the buffer is only filled at |
| 1594 | init time and the data in it is reused when possible but if any of |
| 1595 | :option:`verify`, :option:`buffer_compress_percentage` or |
| 1596 | :option:`dedupe_percentage` are enabled then `refill_buffers` is also |
| 1597 | automatically enabled. |
| 1598 | |
| 1599 | .. option:: scramble_buffers=bool |
| 1600 | |
| 1601 | If :option:`refill_buffers` is too costly and the target is using data |
| 1602 | deduplication, then setting this option will slightly modify the I/O buffer |
| 1603 | contents to defeat normal de-dupe attempts. This is not enough to defeat |
| 1604 | more clever block compression attempts, but it will stop naive dedupe of |
| 1605 | blocks. Default: true. |
| 1606 | |
| 1607 | .. option:: buffer_compress_percentage=int |
| 1608 | |
| 1609 | If this is set, then fio will attempt to provide I/O buffer content |
| 1610 | (on WRITEs) that compresses to the specified level. Fio does this by |
| 1611 | providing a mix of random data followed by fixed pattern data. The |
| 1612 | fixed pattern is either zeros, or the pattern specified by |
| 1613 | :option:`buffer_pattern`. If the `buffer_pattern` option is used, it |
| 1614 | might skew the compression ratio slightly. Setting |
| 1615 | `buffer_compress_percentage` to a value other than 100 will also |
| 1616 | enable :option:`refill_buffers` in order to reduce the likelihood that |
| 1617 | adjacent blocks are so similar that they over compress when seen |
| 1618 | together. See :option:`buffer_compress_chunk` for how to set a finer or |
| 1619 | coarser granularity for the random/fixed data region. Defaults to unset |
| 1620 | i.e., buffer data will not adhere to any compression level. |
| 1621 | |
| 1622 | .. option:: buffer_compress_chunk=int |
| 1623 | |
| 1624 | This setting allows fio to manage how big the random/fixed data region |
| 1625 | is when using :option:`buffer_compress_percentage`. When |
| 1626 | `buffer_compress_chunk` is set to some non-zero value smaller than the |
| 1627 | block size, fio can repeat the random/fixed region throughout the I/O |
| 1628 | buffer at the specified interval (which particularly useful when |
| 1629 | bigger block sizes are used for a job). When set to 0, fio will use a |
| 1630 | chunk size that matches the block size resulting in a single |
| 1631 | random/fixed region within the I/O buffer. Defaults to 512. When the |
| 1632 | unit is omitted, the value is interpreted in bytes. |
| 1633 | |
| 1634 | .. option:: buffer_pattern=str |
| 1635 | |
| 1636 | If set, fio will fill the I/O buffers with this pattern or with the contents |
| 1637 | of a file. If not set, the contents of I/O buffers are defined by the other |
| 1638 | options related to buffer contents. The setting can be any pattern of bytes, |
| 1639 | and can be prefixed with 0x for hex values. It may also be a string, where |
| 1640 | the string must then be wrapped with ``""``. Or it may also be a filename, |
| 1641 | where the filename must be wrapped with ``''`` in which case the file is |
| 1642 | opened and read. Note that not all the file contents will be read if that |
| 1643 | would cause the buffers to overflow. So, for example:: |
| 1644 | |
| 1645 | buffer_pattern='filename' |
| 1646 | |
| 1647 | or:: |
| 1648 | |
| 1649 | buffer_pattern="abcd" |
| 1650 | |
| 1651 | or:: |
| 1652 | |
| 1653 | buffer_pattern=-12 |
| 1654 | |
| 1655 | or:: |
| 1656 | |
| 1657 | buffer_pattern=0xdeadface |
| 1658 | |
| 1659 | Also you can combine everything together in any order:: |
| 1660 | |
| 1661 | buffer_pattern=0xdeadface"abcd"-12'filename' |
| 1662 | |
| 1663 | .. option:: dedupe_percentage=int |
| 1664 | |
| 1665 | If set, fio will generate this percentage of identical buffers when |
| 1666 | writing. These buffers will be naturally dedupable. The contents of the |
| 1667 | buffers depend on what other buffer compression settings have been set. It's |
| 1668 | possible to have the individual buffers either fully compressible, or not at |
| 1669 | all -- this option only controls the distribution of unique buffers. Setting |
| 1670 | this option will also enable :option:`refill_buffers` to prevent every buffer |
| 1671 | being identical. |
| 1672 | |
| 1673 | .. option:: invalidate=bool |
| 1674 | |
| 1675 | Invalidate the buffer/page cache parts of the files to be used prior to |
| 1676 | starting I/O if the platform and file type support it. Defaults to true. |
| 1677 | This will be ignored if :option:`pre_read` is also specified for the |
| 1678 | same job. |
| 1679 | |
| 1680 | .. option:: sync=bool |
| 1681 | |
| 1682 | Use synchronous I/O for buffered writes. For the majority of I/O engines, |
| 1683 | this means using O_SYNC. Default: false. |
| 1684 | |
| 1685 | .. option:: iomem=str, mem=str |
| 1686 | |
| 1687 | Fio can use various types of memory as the I/O unit buffer. The allowed |
| 1688 | values are: |
| 1689 | |
| 1690 | **malloc** |
| 1691 | Use memory from :manpage:`malloc(3)` as the buffers. Default memory |
| 1692 | type. |
| 1693 | |
| 1694 | **shm** |
| 1695 | Use shared memory as the buffers. Allocated through |
| 1696 | :manpage:`shmget(2)`. |
| 1697 | |
| 1698 | **shmhuge** |
| 1699 | Same as shm, but use huge pages as backing. |
| 1700 | |
| 1701 | **mmap** |
| 1702 | Use :manpage:`mmap(2)` to allocate buffers. May either be anonymous memory, or can |
| 1703 | be file backed if a filename is given after the option. The format |
| 1704 | is `mem=mmap:/path/to/file`. |
| 1705 | |
| 1706 | **mmaphuge** |
| 1707 | Use a memory mapped huge file as the buffer backing. Append filename |
| 1708 | after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`. |
| 1709 | |
| 1710 | **mmapshared** |
| 1711 | Same as mmap, but use a MMAP_SHARED mapping. |
| 1712 | |
| 1713 | **cudamalloc** |
| 1714 | Use GPU memory as the buffers for GPUDirect RDMA benchmark. |
| 1715 | The :option:`ioengine` must be `rdma`. |
| 1716 | |
| 1717 | The area allocated is a function of the maximum allowed bs size for the job, |
| 1718 | multiplied by the I/O depth given. Note that for **shmhuge** and |
| 1719 | **mmaphuge** to work, the system must have free huge pages allocated. This |
| 1720 | can normally be checked and set by reading/writing |
| 1721 | :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page |
| 1722 | is 4MiB in size. So to calculate the number of huge pages you need for a |
| 1723 | given job file, add up the I/O depth of all jobs (normally one unless |
| 1724 | :option:`iodepth` is used) and multiply by the maximum bs set. Then divide |
| 1725 | that number by the huge page size. You can see the size of the huge pages in |
| 1726 | :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero |
| 1727 | number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also |
| 1728 | see :option:`hugepage-size`. |
| 1729 | |
| 1730 | **mmaphuge** also needs to have hugetlbfs mounted and the file location |
| 1731 | should point there. So if it's mounted in :file:`/huge`, you would use |
| 1732 | `mem=mmaphuge:/huge/somefile`. |
| 1733 | |
| 1734 | .. option:: iomem_align=int, mem_align=int |
| 1735 | |
| 1736 | This indicates the memory alignment of the I/O memory buffers. Note that |
| 1737 | the given alignment is applied to the first I/O unit buffer, if using |
| 1738 | :option:`iodepth` the alignment of the following buffers are given by the |
| 1739 | :option:`bs` used. In other words, if using a :option:`bs` that is a |
| 1740 | multiple of the page sized in the system, all buffers will be aligned to |
| 1741 | this value. If using a :option:`bs` that is not page aligned, the alignment |
| 1742 | of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and |
| 1743 | :option:`bs` used. |
| 1744 | |
| 1745 | .. option:: hugepage-size=int |
| 1746 | |
| 1747 | Defines the size of a huge page. Must at least be equal to the system |
| 1748 | setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably |
| 1749 | always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the |
| 1750 | preferred way to set this to avoid setting a non-pow-2 bad value. |
| 1751 | |
| 1752 | .. option:: lockmem=int |
| 1753 | |
| 1754 | Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to |
| 1755 | simulate a smaller amount of memory. The amount specified is per worker. |
| 1756 | |
| 1757 | |
| 1758 | I/O size |
| 1759 | ~~~~~~~~ |
| 1760 | |
| 1761 | .. option:: size=int |
| 1762 | |
| 1763 | The total size of file I/O for each thread of this job. Fio will run until |
| 1764 | this many bytes has been transferred, unless runtime is limited by other options |
| 1765 | (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`). |
| 1766 | Fio will divide this size between the available files determined by options |
| 1767 | such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is |
| 1768 | specified by the job. If the result of division happens to be 0, the size is |
| 1769 | set to the physical size of the given files or devices if they exist. |
| 1770 | If this option is not specified, fio will use the full size of the given |
| 1771 | files or devices. If the files do not exist, size must be given. It is also |
| 1772 | possible to give size as a percentage between 1 and 100. If ``size=20%`` is |
| 1773 | given, fio will use 20% of the full size of the given files or devices. |
| 1774 | Can be combined with :option:`offset` to constrain the start and end range |
| 1775 | that I/O will be done within. |
| 1776 | |
| 1777 | .. option:: io_size=int, io_limit=int |
| 1778 | |
| 1779 | Normally fio operates within the region set by :option:`size`, which means |
| 1780 | that the :option:`size` option sets both the region and size of I/O to be |
| 1781 | performed. Sometimes that is not what you want. With this option, it is |
| 1782 | possible to define just the amount of I/O that fio should do. For instance, |
| 1783 | if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio |
| 1784 | will perform I/O within the first 20GiB but exit when 5GiB have been |
| 1785 | done. The opposite is also possible -- if :option:`size` is set to 20GiB, |
| 1786 | and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within |
| 1787 | the 0..20GiB region. |
| 1788 | |
| 1789 | .. option:: filesize=irange(int) |
| 1790 | |
| 1791 | Individual file sizes. May be a range, in which case fio will select sizes |
| 1792 | for files at random within the given range and limited to :option:`size` in |
| 1793 | total (if that is given). If not given, each created file is the same size. |
| 1794 | This option overrides :option:`size` in terms of file size, which means |
| 1795 | this value is used as a fixed size or possible range of each file. |
| 1796 | |
| 1797 | .. option:: file_append=bool |
| 1798 | |
| 1799 | Perform I/O after the end of the file. Normally fio will operate within the |
| 1800 | size of a file. If this option is set, then fio will append to the file |
| 1801 | instead. This has identical behavior to setting :option:`offset` to the size |
| 1802 | of a file. This option is ignored on non-regular files. |
| 1803 | |
| 1804 | .. option:: fill_device=bool, fill_fs=bool |
| 1805 | |
| 1806 | Sets size to something really large and waits for ENOSPC (no space left on |
| 1807 | device) as the terminating condition. Only makes sense with sequential |
| 1808 | write. For a read workload, the mount point will be filled first then I/O |
| 1809 | started on the result. This option doesn't make sense if operating on a raw |
| 1810 | device node, since the size of that is already known by the file system. |
| 1811 | Additionally, writing beyond end-of-device will not return ENOSPC there. |
| 1812 | |
| 1813 | |
| 1814 | I/O engine |
| 1815 | ~~~~~~~~~~ |
| 1816 | |
| 1817 | .. option:: ioengine=str |
| 1818 | |
| 1819 | Defines how the job issues I/O to the file. The following types are defined: |
| 1820 | |
| 1821 | **sync** |
| 1822 | Basic :manpage:`read(2)` or :manpage:`write(2)` |
| 1823 | I/O. :manpage:`lseek(2)` is used to position the I/O location. |
| 1824 | See :option:`fsync` and :option:`fdatasync` for syncing write I/Os. |
| 1825 | |
| 1826 | **psync** |
| 1827 | Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on |
| 1828 | all supported operating systems except for Windows. |
| 1829 | |
| 1830 | **vsync** |
| 1831 | Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate |
| 1832 | queuing by coalescing adjacent I/Os into a single submission. |
| 1833 | |
| 1834 | **pvsync** |
| 1835 | Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O. |
| 1836 | |
| 1837 | **pvsync2** |
| 1838 | Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O. |
| 1839 | |
| 1840 | **io_uring** |
| 1841 | Fast Linux native asynchronous I/O. Supports async IO |
| 1842 | for both direct and buffered IO. |
| 1843 | This engine defines engine specific options. |
| 1844 | |
| 1845 | **libaio** |
| 1846 | Linux native asynchronous I/O. Note that Linux may only support |
| 1847 | queued behavior with non-buffered I/O (set ``direct=1`` or |
| 1848 | ``buffered=0``). |
| 1849 | This engine defines engine specific options. |
| 1850 | |
| 1851 | **posixaio** |
| 1852 | POSIX asynchronous I/O using :manpage:`aio_read(3)` and |
| 1853 | :manpage:`aio_write(3)`. |
| 1854 | |
| 1855 | **solarisaio** |
| 1856 | Solaris native asynchronous I/O. |
| 1857 | |
| 1858 | **windowsaio** |
| 1859 | Windows native asynchronous I/O. Default on Windows. |
| 1860 | |
| 1861 | **mmap** |
| 1862 | File is memory mapped with :manpage:`mmap(2)` and data copied |
| 1863 | to/from using :manpage:`memcpy(3)`. |
| 1864 | |
| 1865 | **splice** |
| 1866 | :manpage:`splice(2)` is used to transfer the data and |
| 1867 | :manpage:`vmsplice(2)` to transfer data from user space to the |
| 1868 | kernel. |
| 1869 | |
| 1870 | **sg** |
| 1871 | SCSI generic sg v3 I/O. May either be synchronous using the SG_IO |
| 1872 | ioctl, or if the target is an sg character device we use |
| 1873 | :manpage:`read(2)` and :manpage:`write(2)` for asynchronous |
| 1874 | I/O. Requires :option:`filename` option to specify either block or |
| 1875 | character devices. This engine supports trim operations. |
| 1876 | The sg engine includes engine specific options. |
| 1877 | |
| 1878 | **null** |
| 1879 | Doesn't transfer any data, just pretends to. This is mainly used to |
| 1880 | exercise fio itself and for debugging/testing purposes. |
| 1881 | |
| 1882 | **net** |
| 1883 | Transfer over the network to given ``host:port``. Depending on the |
| 1884 | :option:`protocol` used, the :option:`hostname`, :option:`port`, |
| 1885 | :option:`listen` and :option:`filename` options are used to specify |
| 1886 | what sort of connection to make, while the :option:`protocol` option |
| 1887 | determines which protocol will be used. This engine defines engine |
| 1888 | specific options. |
| 1889 | |
| 1890 | **netsplice** |
| 1891 | Like **net**, but uses :manpage:`splice(2)` and |
| 1892 | :manpage:`vmsplice(2)` to map data and send/receive. |
| 1893 | This engine defines engine specific options. |
| 1894 | |
| 1895 | **cpuio** |
| 1896 | Doesn't transfer any data, but burns CPU cycles according to the |
| 1897 | :option:`cpuload` and :option:`cpuchunks` options. Setting |
| 1898 | :option:`cpuload`\=85 will cause that job to do nothing but burn 85% |
| 1899 | of the CPU. In case of SMP machines, use :option:`numjobs`\=<nr_of_cpu> |
| 1900 | to get desired CPU usage, as the cpuload only loads a |
| 1901 | single CPU at the desired rate. A job never finishes unless there is |
| 1902 | at least one non-cpuio job. |
| 1903 | |
| 1904 | **guasi** |
| 1905 | The GUASI I/O engine is the Generic Userspace Asynchronous Syscall |
| 1906 | Interface approach to async I/O. See |
| 1907 | |
| 1908 | http://www.xmailserver.org/guasi-lib.html |
| 1909 | |
| 1910 | for more info on GUASI. |
| 1911 | |
| 1912 | **rdma** |
| 1913 | The RDMA I/O engine supports both RDMA memory semantics |
| 1914 | (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the |
| 1915 | InfiniBand, RoCE and iWARP protocols. This engine defines engine |
| 1916 | specific options. |
| 1917 | |
| 1918 | **falloc** |
| 1919 | I/O engine that does regular fallocate to simulate data transfer as |
| 1920 | fio ioengine. |
| 1921 | |
| 1922 | DDIR_READ |
| 1923 | does fallocate(,mode = FALLOC_FL_KEEP_SIZE,). |
| 1924 | |
| 1925 | DDIR_WRITE |
| 1926 | does fallocate(,mode = 0). |
| 1927 | |
| 1928 | DDIR_TRIM |
| 1929 | does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE). |
| 1930 | |
| 1931 | **ftruncate** |
| 1932 | I/O engine that sends :manpage:`ftruncate(2)` operations in response |
| 1933 | to write (DDIR_WRITE) events. Each ftruncate issued sets the file's |
| 1934 | size to the current block offset. :option:`blocksize` is ignored. |
| 1935 | |
| 1936 | **e4defrag** |
| 1937 | I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate |
| 1938 | defragment activity in request to DDIR_WRITE event. |
| 1939 | |
| 1940 | **rados** |
| 1941 | I/O engine supporting direct access to Ceph Reliable Autonomic |
| 1942 | Distributed Object Store (RADOS) via librados. This ioengine |
| 1943 | defines engine specific options. |
| 1944 | |
| 1945 | **rbd** |
| 1946 | I/O engine supporting direct access to Ceph Rados Block Devices |
| 1947 | (RBD) via librbd without the need to use the kernel rbd driver. This |
| 1948 | ioengine defines engine specific options. |
| 1949 | |
| 1950 | **http** |
| 1951 | I/O engine supporting GET/PUT requests over HTTP(S) with libcurl to |
| 1952 | a WebDAV or S3 endpoint. This ioengine defines engine specific options. |
| 1953 | |
| 1954 | This engine only supports direct IO of iodepth=1; you need to scale this |
| 1955 | via numjobs. blocksize defines the size of the objects to be created. |
| 1956 | |
| 1957 | TRIM is translated to object deletion. |
| 1958 | |
| 1959 | **gfapi** |
| 1960 | Using GlusterFS libgfapi sync interface to direct access to |
| 1961 | GlusterFS volumes without having to go through FUSE. This ioengine |
| 1962 | defines engine specific options. |
| 1963 | |
| 1964 | **gfapi_async** |
| 1965 | Using GlusterFS libgfapi async interface to direct access to |
| 1966 | GlusterFS volumes without having to go through FUSE. This ioengine |
| 1967 | defines engine specific options. |
| 1968 | |
| 1969 | **libhdfs** |
| 1970 | Read and write through Hadoop (HDFS). The :option:`filename` option |
| 1971 | is used to specify host,port of the hdfs name-node to connect. This |
| 1972 | engine interprets offsets a little differently. In HDFS, files once |
| 1973 | created cannot be modified so random writes are not possible. To |
| 1974 | imitate this the libhdfs engine expects a bunch of small files to be |
| 1975 | created over HDFS and will randomly pick a file from them |
| 1976 | based on the offset generated by fio backend (see the example |
| 1977 | job file to create such files, use ``rw=write`` option). Please |
| 1978 | note, it may be necessary to set environment variables to work |
| 1979 | with HDFS/libhdfs properly. Each job uses its own connection to |
| 1980 | HDFS. |
| 1981 | |
| 1982 | **mtd** |
| 1983 | Read, write and erase an MTD character device (e.g., |
| 1984 | :file:`/dev/mtd0`). Discards are treated as erases. Depending on the |
| 1985 | underlying device type, the I/O may have to go in a certain pattern, |
| 1986 | e.g., on NAND, writing sequentially to erase blocks and discarding |
| 1987 | before overwriting. The `trimwrite` mode works well for this |
| 1988 | constraint. |
| 1989 | |
| 1990 | **pmemblk** |
| 1991 | Read and write using filesystem DAX to a file on a filesystem |
| 1992 | mounted with DAX on a persistent memory device through the PMDK |
| 1993 | libpmemblk library. |
| 1994 | |
| 1995 | **dev-dax** |
| 1996 | Read and write using device DAX to a persistent memory device (e.g., |
| 1997 | /dev/dax0.0) through the PMDK libpmem library. |
| 1998 | |
| 1999 | **external** |
| 2000 | Prefix to specify loading an external I/O engine object file. Append |
| 2001 | the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load |
| 2002 | ioengine :file:`foo.o` in :file:`/tmp`. The path can be either |
| 2003 | absolute or relative. See :file:`engines/skeleton_external.c` for |
| 2004 | details of writing an external I/O engine. |
| 2005 | |
| 2006 | **filecreate** |
| 2007 | Simply create the files and do no I/O to them. You still need to |
| 2008 | set `filesize` so that all the accounting still occurs, but no |
| 2009 | actual I/O will be done other than creating the file. |
| 2010 | |
| 2011 | **filestat** |
| 2012 | Simply do stat() and do no I/O to the file. You need to set 'filesize' |
| 2013 | and 'nrfiles', so that files will be created. |
| 2014 | This engine is to measure file lookup and meta data access. |
| 2015 | |
| 2016 | **libpmem** |
| 2017 | Read and write using mmap I/O to a file on a filesystem |
| 2018 | mounted with DAX on a persistent memory device through the PMDK |
| 2019 | libpmem library. |
| 2020 | |
| 2021 | **ime_psync** |
| 2022 | Synchronous read and write using DDN's Infinite Memory Engine (IME). |
| 2023 | This engine is very basic and issues calls to IME whenever an IO is |
| 2024 | queued. |
| 2025 | |
| 2026 | **ime_psyncv** |
| 2027 | Synchronous read and write using DDN's Infinite Memory Engine (IME). |
| 2028 | This engine uses iovecs and will try to stack as much IOs as possible |
| 2029 | (if the IOs are "contiguous" and the IO depth is not exceeded) |
| 2030 | before issuing a call to IME. |
| 2031 | |
| 2032 | **ime_aio** |
| 2033 | Asynchronous read and write using DDN's Infinite Memory Engine (IME). |
| 2034 | This engine will try to stack as much IOs as possible by creating |
| 2035 | requests for IME. FIO will then decide when to commit these requests. |
| 2036 | **libiscsi** |
| 2037 | Read and write iscsi lun with libiscsi. |
| 2038 | **nbd** |
| 2039 | Read and write a Network Block Device (NBD). |
| 2040 | |
| 2041 | I/O engine specific parameters |
| 2042 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| 2043 | |
| 2044 | In addition, there are some parameters which are only valid when a specific |
| 2045 | :option:`ioengine` is in use. These are used identically to normal parameters, |
| 2046 | with the caveat that when used on the command line, they must come after the |
| 2047 | :option:`ioengine` that defines them is selected. |
| 2048 | |
| 2049 | .. option:: cmdprio_percentage=int : [io_uring] [libaio] |
| 2050 | |
| 2051 | Set the percentage of I/O that will be issued with higher priority by setting |
| 2052 | the priority bit. Non-read I/O is likely unaffected by ``cmdprio_percentage``. |
| 2053 | This option cannot be used with the `prio` or `prioclass` options. For this |
| 2054 | option to set the priority bit properly, NCQ priority must be supported and |
| 2055 | enabled and :option:`direct`\=1 option must be used. fio must also be run as |
| 2056 | the root user. |
| 2057 | |
| 2058 | .. option:: fixedbufs : [io_uring] |
| 2059 | |
| 2060 | If fio is asked to do direct IO, then Linux will map pages for each |
| 2061 | IO call, and release them when IO is done. If this option is set, the |
| 2062 | pages are pre-mapped before IO is started. This eliminates the need to |
| 2063 | map and release for each IO. This is more efficient, and reduces the |
| 2064 | IO latency as well. |
| 2065 | |
| 2066 | .. option:: hipri : [io_uring] |
| 2067 | |
| 2068 | If this option is set, fio will attempt to use polled IO completions. |
| 2069 | Normal IO completions generate interrupts to signal the completion of |
| 2070 | IO, polled completions do not. Hence they are require active reaping |
| 2071 | by the application. The benefits are more efficient IO for high IOPS |
| 2072 | scenarios, and lower latencies for low queue depth IO. |
| 2073 | |
| 2074 | .. option:: registerfiles : [io_uring] |
| 2075 | |
| 2076 | With this option, fio registers the set of files being used with the |
| 2077 | kernel. This avoids the overhead of managing file counts in the kernel, |
| 2078 | making the submission and completion part more lightweight. Required |
| 2079 | for the below :option:`sqthread_poll` option. |
| 2080 | |
| 2081 | .. option:: sqthread_poll : [io_uring] |
| 2082 | |
| 2083 | Normally fio will submit IO by issuing a system call to notify the |
| 2084 | kernel of available items in the SQ ring. If this option is set, the |
| 2085 | act of submitting IO will be done by a polling thread in the kernel. |
| 2086 | This frees up cycles for fio, at the cost of using more CPU in the |
| 2087 | system. |
| 2088 | |
| 2089 | .. option:: sqthread_poll_cpu : [io_uring] |
| 2090 | |
| 2091 | When :option:`sqthread_poll` is set, this option provides a way to |
| 2092 | define which CPU should be used for the polling thread. |
| 2093 | |
| 2094 | .. option:: userspace_reap : [libaio] |
| 2095 | |
| 2096 | Normally, with the libaio engine in use, fio will use the |
| 2097 | :manpage:`io_getevents(2)` system call to reap newly returned events. With |
| 2098 | this flag turned on, the AIO ring will be read directly from user-space to |
| 2099 | reap events. The reaping mode is only enabled when polling for a minimum of |
| 2100 | 0 events (e.g. when :option:`iodepth_batch_complete` `=0`). |
| 2101 | |
| 2102 | .. option:: hipri : [pvsync2] |
| 2103 | |
| 2104 | Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority |
| 2105 | than normal. |
| 2106 | |
| 2107 | .. option:: hipri_percentage : [pvsync2] |
| 2108 | |
| 2109 | When hipri is set this determines the probability of a pvsync2 I/O being high |
| 2110 | priority. The default is 100%. |
| 2111 | |
| 2112 | .. option:: nowait : [pvsync2] [libaio] [io_uring] |
| 2113 | |
| 2114 | By default if a request cannot be executed immediately (e.g. resource starvation, |
| 2115 | waiting on locks) it is queued and the initiating process will be blocked until |
| 2116 | the required resource becomes free. |
| 2117 | |
| 2118 | This option sets the RWF_NOWAIT flag (supported from the 4.14 Linux kernel) and |
| 2119 | the call will return instantly with EAGAIN or a partial result rather than waiting. |
| 2120 | |
| 2121 | It is useful to also use ignore_error=EAGAIN when using this option. |
| 2122 | |
| 2123 | Note: glibc 2.27, 2.28 have a bug in syscall wrappers preadv2, pwritev2. |
| 2124 | They return EOPNOTSUP instead of EAGAIN. |
| 2125 | |
| 2126 | For cached I/O, using this option usually means a request operates only with |
| 2127 | cached data. Currently the RWF_NOWAIT flag does not supported for cached write. |
| 2128 | |
| 2129 | For direct I/O, requests will only succeed if cache invalidation isn't required, |
| 2130 | file blocks are fully allocated and the disk request could be issued immediately. |
| 2131 | |
| 2132 | .. option:: cpuload=int : [cpuio] |
| 2133 | |
| 2134 | Attempt to use the specified percentage of CPU cycles. This is a mandatory |
| 2135 | option when using cpuio I/O engine. |
| 2136 | |
| 2137 | .. option:: cpuchunks=int : [cpuio] |
| 2138 | |
| 2139 | Split the load into cycles of the given time. In microseconds. |
| 2140 | |
| 2141 | .. option:: exit_on_io_done=bool : [cpuio] |
| 2142 | |
| 2143 | Detect when I/O threads are done, then exit. |
| 2144 | |
| 2145 | .. option:: namenode=str : [libhdfs] |
| 2146 | |
| 2147 | The hostname or IP address of a HDFS cluster namenode to contact. |
| 2148 | |
| 2149 | .. option:: port=int |
| 2150 | |
| 2151 | [libhdfs] |
| 2152 | |
| 2153 | The listening port of the HFDS cluster namenode. |
| 2154 | |
| 2155 | [netsplice], [net] |
| 2156 | |
| 2157 | The TCP or UDP port to bind to or connect to. If this is used with |
| 2158 | :option:`numjobs` to spawn multiple instances of the same job type, then |
| 2159 | this will be the starting port number since fio will use a range of |
| 2160 | ports. |
| 2161 | |
| 2162 | [rdma] |
| 2163 | |
| 2164 | The port to use for RDMA-CM communication. This should be the same value |
| 2165 | on the client and the server side. |
| 2166 | |
| 2167 | .. option:: hostname=str : [netsplice] [net] [rdma] |
| 2168 | |
| 2169 | The hostname or IP address to use for TCP, UDP or RDMA-CM based I/O. If the job |
| 2170 | is a TCP listener or UDP reader, the hostname is not used and must be omitted |
| 2171 | unless it is a valid UDP multicast address. |
| 2172 | |
| 2173 | .. option:: interface=str : [netsplice] [net] |
| 2174 | |
| 2175 | The IP address of the network interface used to send or receive UDP |
| 2176 | multicast. |
| 2177 | |
| 2178 | .. option:: ttl=int : [netsplice] [net] |
| 2179 | |
| 2180 | Time-to-live value for outgoing UDP multicast packets. Default: 1. |
| 2181 | |
| 2182 | .. option:: nodelay=bool : [netsplice] [net] |
| 2183 | |
| 2184 | Set TCP_NODELAY on TCP connections. |
| 2185 | |
| 2186 | .. option:: protocol=str, proto=str : [netsplice] [net] |
| 2187 | |
| 2188 | The network protocol to use. Accepted values are: |
| 2189 | |
| 2190 | **tcp** |
| 2191 | Transmission control protocol. |
| 2192 | **tcpv6** |
| 2193 | Transmission control protocol V6. |
| 2194 | **udp** |
| 2195 | User datagram protocol. |
| 2196 | **udpv6** |
| 2197 | User datagram protocol V6. |
| 2198 | **unix** |
| 2199 | UNIX domain socket. |
| 2200 | |
| 2201 | When the protocol is TCP or UDP, the port must also be given, as well as the |
| 2202 | hostname if the job is a TCP listener or UDP reader. For unix sockets, the |
| 2203 | normal :option:`filename` option should be used and the port is invalid. |
| 2204 | |
| 2205 | .. option:: listen : [netsplice] [net] |
| 2206 | |
| 2207 | For TCP network connections, tell fio to listen for incoming connections |
| 2208 | rather than initiating an outgoing connection. The :option:`hostname` must |
| 2209 | be omitted if this option is used. |
| 2210 | |
| 2211 | .. option:: pingpong : [netsplice] [net] |
| 2212 | |
| 2213 | Normally a network writer will just continue writing data, and a network |
| 2214 | reader will just consume packages. If ``pingpong=1`` is set, a writer will |
| 2215 | send its normal payload to the reader, then wait for the reader to send the |
| 2216 | same payload back. This allows fio to measure network latencies. The |
| 2217 | submission and completion latencies then measure local time spent sending or |
| 2218 | receiving, and the completion latency measures how long it took for the |
| 2219 | other end to receive and send back. For UDP multicast traffic |
| 2220 | ``pingpong=1`` should only be set for a single reader when multiple readers |
| 2221 | are listening to the same address. |
| 2222 | |
| 2223 | .. option:: window_size : [netsplice] [net] |
| 2224 | |
| 2225 | Set the desired socket buffer size for the connection. |
| 2226 | |
| 2227 | .. option:: mss : [netsplice] [net] |
| 2228 | |
| 2229 | Set the TCP maximum segment size (TCP_MAXSEG). |
| 2230 | |
| 2231 | .. option:: donorname=str : [e4defrag] |
| 2232 | |
| 2233 | File will be used as a block donor (swap extents between files). |
| 2234 | |
| 2235 | .. option:: inplace=int : [e4defrag] |
| 2236 | |
| 2237 | Configure donor file blocks allocation strategy: |
| 2238 | |
| 2239 | **0** |
| 2240 | Default. Preallocate donor's file on init. |
| 2241 | **1** |
| 2242 | Allocate space immediately inside defragment event, and free right |
| 2243 | after event. |
| 2244 | |
| 2245 | .. option:: clustername=str : [rbd,rados] |
| 2246 | |
| 2247 | Specifies the name of the Ceph cluster. |
| 2248 | |
| 2249 | .. option:: rbdname=str : [rbd] |
| 2250 | |
| 2251 | Specifies the name of the RBD. |
| 2252 | |
| 2253 | .. option:: pool=str : [rbd,rados] |
| 2254 | |
| 2255 | Specifies the name of the Ceph pool containing RBD or RADOS data. |
| 2256 | |
| 2257 | .. option:: clientname=str : [rbd,rados] |
| 2258 | |
| 2259 | Specifies the username (without the 'client.' prefix) used to access the |
| 2260 | Ceph cluster. If the *clustername* is specified, the *clientname* shall be |
| 2261 | the full *type.id* string. If no type. prefix is given, fio will add |
| 2262 | 'client.' by default. |
| 2263 | |
| 2264 | .. option:: busy_poll=bool : [rbd,rados] |
| 2265 | |
| 2266 | Poll store instead of waiting for completion. Usually this provides better |
| 2267 | throughput at cost of higher(up to 100%) CPU utilization. |
| 2268 | |
| 2269 | .. option:: skip_bad=bool : [mtd] |
| 2270 | |
| 2271 | Skip operations against known bad blocks. |
| 2272 | |
| 2273 | .. option:: hdfsdirectory : [libhdfs] |
| 2274 | |
| 2275 | libhdfs will create chunk in this HDFS directory. |
| 2276 | |
| 2277 | .. option:: chunk_size : [libhdfs] |
| 2278 | |
| 2279 | The size of the chunk to use for each file. |
| 2280 | |
| 2281 | .. option:: verb=str : [rdma] |
| 2282 | |
| 2283 | The RDMA verb to use on this side of the RDMA ioengine connection. Valid |
| 2284 | values are write, read, send and recv. These correspond to the equivalent |
| 2285 | RDMA verbs (e.g. write = rdma_write etc.). Note that this only needs to be |
| 2286 | specified on the client side of the connection. See the examples folder. |
| 2287 | |
| 2288 | .. option:: bindname=str : [rdma] |
| 2289 | |
| 2290 | The name to use to bind the local RDMA-CM connection to a local RDMA device. |
| 2291 | This could be a hostname or an IPv4 or IPv6 address. On the server side this |
| 2292 | will be passed into the rdma_bind_addr() function and on the client site it |
| 2293 | will be used in the rdma_resolve_add() function. This can be useful when |
| 2294 | multiple paths exist between the client and the server or in certain loopback |
| 2295 | configurations. |
| 2296 | |
| 2297 | .. option:: stat_type=str : [filestat] |
| 2298 | |
| 2299 | Specify stat system call type to measure lookup/getattr performance. |
| 2300 | Default is **stat** for :manpage:`stat(2)`. |
| 2301 | |
| 2302 | .. option:: readfua=bool : [sg] |
| 2303 | |
| 2304 | With readfua option set to 1, read operations include |
| 2305 | the force unit access (fua) flag. Default is 0. |
| 2306 | |
| 2307 | .. option:: writefua=bool : [sg] |
| 2308 | |
| 2309 | With writefua option set to 1, write operations include |
| 2310 | the force unit access (fua) flag. Default is 0. |
| 2311 | |
| 2312 | .. option:: sg_write_mode=str : [sg] |
| 2313 | |
| 2314 | Specify the type of write commands to issue. This option can take three values: |
| 2315 | |
| 2316 | **write** |
| 2317 | This is the default where write opcodes are issued as usual. |
| 2318 | **verify** |
| 2319 | Issue WRITE AND VERIFY commands. The BYTCHK bit is set to 0. This |
| 2320 | directs the device to carry out a medium verification with no data |
| 2321 | comparison. The writefua option is ignored with this selection. |
| 2322 | **same** |
| 2323 | Issue WRITE SAME commands. This transfers a single block to the device |
| 2324 | and writes this same block of data to a contiguous sequence of LBAs |
| 2325 | beginning at the specified offset. fio's block size parameter specifies |
| 2326 | the amount of data written with each command. However, the amount of data |
| 2327 | actually transferred to the device is equal to the device's block |
| 2328 | (sector) size. For a device with 512 byte sectors, blocksize=8k will |
| 2329 | write 16 sectors with each command. fio will still generate 8k of data |
| 2330 | for each command but only the first 512 bytes will be used and |
| 2331 | transferred to the device. The writefua option is ignored with this |
| 2332 | selection. |
| 2333 | |
| 2334 | .. option:: http_host=str : [http] |
| 2335 | |
| 2336 | Hostname to connect to. For S3, this could be the bucket hostname. |
| 2337 | Default is **localhost** |
| 2338 | |
| 2339 | .. option:: http_user=str : [http] |
| 2340 | |
| 2341 | Username for HTTP authentication. |
| 2342 | |
| 2343 | .. option:: http_pass=str : [http] |
| 2344 | |
| 2345 | Password for HTTP authentication. |
| 2346 | |
| 2347 | .. option:: https=str : [http] |
| 2348 | |
| 2349 | Enable HTTPS instead of http. *on* enables HTTPS; *insecure* |
| 2350 | will enable HTTPS, but disable SSL peer verification (use with |
| 2351 | caution!). Default is **off** |
| 2352 | |
| 2353 | .. option:: http_mode=str : [http] |
| 2354 | |
| 2355 | Which HTTP access mode to use: *webdav*, *swift*, or *s3*. |
| 2356 | Default is **webdav** |
| 2357 | |
| 2358 | .. option:: http_s3_region=str : [http] |
| 2359 | |
| 2360 | The S3 region/zone string. |
| 2361 | Default is **us-east-1** |
| 2362 | |
| 2363 | .. option:: http_s3_key=str : [http] |
| 2364 | |
| 2365 | The S3 secret key. |
| 2366 | |
| 2367 | .. option:: http_s3_keyid=str : [http] |
| 2368 | |
| 2369 | The S3 key/access id. |
| 2370 | |
| 2371 | .. option:: http_swift_auth_token=str : [http] |
| 2372 | |
| 2373 | The Swift auth token. See the example configuration file on how |
| 2374 | to retrieve this. |
| 2375 | |
| 2376 | .. option:: http_verbose=int : [http] |
| 2377 | |
| 2378 | Enable verbose requests from libcurl. Useful for debugging. 1 |
| 2379 | turns on verbose logging from libcurl, 2 additionally enables |
| 2380 | HTTP IO tracing. Default is **0** |
| 2381 | |
| 2382 | .. option:: uri=str : [nbd] |
| 2383 | |
| 2384 | Specify the NBD URI of the server to test. The string |
| 2385 | is a standard NBD URI |
| 2386 | (see https://github.com/NetworkBlockDevice/nbd/tree/master/doc). |
| 2387 | Example URIs: nbd://localhost:10809 |
| 2388 | nbd+unix:///?socket=/tmp/socket |
| 2389 | nbds://tlshost/exportname |
| 2390 | |
| 2391 | I/O depth |
| 2392 | ~~~~~~~~~ |
| 2393 | |
| 2394 | .. option:: iodepth=int |
| 2395 | |
| 2396 | Number of I/O units to keep in flight against the file. Note that |
| 2397 | increasing *iodepth* beyond 1 will not affect synchronous ioengines (except |
| 2398 | for small degrees when :option:`verify_async` is in use). Even async |
| 2399 | engines may impose OS restrictions causing the desired depth not to be |
| 2400 | achieved. This may happen on Linux when using libaio and not setting |
| 2401 | :option:`direct`\=1, since buffered I/O is not async on that OS. Keep an |
| 2402 | eye on the I/O depth distribution in the fio output to verify that the |
| 2403 | achieved depth is as expected. Default: 1. |
| 2404 | |
| 2405 | .. option:: iodepth_batch_submit=int, iodepth_batch=int |
| 2406 | |
| 2407 | This defines how many pieces of I/O to submit at once. It defaults to 1 |
| 2408 | which means that we submit each I/O as soon as it is available, but can be |
| 2409 | raised to submit bigger batches of I/O at the time. If it is set to 0 the |
| 2410 | :option:`iodepth` value will be used. |
| 2411 | |
| 2412 | .. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int |
| 2413 | |
| 2414 | This defines how many pieces of I/O to retrieve at once. It defaults to 1 |
| 2415 | which means that we'll ask for a minimum of 1 I/O in the retrieval process |
| 2416 | from the kernel. The I/O retrieval will go on until we hit the limit set by |
| 2417 | :option:`iodepth_low`. If this variable is set to 0, then fio will always |
| 2418 | check for completed events before queuing more I/O. This helps reduce I/O |
| 2419 | latency, at the cost of more retrieval system calls. |
| 2420 | |
| 2421 | .. option:: iodepth_batch_complete_max=int |
| 2422 | |
| 2423 | This defines maximum pieces of I/O to retrieve at once. This variable should |
| 2424 | be used along with :option:`iodepth_batch_complete_min`\=int variable, |
| 2425 | specifying the range of min and max amount of I/O which should be |
| 2426 | retrieved. By default it is equal to the :option:`iodepth_batch_complete_min` |
| 2427 | value. |
| 2428 | |
| 2429 | Example #1:: |
| 2430 | |
| 2431 | iodepth_batch_complete_min=1 |
| 2432 | iodepth_batch_complete_max=<iodepth> |
| 2433 | |
| 2434 | which means that we will retrieve at least 1 I/O and up to the whole |
| 2435 | submitted queue depth. If none of I/O has been completed yet, we will wait. |
| 2436 | |
| 2437 | Example #2:: |
| 2438 | |
| 2439 | iodepth_batch_complete_min=0 |
| 2440 | iodepth_batch_complete_max=<iodepth> |
| 2441 | |
| 2442 | which means that we can retrieve up to the whole submitted queue depth, but |
| 2443 | if none of I/O has been completed yet, we will NOT wait and immediately exit |
| 2444 | the system call. In this example we simply do polling. |
| 2445 | |
| 2446 | .. option:: iodepth_low=int |
| 2447 | |
| 2448 | The low water mark indicating when to start filling the queue |
| 2449 | again. Defaults to the same as :option:`iodepth`, meaning that fio will |
| 2450 | attempt to keep the queue full at all times. If :option:`iodepth` is set to |
| 2451 | e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of |
| 2452 | 16 requests, it will let the depth drain down to 4 before starting to fill |
| 2453 | it again. |
| 2454 | |
| 2455 | .. option:: serialize_overlap=bool |
| 2456 | |
| 2457 | Serialize in-flight I/Os that might otherwise cause or suffer from data races. |
| 2458 | When two or more I/Os are submitted simultaneously, there is no guarantee that |
| 2459 | the I/Os will be processed or completed in the submitted order. Further, if |
| 2460 | two or more of those I/Os are writes, any overlapping region between them can |
| 2461 | become indeterminate/undefined on certain storage. These issues can cause |
| 2462 | verification to fail erratically when at least one of the racing I/Os is |
| 2463 | changing data and the overlapping region has a non-zero size. Setting |
| 2464 | ``serialize_overlap`` tells fio to avoid provoking this behavior by explicitly |
| 2465 | serializing in-flight I/Os that have a non-zero overlap. Note that setting |
| 2466 | this option can reduce both performance and the :option:`iodepth` achieved. |
| 2467 | |
| 2468 | This option only applies to I/Os issued for a single job except when it is |
| 2469 | enabled along with :option:`io_submit_mode`\=offload. In offload mode, fio |
| 2470 | will check for overlap among all I/Os submitted by offload jobs with :option:`serialize_overlap` |
| 2471 | enabled. |
| 2472 | |
| 2473 | Default: false. |
| 2474 | |
| 2475 | .. option:: io_submit_mode=str |
| 2476 | |
| 2477 | This option controls how fio submits the I/O to the I/O engine. The default |
| 2478 | is `inline`, which means that the fio job threads submit and reap I/O |
| 2479 | directly. If set to `offload`, the job threads will offload I/O submission |
| 2480 | to a dedicated pool of I/O threads. This requires some coordination and thus |
| 2481 | has a bit of extra overhead, especially for lower queue depth I/O where it |
| 2482 | can increase latencies. The benefit is that fio can manage submission rates |
| 2483 | independently of the device completion rates. This avoids skewed latency |
| 2484 | reporting if I/O gets backed up on the device side (the coordinated omission |
| 2485 | problem). |
| 2486 | |
| 2487 | |
| 2488 | I/O rate |
| 2489 | ~~~~~~~~ |
| 2490 | |
| 2491 | .. option:: thinktime=time |
| 2492 | |
| 2493 | Stall the job for the specified period of time after an I/O has completed before issuing the |
| 2494 | next. May be used to simulate processing being done by an application. |
| 2495 | When the unit is omitted, the value is interpreted in microseconds. See |
| 2496 | :option:`thinktime_blocks` and :option:`thinktime_spin`. |
| 2497 | |
| 2498 | .. option:: thinktime_spin=time |
| 2499 | |
| 2500 | Only valid if :option:`thinktime` is set - pretend to spend CPU time doing |
| 2501 | something with the data received, before falling back to sleeping for the |
| 2502 | rest of the period specified by :option:`thinktime`. When the unit is |
| 2503 | omitted, the value is interpreted in microseconds. |
| 2504 | |
| 2505 | .. option:: thinktime_blocks=int |
| 2506 | |
| 2507 | Only valid if :option:`thinktime` is set - control how many blocks to issue, |
| 2508 | before waiting :option:`thinktime` usecs. If not set, defaults to 1 which will make |
| 2509 | fio wait :option:`thinktime` usecs after every block. This effectively makes any |
| 2510 | queue depth setting redundant, since no more than 1 I/O will be queued |
| 2511 | before we have to complete it and do our :option:`thinktime`. In other words, this |
| 2512 | setting effectively caps the queue depth if the latter is larger. |
| 2513 | |
| 2514 | .. option:: rate=int[,int][,int] |
| 2515 | |
| 2516 | Cap the bandwidth used by this job. The number is in bytes/sec, the normal |
| 2517 | suffix rules apply. Comma-separated values may be specified for reads, |
| 2518 | writes, and trims as described in :option:`blocksize`. |
| 2519 | |
| 2520 | For example, using `rate=1m,500k` would limit reads to 1MiB/sec and writes to |
| 2521 | 500KiB/sec. Capping only reads or writes can be done with `rate=,500k` or |
| 2522 | `rate=500k,` where the former will only limit writes (to 500KiB/sec) and the |
| 2523 | latter will only limit reads. |
| 2524 | |
| 2525 | .. option:: rate_min=int[,int][,int] |
| 2526 | |
| 2527 | Tell fio to do whatever it can to maintain at least this bandwidth. Failing |
| 2528 | to meet this requirement will cause the job to exit. Comma-separated values |
| 2529 | may be specified for reads, writes, and trims as described in |
| 2530 | :option:`blocksize`. |
| 2531 | |
| 2532 | .. option:: rate_iops=int[,int][,int] |
| 2533 | |
| 2534 | Cap the bandwidth to this number of IOPS. Basically the same as |
| 2535 | :option:`rate`, just specified independently of bandwidth. If the job is |
| 2536 | given a block size range instead of a fixed value, the smallest block size |
| 2537 | is used as the metric. Comma-separated values may be specified for reads, |
| 2538 | writes, and trims as described in :option:`blocksize`. |
| 2539 | |
| 2540 | .. option:: rate_iops_min=int[,int][,int] |
| 2541 | |
| 2542 | If fio doesn't meet this rate of I/O, it will cause the job to exit. |
| 2543 | Comma-separated values may be specified for reads, writes, and trims as |
| 2544 | described in :option:`blocksize`. |
| 2545 | |
| 2546 | .. option:: rate_process=str |
| 2547 | |
| 2548 | This option controls how fio manages rated I/O submissions. The default is |
| 2549 | `linear`, which submits I/O in a linear fashion with fixed delays between |
| 2550 | I/Os that gets adjusted based on I/O completion rates. If this is set to |
| 2551 | `poisson`, fio will submit I/O based on a more real world random request |
| 2552 | flow, known as the Poisson process |
| 2553 | (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be |
| 2554 | 10^6 / IOPS for the given workload. |
| 2555 | |
| 2556 | .. option:: rate_ignore_thinktime=bool |
| 2557 | |
| 2558 | By default, fio will attempt to catch up to the specified rate setting, |
| 2559 | if any kind of thinktime setting was used. If this option is set, then |
| 2560 | fio will ignore the thinktime and continue doing IO at the specified |
| 2561 | rate, instead of entering a catch-up mode after thinktime is done. |
| 2562 | |
| 2563 | |
| 2564 | I/O latency |
| 2565 | ~~~~~~~~~~~ |
| 2566 | |
| 2567 | .. option:: latency_target=time |
| 2568 | |
| 2569 | If set, fio will attempt to find the max performance point that the given |
| 2570 | workload will run at while maintaining a latency below this target. When |
| 2571 | the unit is omitted, the value is interpreted in microseconds. See |
| 2572 | :option:`latency_window` and :option:`latency_percentile`. |
| 2573 | |
| 2574 | .. option:: latency_window=time |
| 2575 | |
| 2576 | Used with :option:`latency_target` to specify the sample window that the job |
| 2577 | is run at varying queue depths to test the performance. When the unit is |
| 2578 | omitted, the value is interpreted in microseconds. |
| 2579 | |
| 2580 | .. option:: latency_percentile=float |
| 2581 | |
| 2582 | The percentage of I/Os that must fall within the criteria specified by |
| 2583 | :option:`latency_target` and :option:`latency_window`. If not set, this |
| 2584 | defaults to 100.0, meaning that all I/Os must be equal or below to the value |
| 2585 | set by :option:`latency_target`. |
| 2586 | |
| 2587 | .. option:: latency_run=bool |
| 2588 | |
| 2589 | Used with :option:`latency_target`. If false (default), fio will find |
| 2590 | the highest queue depth that meets :option:`latency_target` and exit. If |
| 2591 | true, fio will continue running and try to meet :option:`latency_target` |
| 2592 | by adjusting queue depth. |
| 2593 | |
| 2594 | .. option:: max_latency=time |
| 2595 | |
| 2596 | If set, fio will exit the job with an ETIMEDOUT error if it exceeds this |
| 2597 | maximum latency. When the unit is omitted, the value is interpreted in |
| 2598 | microseconds. |
| 2599 | |
| 2600 | .. option:: rate_cycle=int |
| 2601 | |
| 2602 | Average bandwidth for :option:`rate` and :option:`rate_min` over this number |
| 2603 | of milliseconds. Defaults to 1000. |
| 2604 | |
| 2605 | |
| 2606 | I/O replay |
| 2607 | ~~~~~~~~~~ |
| 2608 | |
| 2609 | .. option:: write_iolog=str |
| 2610 | |
| 2611 | Write the issued I/O patterns to the specified file. See |
| 2612 | :option:`read_iolog`. Specify a separate file for each job, otherwise the |
| 2613 | iologs will be interspersed and the file may be corrupt. |
| 2614 | |
| 2615 | .. option:: read_iolog=str |
| 2616 | |
| 2617 | Open an iolog with the specified filename and replay the I/O patterns it |
| 2618 | contains. This can be used to store a workload and replay it sometime |
| 2619 | later. The iolog given may also be a blktrace binary file, which allows fio |
| 2620 | to replay a workload captured by :command:`blktrace`. See |
| 2621 | :manpage:`blktrace(8)` for how to capture such logging data. For blktrace |
| 2622 | replay, the file needs to be turned into a blkparse binary data file first |
| 2623 | (``blkparse <device> -o /dev/null -d file_for_fio.bin``). |
| 2624 | You can specify a number of files by separating the names with a ':' |
| 2625 | character. See the :option:`filename` option for information on how to |
| 2626 | escape ':' characters within the file names. These files will |
| 2627 | be sequentially assigned to job clones created by :option:`numjobs`. |
| 2628 | '-' is a reserved name, meaning read from stdin, notably if |
| 2629 | :option:`filename` is set to '-' which means stdin as well, then |
| 2630 | this flag can't be set to '-'. |
| 2631 | |
| 2632 | .. option:: read_iolog_chunked=bool |
| 2633 | |
| 2634 | Determines how iolog is read. If false(default) entire :option:`read_iolog` |
| 2635 | will be read at once. If selected true, input from iolog will be read |
| 2636 | gradually. Useful when iolog is very large, or it is generated. |
| 2637 | |
| 2638 | .. option:: merge_blktrace_file=str |
| 2639 | |
| 2640 | When specified, rather than replaying the logs passed to :option:`read_iolog`, |
| 2641 | the logs go through a merge phase which aggregates them into a single |
| 2642 | blktrace. The resulting file is then passed on as the :option:`read_iolog` |
| 2643 | parameter. The intention here is to make the order of events consistent. |
| 2644 | This limits the influence of the scheduler compared to replaying multiple |
| 2645 | blktraces via concurrent jobs. |
| 2646 | |
| 2647 | .. option:: merge_blktrace_scalars=float_list |
| 2648 | |
| 2649 | This is a percentage based option that is index paired with the list of |
| 2650 | files passed to :option:`read_iolog`. When merging is performed, scale |
| 2651 | the time of each event by the corresponding amount. For example, |
| 2652 | ``--merge_blktrace_scalars="50:100"`` runs the first trace in halftime |
| 2653 | and the second trace in realtime. This knob is separately tunable from |
| 2654 | :option:`replay_time_scale` which scales the trace during runtime and |
| 2655 | does not change the output of the merge unlike this option. |
| 2656 | |
| 2657 | .. option:: merge_blktrace_iters=float_list |
| 2658 | |
| 2659 | This is a whole number option that is index paired with the list of files |
| 2660 | passed to :option:`read_iolog`. When merging is performed, run each trace |
| 2661 | for the specified number of iterations. For example, |
| 2662 | ``--merge_blktrace_iters="2:1"`` runs the first trace for two iterations |
| 2663 | and the second trace for one iteration. |
| 2664 | |
| 2665 | .. option:: replay_no_stall=bool |
| 2666 | |
| 2667 | When replaying I/O with :option:`read_iolog` the default behavior is to |
| 2668 | attempt to respect the timestamps within the log and replay them with the |
| 2669 | appropriate delay between IOPS. By setting this variable fio will not |
| 2670 | respect the timestamps and attempt to replay them as fast as possible while |
| 2671 | still respecting ordering. The result is the same I/O pattern to a given |
| 2672 | device, but different timings. |
| 2673 | |
| 2674 | .. option:: replay_time_scale=int |
| 2675 | |
| 2676 | When replaying I/O with :option:`read_iolog`, fio will honor the |
| 2677 | original timing in the trace. With this option, it's possible to scale |
| 2678 | the time. It's a percentage option, if set to 50 it means run at 50% |
| 2679 | the original IO rate in the trace. If set to 200, run at twice the |
| 2680 | original IO rate. Defaults to 100. |
| 2681 | |
| 2682 | .. option:: replay_redirect=str |
| 2683 | |
| 2684 | While replaying I/O patterns using :option:`read_iolog` the default behavior |
| 2685 | is to replay the IOPS onto the major/minor device that each IOP was recorded |
| 2686 | from. This is sometimes undesirable because on a different machine those |
| 2687 | major/minor numbers can map to a different device. Changing hardware on the |
| 2688 | same system can also result in a different major/minor mapping. |
| 2689 | ``replay_redirect`` causes all I/Os to be replayed onto the single specified |
| 2690 | device regardless of the device it was recorded |
| 2691 | from. i.e. :option:`replay_redirect`\= :file:`/dev/sdc` would cause all I/O |
| 2692 | in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means |
| 2693 | multiple devices will be replayed onto a single device, if the trace |
| 2694 | contains multiple devices. If you want multiple devices to be replayed |
| 2695 | concurrently to multiple redirected devices you must blkparse your trace |
| 2696 | into separate traces and replay them with independent fio invocations. |
| 2697 | Unfortunately this also breaks the strict time ordering between multiple |
| 2698 | device accesses. |
| 2699 | |
| 2700 | .. option:: replay_align=int |
| 2701 | |
| 2702 | Force alignment of the byte offsets in a trace to this value. The value |
| 2703 | must be a power of 2. |
| 2704 | |
| 2705 | .. option:: replay_scale=int |
| 2706 | |
| 2707 | Scale byte offsets down by this factor when replaying traces. Should most |
| 2708 | likely use :option:`replay_align` as well. |
| 2709 | |
| 2710 | .. option:: replay_skip=str |
| 2711 | |
| 2712 | Sometimes it's useful to skip certain IO types in a replay trace. |
| 2713 | This could be, for instance, eliminating the writes in the trace. |
| 2714 | Or not replaying the trims/discards, if you are redirecting to |
| 2715 | a device that doesn't support them. This option takes a comma |
| 2716 | separated list of read, write, trim, sync. |
| 2717 | |
| 2718 | |
| 2719 | Threads, processes and job synchronization |
| 2720 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| 2721 | |
| 2722 | .. option:: thread |
| 2723 | |
| 2724 | Fio defaults to creating jobs by using fork, however if this option is |
| 2725 | given, fio will create jobs by using POSIX Threads' function |
| 2726 | :manpage:`pthread_create(3)` to create threads instead. |
| 2727 | |
| 2728 | .. option:: wait_for=str |
| 2729 | |
| 2730 | If set, the current job won't be started until all workers of the specified |
| 2731 | waitee job are done. |
| 2732 | |
| 2733 | ``wait_for`` operates on the job name basis, so there are a few |
| 2734 | limitations. First, the waitee must be defined prior to the waiter job |
| 2735 | (meaning no forward references). Second, if a job is being referenced as a |
| 2736 | waitee, it must have a unique name (no duplicate waitees). |
| 2737 | |
| 2738 | .. option:: nice=int |
| 2739 | |
| 2740 | Run the job with the given nice value. See man :manpage:`nice(2)`. |
| 2741 | |
| 2742 | On Windows, values less than -15 set the process class to "High"; -1 through |
| 2743 | -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle" |
| 2744 | priority class. |
| 2745 | |
| 2746 | .. option:: prio=int |
| 2747 | |
| 2748 | Set the I/O priority value of this job. Linux limits us to a positive value |
| 2749 | between 0 and 7, with 0 being the highest. See man |
| 2750 | :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating |
| 2751 | systems since meaning of priority may differ. For per-command priority |
| 2752 | setting, see I/O engine specific `cmdprio_percentage` and `hipri_percentage` |
| 2753 | options. |
| 2754 | |
| 2755 | .. option:: prioclass=int |
| 2756 | |
| 2757 | Set the I/O priority class. See man :manpage:`ionice(1)`. For per-command |
| 2758 | priority setting, see I/O engine specific `cmdprio_percentage` and |
| 2759 | `hipri_percentage` options. |
| 2760 | |
| 2761 | .. option:: cpus_allowed=str |
| 2762 | |
| 2763 | Controls the same options as :option:`cpumask`, but accepts a textual |
| 2764 | specification of the permitted CPUs instead and CPUs are indexed from 0. So |
| 2765 | to use CPUs 0 and 5 you would specify ``cpus_allowed=0,5``. This option also |
| 2766 | allows a range of CPUs to be specified -- say you wanted a binding to CPUs |
| 2767 | 0, 5, and 8 to 15, you would set ``cpus_allowed=0,5,8-15``. |
| 2768 | |
| 2769 | On Windows, when ``cpus_allowed`` is unset only CPUs from fio's current |
| 2770 | processor group will be used and affinity settings are inherited from the |
| 2771 | system. An fio build configured to target Windows 7 makes options that set |
| 2772 | CPUs processor group aware and values will set both the processor group |
| 2773 | and a CPU from within that group. For example, on a system where processor |
| 2774 | group 0 has 40 CPUs and processor group 1 has 32 CPUs, ``cpus_allowed`` |
| 2775 | values between 0 and 39 will bind CPUs from processor group 0 and |
| 2776 | ``cpus_allowed`` values between 40 and 71 will bind CPUs from processor |
| 2777 | group 1. When using ``cpus_allowed_policy=shared`` all CPUs specified by a |
| 2778 | single ``cpus_allowed`` option must be from the same processor group. For |
| 2779 | Windows fio builds not built for Windows 7, CPUs will only be selected from |
| 2780 | (and be relative to) whatever processor group fio happens to be running in |
| 2781 | and CPUs from other processor groups cannot be used. |
| 2782 | |
| 2783 | .. option:: cpus_allowed_policy=str |
| 2784 | |
| 2785 | Set the policy of how fio distributes the CPUs specified by |
| 2786 | :option:`cpus_allowed` or :option:`cpumask`. Two policies are supported: |
| 2787 | |
| 2788 | **shared** |
| 2789 | All jobs will share the CPU set specified. |
| 2790 | **split** |
| 2791 | Each job will get a unique CPU from the CPU set. |
| 2792 | |
| 2793 | **shared** is the default behavior, if the option isn't specified. If |
| 2794 | **split** is specified, then fio will assign one cpu per job. If not |
| 2795 | enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs |
| 2796 | in the set. |
| 2797 | |
| 2798 | .. option:: cpumask=int |
| 2799 | |
| 2800 | Set the CPU affinity of this job. The parameter given is a bit mask of |
| 2801 | allowed CPUs the job may run on. So if you want the allowed CPUs to be 1 |
| 2802 | and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man |
| 2803 | :manpage:`sched_setaffinity(2)`. This may not work on all supported |
| 2804 | operating systems or kernel versions. This option doesn't work well for a |
| 2805 | higher CPU count than what you can store in an integer mask, so it can only |
| 2806 | control cpus 1-32. For boxes with larger CPU counts, use |
| 2807 | :option:`cpus_allowed`. |
| 2808 | |
| 2809 | .. option:: numa_cpu_nodes=str |
| 2810 | |
| 2811 | Set this job running on specified NUMA nodes' CPUs. The arguments allow |
| 2812 | comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable |
| 2813 | NUMA options support, fio must be built on a system with libnuma-dev(el) |
| 2814 | installed. |
| 2815 | |
| 2816 | .. option:: numa_mem_policy=str |
| 2817 | |
| 2818 | Set this job's memory policy and corresponding NUMA nodes. Format of the |
| 2819 | arguments:: |
| 2820 | |
| 2821 | <mode>[:<nodelist>] |
| 2822 | |
| 2823 | ``mode`` is one of the following memory policies: ``default``, ``prefer``, |
| 2824 | ``bind``, ``interleave`` or ``local``. For ``default`` and ``local`` memory |
| 2825 | policies, no node needs to be specified. For ``prefer``, only one node is |
| 2826 | allowed. For ``bind`` and ``interleave`` the ``nodelist`` may be as |
| 2827 | follows: a comma delimited list of numbers, A-B ranges, or `all`. |
| 2828 | |
| 2829 | .. option:: cgroup=str |
| 2830 | |
| 2831 | Add job to this control group. If it doesn't exist, it will be created. The |
| 2832 | system must have a mounted cgroup blkio mount point for this to work. If |
| 2833 | your system doesn't have it mounted, you can do so with:: |
| 2834 | |
| 2835 | # mount -t cgroup -o blkio none /cgroup |
| 2836 | |
| 2837 | .. option:: cgroup_weight=int |
| 2838 | |
| 2839 | Set the weight of the cgroup to this value. See the documentation that comes |
| 2840 | with the kernel, allowed values are in the range of 100..1000. |
| 2841 | |
| 2842 | .. option:: cgroup_nodelete=bool |
| 2843 | |
| 2844 | Normally fio will delete the cgroups it has created after the job |
| 2845 | completion. To override this behavior and to leave cgroups around after the |
| 2846 | job completion, set ``cgroup_nodelete=1``. This can be useful if one wants |
| 2847 | to inspect various cgroup files after job completion. Default: false. |
| 2848 | |
| 2849 | .. option:: flow_id=int |
| 2850 | |
| 2851 | The ID of the flow. If not specified, it defaults to being a global |
| 2852 | flow. See :option:`flow`. |
| 2853 | |
| 2854 | .. option:: flow=int |
| 2855 | |
| 2856 | Weight in token-based flow control. If this value is used, then there is a |
| 2857 | 'flow counter' which is used to regulate the proportion of activity between |
| 2858 | two or more jobs. Fio attempts to keep this flow counter near zero. The |
| 2859 | ``flow`` parameter stands for how much should be added or subtracted to the |
| 2860 | flow counter on each iteration of the main I/O loop. That is, if one job has |
| 2861 | ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8 |
| 2862 | ratio in how much one runs vs the other. |
| 2863 | |
| 2864 | .. option:: flow_watermark=int |
| 2865 | |
| 2866 | The maximum value that the absolute value of the flow counter is allowed to |
| 2867 | reach before the job must wait for a lower value of the counter. |
| 2868 | |
| 2869 | .. option:: flow_sleep=int |
| 2870 | |
| 2871 | The period of time, in microseconds, to wait after the flow watermark has |
| 2872 | been exceeded before retrying operations. |
| 2873 | |
| 2874 | .. option:: stonewall, wait_for_previous |
| 2875 | |
| 2876 | Wait for preceding jobs in the job file to exit, before starting this |
| 2877 | one. Can be used to insert serialization points in the job file. A stone |
| 2878 | wall also implies starting a new reporting group, see |
| 2879 | :option:`group_reporting`. |
| 2880 | |
| 2881 | .. option:: exitall |
| 2882 | |
| 2883 | By default, fio will continue running all other jobs when one job finishes. |
| 2884 | Sometimes this is not the desired action. Setting ``exitall`` will instead |
| 2885 | make fio terminate all jobs in the same group, as soon as one job of that |
| 2886 | group finishes. |
| 2887 | |
| 2888 | .. option:: exit_what |
| 2889 | |
| 2890 | By default, fio will continue running all other jobs when one job finishes. |
| 2891 | Sometimes this is not the desired action. Setting ``exit_all`` will |
| 2892 | instead make fio terminate all jobs in the same group. The option |
| 2893 | ``exit_what`` allows to control which jobs get terminated when ``exitall`` is |
| 2894 | enabled. The default is ``group`` and does not change the behaviour of |
| 2895 | ``exitall``. The setting ``all`` terminates all jobs. The setting ``stonewall`` |
| 2896 | terminates all currently running jobs across all groups and continues execution |
| 2897 | with the next stonewalled group. |
| 2898 | |
| 2899 | .. option:: exec_prerun=str |
| 2900 | |
| 2901 | Before running this job, issue the command specified through |
| 2902 | :manpage:`system(3)`. Output is redirected in a file called |
| 2903 | :file:`jobname.prerun.txt`. |
| 2904 | |
| 2905 | .. option:: exec_postrun=str |
| 2906 | |
| 2907 | After the job completes, issue the command specified though |
| 2908 | :manpage:`system(3)`. Output is redirected in a file called |
| 2909 | :file:`jobname.postrun.txt`. |
| 2910 | |
| 2911 | .. option:: uid=int |
| 2912 | |
| 2913 | Instead of running as the invoking user, set the user ID to this value |
| 2914 | before the thread/process does any work. |
| 2915 | |
| 2916 | .. option:: gid=int |
| 2917 | |
| 2918 | Set group ID, see :option:`uid`. |
| 2919 | |
| 2920 | |
| 2921 | Verification |
| 2922 | ~~~~~~~~~~~~ |
| 2923 | |
| 2924 | .. option:: verify_only |
| 2925 | |
| 2926 | Do not perform specified workload, only verify data still matches previous |
| 2927 | invocation of this workload. This option allows one to check data multiple |
| 2928 | times at a later date without overwriting it. This option makes sense only |
| 2929 | for workloads that write data, and does not support workloads with the |
| 2930 | :option:`time_based` option set. |
| 2931 | |
| 2932 | .. option:: do_verify=bool |
| 2933 | |
| 2934 | Run the verify phase after a write phase. Only valid if :option:`verify` is |
| 2935 | set. Default: true. |
| 2936 | |
| 2937 | .. option:: verify=str |
| 2938 | |
| 2939 | If writing to a file, fio can verify the file contents after each iteration |
| 2940 | of the job. Each verification method also implies verification of special |
| 2941 | header, which is written to the beginning of each block. This header also |
| 2942 | includes meta information, like offset of the block, block number, timestamp |
| 2943 | when block was written, etc. :option:`verify` can be combined with |
| 2944 | :option:`verify_pattern` option. The allowed values are: |
| 2945 | |
| 2946 | **md5** |
| 2947 | Use an md5 sum of the data area and store it in the header of |
| 2948 | each block. |
| 2949 | |
| 2950 | **crc64** |
| 2951 | Use an experimental crc64 sum of the data area and store it in the |
| 2952 | header of each block. |
| 2953 | |
| 2954 | **crc32c** |
| 2955 | Use a crc32c sum of the data area and store it in the header of |
| 2956 | each block. This will automatically use hardware acceleration |
| 2957 | (e.g. SSE4.2 on an x86 or CRC crypto extensions on ARM64) but will |
| 2958 | fall back to software crc32c if none is found. Generally the |
| 2959 | fastest checksum fio supports when hardware accelerated. |
| 2960 | |
| 2961 | **crc32c-intel** |
| 2962 | Synonym for crc32c. |
| 2963 | |
| 2964 | **crc32** |
| 2965 | Use a crc32 sum of the data area and store it in the header of each |
| 2966 | block. |
| 2967 | |
| 2968 | **crc16** |
| 2969 | Use a crc16 sum of the data area and store it in the header of each |
| 2970 | block. |
| 2971 | |
| 2972 | **crc7** |
| 2973 | Use a crc7 sum of the data area and store it in the header of each |
| 2974 | block. |
| 2975 | |
| 2976 | **xxhash** |
| 2977 | Use xxhash as the checksum function. Generally the fastest software |
| 2978 | checksum that fio supports. |
| 2979 | |
| 2980 | **sha512** |
| 2981 | Use sha512 as the checksum function. |
| 2982 | |
| 2983 | **sha256** |
| 2984 | Use sha256 as the checksum function. |
| 2985 | |
| 2986 | **sha1** |
| 2987 | Use optimized sha1 as the checksum function. |
| 2988 | |
| 2989 | **sha3-224** |
| 2990 | Use optimized sha3-224 as the checksum function. |
| 2991 | |
| 2992 | **sha3-256** |
| 2993 | Use optimized sha3-256 as the checksum function. |
| 2994 | |
| 2995 | **sha3-384** |
| 2996 | Use optimized sha3-384 as the checksum function. |
| 2997 | |
| 2998 | **sha3-512** |
| 2999 | Use optimized sha3-512 as the checksum function. |
| 3000 | |
| 3001 | **meta** |
| 3002 | This option is deprecated, since now meta information is included in |
| 3003 | generic verification header and meta verification happens by |
| 3004 | default. For detailed information see the description of the |
| 3005 | :option:`verify` setting. This option is kept because of |
| 3006 | compatibility's sake with old configurations. Do not use it. |
| 3007 | |
| 3008 | **pattern** |
| 3009 | Verify a strict pattern. Normally fio includes a header with some |
| 3010 | basic information and checksumming, but if this option is set, only |
| 3011 | the specific pattern set with :option:`verify_pattern` is verified. |
| 3012 | |
| 3013 | **null** |
| 3014 | Only pretend to verify. Useful for testing internals with |
| 3015 | :option:`ioengine`\=null, not for much else. |
| 3016 | |
| 3017 | This option can be used for repeated burn-in tests of a system to make sure |
| 3018 | that the written data is also correctly read back. If the data direction |
| 3019 | given is a read or random read, fio will assume that it should verify a |
| 3020 | previously written file. If the data direction includes any form of write, |
| 3021 | the verify will be of the newly written data. |
| 3022 | |
| 3023 | To avoid false verification errors, do not use the norandommap option when |
| 3024 | verifying data with async I/O engines and I/O depths > 1. Or use the |
| 3025 | norandommap and the lfsr random generator together to avoid writing to the |
| 3026 | same offset with muliple outstanding I/Os. |
| 3027 | |
| 3028 | .. option:: verify_offset=int |
| 3029 | |
| 3030 | Swap the verification header with data somewhere else in the block before |
| 3031 | writing. It is swapped back before verifying. |
| 3032 | |
| 3033 | .. option:: verify_interval=int |
| 3034 | |
| 3035 | Write the verification header at a finer granularity than the |
| 3036 | :option:`blocksize`. It will be written for chunks the size of |
| 3037 | ``verify_interval``. :option:`blocksize` should divide this evenly. |
| 3038 | |
| 3039 | .. option:: verify_pattern=str |
| 3040 | |
| 3041 | If set, fio will fill the I/O buffers with this pattern. Fio defaults to |
| 3042 | filling with totally random bytes, but sometimes it's interesting to fill |
| 3043 | with a known pattern for I/O verification purposes. Depending on the width |
| 3044 | of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time (it can |
| 3045 | be either a decimal or a hex number). The ``verify_pattern`` if larger than |
| 3046 | a 32-bit quantity has to be a hex number that starts with either "0x" or |
| 3047 | "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o |
| 3048 | format, which means that for each block offset will be written and then |
| 3049 | verified back, e.g.:: |
| 3050 | |
| 3051 | verify_pattern=%o |
| 3052 | |
| 3053 | Or use combination of everything:: |
| 3054 | |
| 3055 | verify_pattern=0xff%o"abcd"-12 |
| 3056 | |
| 3057 | .. option:: verify_fatal=bool |
| 3058 | |
| 3059 | Normally fio will keep checking the entire contents before quitting on a |
| 3060 | block verification failure. If this option is set, fio will exit the job on |
| 3061 | the first observed failure. Default: false. |
| 3062 | |
| 3063 | .. option:: verify_dump=bool |
| 3064 | |
| 3065 | If set, dump the contents of both the original data block and the data block |
| 3066 | we read off disk to files. This allows later analysis to inspect just what |
| 3067 | kind of data corruption occurred. Off by default. |
| 3068 | |
| 3069 | .. option:: verify_async=int |
| 3070 | |
| 3071 | Fio will normally verify I/O inline from the submitting thread. This option |
| 3072 | takes an integer describing how many async offload threads to create for I/O |
| 3073 | verification instead, causing fio to offload the duty of verifying I/O |
| 3074 | contents to one or more separate threads. If using this offload option, even |
| 3075 | sync I/O engines can benefit from using an :option:`iodepth` setting higher |
| 3076 | than 1, as it allows them to have I/O in flight while verifies are running. |
| 3077 | Defaults to 0 async threads, i.e. verification is not asynchronous. |
| 3078 | |
| 3079 | .. option:: verify_async_cpus=str |
| 3080 | |
| 3081 | Tell fio to set the given CPU affinity on the async I/O verification |
| 3082 | threads. See :option:`cpus_allowed` for the format used. |
| 3083 | |
| 3084 | .. option:: verify_backlog=int |
| 3085 | |
| 3086 | Fio will normally verify the written contents of a job that utilizes verify |
| 3087 | once that job has completed. In other words, everything is written then |
| 3088 | everything is read back and verified. You may want to verify continually |
| 3089 | instead for a variety of reasons. Fio stores the meta data associated with |
| 3090 | an I/O block in memory, so for large verify workloads, quite a bit of memory |
| 3091 | would be used up holding this meta data. If this option is enabled, fio will |
| 3092 | write only N blocks before verifying these blocks. |
| 3093 | |
| 3094 | .. option:: verify_backlog_batch=int |
| 3095 | |
| 3096 | Control how many blocks fio will verify if :option:`verify_backlog` is |
| 3097 | set. If not set, will default to the value of :option:`verify_backlog` |
| 3098 | (meaning the entire queue is read back and verified). If |
| 3099 | ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all |
| 3100 | blocks will be verified, if ``verify_backlog_batch`` is larger than |
| 3101 | :option:`verify_backlog`, some blocks will be verified more than once. |
| 3102 | |
| 3103 | .. option:: verify_state_save=bool |
| 3104 | |
| 3105 | When a job exits during the write phase of a verify workload, save its |
| 3106 | current state. This allows fio to replay up until that point, if the verify |
| 3107 | state is loaded for the verify read phase. The format of the filename is, |
| 3108 | roughly:: |
| 3109 | |
| 3110 | <type>-<jobname>-<jobindex>-verify.state. |
| 3111 | |
| 3112 | <type> is "local" for a local run, "sock" for a client/server socket |
| 3113 | connection, and "ip" (192.168.0.1, for instance) for a networked |
| 3114 | client/server connection. Defaults to true. |
| 3115 | |
| 3116 | .. option:: verify_state_load=bool |
| 3117 | |
| 3118 | If a verify termination trigger was used, fio stores the current write state |
| 3119 | of each thread. This can be used at verification time so that fio knows how |
| 3120 | far it should verify. Without this information, fio will run a full |
| 3121 | verification pass, according to the settings in the job file used. Default |
| 3122 | false. |
| 3123 | |
| 3124 | .. option:: trim_percentage=int |
| 3125 | |
| 3126 | Number of verify blocks to discard/trim. |
| 3127 | |
| 3128 | .. option:: trim_verify_zero=bool |
| 3129 | |
| 3130 | Verify that trim/discarded blocks are returned as zeros. |
| 3131 | |
| 3132 | .. option:: trim_backlog=int |
| 3133 | |
| 3134 | Trim after this number of blocks are written. |
| 3135 | |
| 3136 | .. option:: trim_backlog_batch=int |
| 3137 | |
| 3138 | Trim this number of I/O blocks. |
| 3139 | |
| 3140 | .. option:: experimental_verify=bool |
| 3141 | |
| 3142 | Enable experimental verification. |
| 3143 | |
| 3144 | Steady state |
| 3145 | ~~~~~~~~~~~~ |
| 3146 | |
| 3147 | .. option:: steadystate=str:float, ss=str:float |
| 3148 | |
| 3149 | Define the criterion and limit for assessing steady state performance. The |
| 3150 | first parameter designates the criterion whereas the second parameter sets |
| 3151 | the threshold. When the criterion falls below the threshold for the |
| 3152 | specified duration, the job will stop. For example, `iops_slope:0.1%` will |
| 3153 | direct fio to terminate the job when the least squares regression slope |
| 3154 | falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled |
| 3155 | this will apply to all jobs in the group. Below is the list of available |
| 3156 | steady state assessment criteria. All assessments are carried out using only |
| 3157 | data from the rolling collection window. Threshold limits can be expressed |
| 3158 | as a fixed value or as a percentage of the mean in the collection window. |
| 3159 | |
| 3160 | When using this feature, most jobs should include the :option:`time_based` |
| 3161 | and :option:`runtime` options or the :option:`loops` option so that fio does not |
| 3162 | stop running after it has covered the full size of the specified file(s) or device(s). |
| 3163 | |
| 3164 | **iops** |
| 3165 | Collect IOPS data. Stop the job if all individual IOPS measurements |
| 3166 | are within the specified limit of the mean IOPS (e.g., ``iops:2`` |
| 3167 | means that all individual IOPS values must be within 2 of the mean, |
| 3168 | whereas ``iops:0.2%`` means that all individual IOPS values must be |
| 3169 | within 0.2% of the mean IOPS to terminate the job). |
| 3170 | |
| 3171 | **iops_slope** |
| 3172 | Collect IOPS data and calculate the least squares regression |
| 3173 | slope. Stop the job if the slope falls below the specified limit. |
| 3174 | |
| 3175 | **bw** |
| 3176 | Collect bandwidth data. Stop the job if all individual bandwidth |
| 3177 | measurements are within the specified limit of the mean bandwidth. |
| 3178 | |
| 3179 | **bw_slope** |
| 3180 | Collect bandwidth data and calculate the least squares regression |
| 3181 | slope. Stop the job if the slope falls below the specified limit. |
| 3182 | |
| 3183 | .. option:: steadystate_duration=time, ss_dur=time |
| 3184 | |
| 3185 | A rolling window of this duration will be used to judge whether steady state |
| 3186 | has been reached. Data will be collected once per second. The default is 0 |
| 3187 | which disables steady state detection. When the unit is omitted, the |
| 3188 | value is interpreted in seconds. |
| 3189 | |
| 3190 | .. option:: steadystate_ramp_time=time, ss_ramp=time |
| 3191 | |
| 3192 | Allow the job to run for the specified duration before beginning data |
| 3193 | collection for checking the steady state job termination criterion. The |
| 3194 | default is 0. When the unit is omitted, the value is interpreted in seconds. |
| 3195 | |
| 3196 | |
| 3197 | Measurements and reporting |
| 3198 | ~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| 3199 | |
| 3200 | .. option:: per_job_logs=bool |
| 3201 | |
| 3202 | If set, this generates bw/clat/iops log with per file private filenames. If |
| 3203 | not set, jobs with identical names will share the log filename. Default: |
| 3204 | true. |
| 3205 | |
| 3206 | .. option:: group_reporting |
| 3207 | |
| 3208 | It may sometimes be interesting to display statistics for groups of jobs as |
| 3209 | a whole instead of for each individual job. This is especially true if |
| 3210 | :option:`numjobs` is used; looking at individual thread/process output |
| 3211 | quickly becomes unwieldy. To see the final report per-group instead of |
| 3212 | per-job, use :option:`group_reporting`. Jobs in a file will be part of the |
| 3213 | same reporting group, unless if separated by a :option:`stonewall`, or by |
| 3214 | using :option:`new_group`. |
| 3215 | |
| 3216 | .. option:: new_group |
| 3217 | |
| 3218 | Start a new reporting group. See: :option:`group_reporting`. If not given, |
| 3219 | all jobs in a file will be part of the same reporting group, unless |
| 3220 | separated by a :option:`stonewall`. |
| 3221 | |
| 3222 | .. option:: stats=bool |
| 3223 | |
| 3224 | By default, fio collects and shows final output results for all jobs |
| 3225 | that run. If this option is set to 0, then fio will ignore it in |
| 3226 | the final stat output. |
| 3227 | |
| 3228 | .. option:: write_bw_log=str |
| 3229 | |
| 3230 | If given, write a bandwidth log for this job. Can be used to store data of |
| 3231 | the bandwidth of the jobs in their lifetime. |
| 3232 | |
| 3233 | If no str argument is given, the default filename of |
| 3234 | :file:`jobname_type.x.log` is used. Even when the argument is given, fio |
| 3235 | will still append the type of log. So if one specifies:: |
| 3236 | |
| 3237 | write_bw_log=foo |
| 3238 | |
| 3239 | The actual log name will be :file:`foo_bw.x.log` where `x` is the index |
| 3240 | of the job (`1..N`, where `N` is the number of jobs). If |
| 3241 | :option:`per_job_logs` is false, then the filename will not include the |
| 3242 | `.x` job index. |
| 3243 | |
| 3244 | The included :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these |
| 3245 | text files into nice graphs. See `Log File Formats`_ for how data is |
| 3246 | structured within the file. |
| 3247 | |
| 3248 | .. option:: write_lat_log=str |
| 3249 | |
| 3250 | Same as :option:`write_bw_log`, except this option creates I/O |
| 3251 | submission (e.g., :file:`name_slat.x.log`), completion (e.g., |
| 3252 | :file:`name_clat.x.log`), and total (e.g., :file:`name_lat.x.log`) |
| 3253 | latency files instead. See :option:`write_bw_log` for details about |
| 3254 | the filename format and `Log File Formats`_ for how data is structured |
| 3255 | within the files. |
| 3256 | |
| 3257 | .. option:: write_hist_log=str |
| 3258 | |
| 3259 | Same as :option:`write_bw_log` but writes an I/O completion latency |
| 3260 | histogram file (e.g., :file:`name_hist.x.log`) instead. Note that this |
| 3261 | file will be empty unless :option:`log_hist_msec` has also been set. |
| 3262 | See :option:`write_bw_log` for details about the filename format and |
| 3263 | `Log File Formats`_ for how data is structured within the file. |
| 3264 | |
| 3265 | .. option:: write_iops_log=str |
| 3266 | |
| 3267 | Same as :option:`write_bw_log`, but writes an IOPS file (e.g. |
| 3268 | :file:`name_iops.x.log`) instead. Because fio defaults to individual |
| 3269 | I/O logging, the value entry in the IOPS log will be 1 unless windowed |
| 3270 | logging (see :option:`log_avg_msec`) has been enabled. See |
| 3271 | :option:`write_bw_log` for details about the filename format and `Log |
| 3272 | File Formats`_ for how data is structured within the file. |
| 3273 | |
| 3274 | .. option:: log_avg_msec=int |
| 3275 | |
| 3276 | By default, fio will log an entry in the iops, latency, or bw log for every |
| 3277 | I/O that completes. When writing to the disk log, that can quickly grow to a |
| 3278 | very large size. Setting this option makes fio average the each log entry |
| 3279 | over the specified period of time, reducing the resolution of the log. See |
| 3280 | :option:`log_max_value` as well. Defaults to 0, logging all entries. |
| 3281 | Also see `Log File Formats`_. |
| 3282 | |
| 3283 | .. option:: log_hist_msec=int |
| 3284 | |
| 3285 | Same as :option:`log_avg_msec`, but logs entries for completion latency |
| 3286 | histograms. Computing latency percentiles from averages of intervals using |
| 3287 | :option:`log_avg_msec` is inaccurate. Setting this option makes fio log |
| 3288 | histogram entries over the specified period of time, reducing log sizes for |
| 3289 | high IOPS devices while retaining percentile accuracy. See |
| 3290 | :option:`log_hist_coarseness` and :option:`write_hist_log` as well. |
| 3291 | Defaults to 0, meaning histogram logging is disabled. |
| 3292 | |
| 3293 | .. option:: log_hist_coarseness=int |
| 3294 | |
| 3295 | Integer ranging from 0 to 6, defining the coarseness of the resolution of |
| 3296 | the histogram logs enabled with :option:`log_hist_msec`. For each increment |
| 3297 | in coarseness, fio outputs half as many bins. Defaults to 0, for which |
| 3298 | histogram logs contain 1216 latency bins. See :option:`write_hist_log` |
| 3299 | and `Log File Formats`_. |
| 3300 | |
| 3301 | .. option:: log_max_value=bool |
| 3302 | |
| 3303 | If :option:`log_avg_msec` is set, fio logs the average over that window. If |
| 3304 | you instead want to log the maximum value, set this option to 1. Defaults to |
| 3305 | 0, meaning that averaged values are logged. |
| 3306 | |
| 3307 | .. option:: log_offset=bool |
| 3308 | |
| 3309 | If this is set, the iolog options will include the byte offset for the I/O |
| 3310 | entry as well as the other data values. Defaults to 0 meaning that |
| 3311 | offsets are not present in logs. Also see `Log File Formats`_. |
| 3312 | |
| 3313 | .. option:: log_compression=int |
| 3314 | |
| 3315 | If this is set, fio will compress the I/O logs as it goes, to keep the |
| 3316 | memory footprint lower. When a log reaches the specified size, that chunk is |
| 3317 | removed and compressed in the background. Given that I/O logs are fairly |
| 3318 | highly compressible, this yields a nice memory savings for longer runs. The |
| 3319 | downside is that the compression will consume some background CPU cycles, so |
| 3320 | it may impact the run. This, however, is also true if the logging ends up |
| 3321 | consuming most of the system memory. So pick your poison. The I/O logs are |
| 3322 | saved normally at the end of a run, by decompressing the chunks and storing |
| 3323 | them in the specified log file. This feature depends on the availability of |
| 3324 | zlib. |
| 3325 | |
| 3326 | .. option:: log_compression_cpus=str |
| 3327 | |
| 3328 | Define the set of CPUs that are allowed to handle online log compression for |
| 3329 | the I/O jobs. This can provide better isolation between performance |
| 3330 | sensitive jobs, and background compression work. See |
| 3331 | :option:`cpus_allowed` for the format used. |
| 3332 | |
| 3333 | .. option:: log_store_compressed=bool |
| 3334 | |
| 3335 | If set, fio will store the log files in a compressed format. They can be |
| 3336 | decompressed with fio, using the :option:`--inflate-log` command line |
| 3337 | parameter. The files will be stored with a :file:`.fz` suffix. |
| 3338 | |
| 3339 | .. option:: log_unix_epoch=bool |
| 3340 | |
| 3341 | If set, fio will log Unix timestamps to the log files produced by enabling |
| 3342 | write_type_log for each log type, instead of the default zero-based |
| 3343 | timestamps. |
| 3344 | |
| 3345 | .. option:: block_error_percentiles=bool |
| 3346 | |
| 3347 | If set, record errors in trim block-sized units from writes and trims and |
| 3348 | output a histogram of how many trims it took to get to errors, and what kind |
| 3349 | of error was encountered. |
| 3350 | |
| 3351 | .. option:: bwavgtime=int |
| 3352 | |
| 3353 | Average the calculated bandwidth over the given time. Value is specified in |
| 3354 | milliseconds. If the job also does bandwidth logging through |
| 3355 | :option:`write_bw_log`, then the minimum of this option and |
| 3356 | :option:`log_avg_msec` will be used. Default: 500ms. |
| 3357 | |
| 3358 | .. option:: iopsavgtime=int |
| 3359 | |
| 3360 | Average the calculated IOPS over the given time. Value is specified in |
| 3361 | milliseconds. If the job also does IOPS logging through |
| 3362 | :option:`write_iops_log`, then the minimum of this option and |
| 3363 | :option:`log_avg_msec` will be used. Default: 500ms. |
| 3364 | |
| 3365 | .. option:: disk_util=bool |
| 3366 | |
| 3367 | Generate disk utilization statistics, if the platform supports it. |
| 3368 | Default: true. |
| 3369 | |
| 3370 | .. option:: disable_lat=bool |
| 3371 | |
| 3372 | Disable measurements of total latency numbers. Useful only for cutting back |
| 3373 | the number of calls to :manpage:`gettimeofday(2)`, as that does impact |
| 3374 | performance at really high IOPS rates. Note that to really get rid of a |
| 3375 | large amount of these calls, this option must be used with |
| 3376 | :option:`disable_slat` and :option:`disable_bw_measurement` as well. |
| 3377 | |
| 3378 | .. option:: disable_clat=bool |
| 3379 | |
| 3380 | Disable measurements of completion latency numbers. See |
| 3381 | :option:`disable_lat`. |
| 3382 | |
| 3383 | .. option:: disable_slat=bool |
| 3384 | |
| 3385 | Disable measurements of submission latency numbers. See |
| 3386 | :option:`disable_lat`. |
| 3387 | |
| 3388 | .. option:: disable_bw_measurement=bool, disable_bw=bool |
| 3389 | |
| 3390 | Disable measurements of throughput/bandwidth numbers. See |
| 3391 | :option:`disable_lat`. |
| 3392 | |
| 3393 | .. option:: slat_percentiles=bool |
| 3394 | |
| 3395 | Report submission latency percentiles. Submission latency is not recorded |
| 3396 | for synchronous ioengines. |
| 3397 | |
| 3398 | .. option:: clat_percentiles=bool |
| 3399 | |
| 3400 | Report completion latency percentiles. |
| 3401 | |
| 3402 | .. option:: lat_percentiles=bool |
| 3403 | |
| 3404 | Report total latency percentiles. Total latency is the sum of submission |
| 3405 | latency and completion latency. |
| 3406 | |
| 3407 | .. option:: percentile_list=float_list |
| 3408 | |
| 3409 | Overwrite the default list of percentiles for latencies and the block error |
| 3410 | histogram. Each number is a floating point number in the range (0,100], and |
| 3411 | the maximum length of the list is 20. Use ``:`` to separate the numbers. For |
| 3412 | example, ``--percentile_list=99.5:99.9`` will cause fio to report the |
| 3413 | latency durations below which 99.5% and 99.9% of the observed latencies fell, |
| 3414 | respectively. |
| 3415 | |
| 3416 | .. option:: significant_figures=int |
| 3417 | |
| 3418 | If using :option:`--output-format` of `normal`, set the significant |
| 3419 | figures to this value. Higher values will yield more precise IOPS and |
| 3420 | throughput units, while lower values will round. Requires a minimum |
| 3421 | value of 1 and a maximum value of 10. Defaults to 4. |
| 3422 | |
| 3423 | |
| 3424 | Error handling |
| 3425 | ~~~~~~~~~~~~~~ |
| 3426 | |
| 3427 | .. option:: exitall_on_error |
| 3428 | |
| 3429 | When one job finishes in error, terminate the rest. The default is to wait |
| 3430 | for each job to finish. |
| 3431 | |
| 3432 | .. option:: continue_on_error=str |
| 3433 | |
| 3434 | Normally fio will exit the job on the first observed failure. If this option |
| 3435 | is set, fio will continue the job when there is a 'non-fatal error' (EIO or |
| 3436 | EILSEQ) until the runtime is exceeded or the I/O size specified is |
| 3437 | completed. If this option is used, there are two more stats that are |
| 3438 | appended, the total error count and the first error. The error field given |
| 3439 | in the stats is the first error that was hit during the run. |
| 3440 | |
| 3441 | The allowed values are: |
| 3442 | |
| 3443 | **none** |
| 3444 | Exit on any I/O or verify errors. |
| 3445 | |
| 3446 | **read** |
| 3447 | Continue on read errors, exit on all others. |
| 3448 | |
| 3449 | **write** |
| 3450 | Continue on write errors, exit on all others. |
| 3451 | |
| 3452 | **io** |
| 3453 | Continue on any I/O error, exit on all others. |
| 3454 | |
| 3455 | **verify** |
| 3456 | Continue on verify errors, exit on all others. |
| 3457 | |
| 3458 | **all** |
| 3459 | Continue on all errors. |
| 3460 | |
| 3461 | **0** |
| 3462 | Backward-compatible alias for 'none'. |
| 3463 | |
| 3464 | **1** |
| 3465 | Backward-compatible alias for 'all'. |
| 3466 | |
| 3467 | .. option:: ignore_error=str |
| 3468 | |
| 3469 | Sometimes you want to ignore some errors during test in that case you can |
| 3470 | specify error list for each error type, instead of only being able to |
| 3471 | ignore the default 'non-fatal error' using :option:`continue_on_error`. |
| 3472 | ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for |
| 3473 | given error type is separated with ':'. Error may be symbol ('ENOSPC', |
| 3474 | 'ENOMEM') or integer. Example:: |
| 3475 | |
| 3476 | ignore_error=EAGAIN,ENOSPC:122 |
| 3477 | |
| 3478 | This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from |
| 3479 | WRITE. This option works by overriding :option:`continue_on_error` with |
| 3480 | the list of errors for each error type if any. |
| 3481 | |
| 3482 | .. option:: error_dump=bool |
| 3483 | |
| 3484 | If set dump every error even if it is non fatal, true by default. If |
| 3485 | disabled only fatal error will be dumped. |
| 3486 | |
| 3487 | Running predefined workloads |
| 3488 | ---------------------------- |
| 3489 | |
| 3490 | Fio includes predefined profiles that mimic the I/O workloads generated by |
| 3491 | other tools. |
| 3492 | |
| 3493 | .. option:: profile=str |
| 3494 | |
| 3495 | The predefined workload to run. Current profiles are: |
| 3496 | |
| 3497 | **tiobench** |
| 3498 | Threaded I/O bench (tiotest/tiobench) like workload. |
| 3499 | |
| 3500 | **act** |
| 3501 | Aerospike Certification Tool (ACT) like workload. |
| 3502 | |
| 3503 | To view a profile's additional options use :option:`--cmdhelp` after specifying |
| 3504 | the profile. For example:: |
| 3505 | |
| 3506 | $ fio --profile=act --cmdhelp |
| 3507 | |
| 3508 | Act profile options |
| 3509 | ~~~~~~~~~~~~~~~~~~~ |
| 3510 | |
| 3511 | .. option:: device-names=str |
| 3512 | :noindex: |
| 3513 | |
| 3514 | Devices to use. |
| 3515 | |
| 3516 | .. option:: load=int |
| 3517 | :noindex: |
| 3518 | |
| 3519 | ACT load multiplier. Default: 1. |
| 3520 | |
| 3521 | .. option:: test-duration=time |
| 3522 | :noindex: |
| 3523 | |
| 3524 | How long the entire test takes to run. When the unit is omitted, the value |
| 3525 | is given in seconds. Default: 24h. |
| 3526 | |
| 3527 | .. option:: threads-per-queue=int |
| 3528 | :noindex: |
| 3529 | |
| 3530 | Number of read I/O threads per device. Default: 8. |
| 3531 | |
| 3532 | .. option:: read-req-num-512-blocks=int |
| 3533 | :noindex: |
| 3534 | |
| 3535 | Number of 512B blocks to read at the time. Default: 3. |
| 3536 | |
| 3537 | .. option:: large-block-op-kbytes=int |
| 3538 | :noindex: |
| 3539 | |
| 3540 | Size of large block ops in KiB (writes). Default: 131072. |
| 3541 | |
| 3542 | .. option:: prep |
| 3543 | :noindex: |
| 3544 | |
| 3545 | Set to run ACT prep phase. |
| 3546 | |
| 3547 | Tiobench profile options |
| 3548 | ~~~~~~~~~~~~~~~~~~~~~~~~ |
| 3549 | |
| 3550 | .. option:: size=str |
| 3551 | :noindex: |
| 3552 | |
| 3553 | Size in MiB. |
| 3554 | |
| 3555 | .. option:: block=int |
| 3556 | :noindex: |
| 3557 | |
| 3558 | Block size in bytes. Default: 4096. |
| 3559 | |
| 3560 | .. option:: numruns=int |
| 3561 | :noindex: |
| 3562 | |
| 3563 | Number of runs. |
| 3564 | |
| 3565 | .. option:: dir=str |
| 3566 | :noindex: |
| 3567 | |
| 3568 | Test directory. |
| 3569 | |
| 3570 | .. option:: threads=int |
| 3571 | :noindex: |
| 3572 | |
| 3573 | Number of threads. |
| 3574 | |
| 3575 | Interpreting the output |
| 3576 | ----------------------- |
| 3577 | |
| 3578 | .. |
| 3579 | Example output was based on the following: |
| 3580 | TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --time_based \ |
| 3581 | --rate=1256k --bs=14K --name=quick --runtime=1s --name=mixed \ |
| 3582 | --runtime=2m --rw=rw |
| 3583 | |
| 3584 | Fio spits out a lot of output. While running, fio will display the status of the |
| 3585 | jobs created. An example of that would be:: |
| 3586 | |
| 3587 | 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] |
| 3588 | |
| 3589 | The characters inside the first set of square brackets denote the current status of |
| 3590 | each thread. The first character is the first job defined in the job file, and so |
| 3591 | forth. The possible values (in typical life cycle order) are: |
| 3592 | |
| 3593 | +------+-----+-----------------------------------------------------------+ |
| 3594 | | Idle | Run | | |
| 3595 | +======+=====+===========================================================+ |
| 3596 | | P | | Thread setup, but not started. | |
| 3597 | +------+-----+-----------------------------------------------------------+ |
| 3598 | | C | | Thread created. | |
| 3599 | +------+-----+-----------------------------------------------------------+ |
| 3600 | | I | | Thread initialized, waiting or generating necessary data. | |
| 3601 | +------+-----+-----------------------------------------------------------+ |
| 3602 | | | p | Thread running pre-reading file(s). | |
| 3603 | +------+-----+-----------------------------------------------------------+ |
| 3604 | | | / | Thread is in ramp period. | |
| 3605 | +------+-----+-----------------------------------------------------------+ |
| 3606 | | | R | Running, doing sequential reads. | |
| 3607 | +------+-----+-----------------------------------------------------------+ |
| 3608 | | | r | Running, doing random reads. | |
| 3609 | +------+-----+-----------------------------------------------------------+ |
| 3610 | | | W | Running, doing sequential writes. | |
| 3611 | +------+-----+-----------------------------------------------------------+ |
| 3612 | | | w | Running, doing random writes. | |
| 3613 | +------+-----+-----------------------------------------------------------+ |
| 3614 | | | M | Running, doing mixed sequential reads/writes. | |
| 3615 | +------+-----+-----------------------------------------------------------+ |
| 3616 | | | m | Running, doing mixed random reads/writes. | |
| 3617 | +------+-----+-----------------------------------------------------------+ |
| 3618 | | | D | Running, doing sequential trims. | |
| 3619 | +------+-----+-----------------------------------------------------------+ |
| 3620 | | | d | Running, doing random trims. | |
| 3621 | +------+-----+-----------------------------------------------------------+ |
| 3622 | | | F | Running, currently waiting for :manpage:`fsync(2)`. | |
| 3623 | +------+-----+-----------------------------------------------------------+ |
| 3624 | | | V | Running, doing verification of written data. | |
| 3625 | +------+-----+-----------------------------------------------------------+ |
| 3626 | | f | | Thread finishing. | |
| 3627 | +------+-----+-----------------------------------------------------------+ |
| 3628 | | E | | Thread exited, not reaped by main thread yet. | |
| 3629 | +------+-----+-----------------------------------------------------------+ |
| 3630 | | _ | | Thread reaped. | |
| 3631 | +------+-----+-----------------------------------------------------------+ |
| 3632 | | X | | Thread reaped, exited with an error. | |
| 3633 | +------+-----+-----------------------------------------------------------+ |
| 3634 | | K | | Thread reaped, exited due to signal. | |
| 3635 | +------+-----+-----------------------------------------------------------+ |
| 3636 | |
| 3637 | .. |
| 3638 | Example output was based on the following: |
| 3639 | TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --runtime=58m \ |
| 3640 | --time_based --rate=2512k --bs=256K --numjobs=10 \ |
| 3641 | --name=readers --rw=read --name=writers --rw=write |
| 3642 | |
| 3643 | Fio will condense the thread string as not to take up more space on the command |
| 3644 | line than needed. For instance, if you have 10 readers and 10 writers running, |
| 3645 | the output would look like this:: |
| 3646 | |
| 3647 | 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] |
| 3648 | |
| 3649 | Note that the status string is displayed in order, so it's possible to tell which of |
| 3650 | the jobs are currently doing what. In the example above this means that jobs 1--10 |
| 3651 | are readers and 11--20 are writers. |
| 3652 | |
| 3653 | The other values are fairly self explanatory -- number of threads currently |
| 3654 | running and doing I/O, the number of currently open files (f=), the estimated |
| 3655 | completion percentage, the rate of I/O since last check (read speed listed first, |
| 3656 | then write speed and optionally trim speed) in terms of bandwidth and IOPS, |
| 3657 | and time to completion for the current running group. It's impossible to estimate |
| 3658 | runtime of the following groups (if any). |
| 3659 | |
| 3660 | .. |
| 3661 | Example output was based on the following: |
| 3662 | TZ=UTC fio --iodepth=16 --ioengine=posixaio --filename=/tmp/fiofile \ |
| 3663 | --direct=1 --size=100M --time_based --runtime=50s --rate_iops=89 \ |
| 3664 | --bs=7K --name=Client1 --rw=write |
| 3665 | |
| 3666 | When fio is done (or interrupted by :kbd:`Ctrl-C`), it will show the data for |
| 3667 | each thread, group of threads, and disks in that order. For each overall thread (or |
| 3668 | group) the output looks like:: |
| 3669 | |
| 3670 | Client1: (groupid=0, jobs=1): err= 0: pid=16109: Sat Jun 24 12:07:54 2017 |
| 3671 | write: IOPS=88, BW=623KiB/s (638kB/s)(30.4MiB/50032msec) |
| 3672 | slat (nsec): min=500, max=145500, avg=8318.00, stdev=4781.50 |
| 3673 | clat (usec): min=170, max=78367, avg=4019.02, stdev=8293.31 |
| 3674 | lat (usec): min=174, max=78375, avg=4027.34, stdev=8291.79 |
| 3675 | clat percentiles (usec): |
| 3676 | | 1.00th=[ 302], 5.00th=[ 326], 10.00th=[ 343], 20.00th=[ 363], |
| 3677 | | 30.00th=[ 392], 40.00th=[ 404], 50.00th=[ 416], 60.00th=[ 445], |
| 3678 | | 70.00th=[ 816], 80.00th=[ 6718], 90.00th=[12911], 95.00th=[21627], |
| 3679 | | 99.00th=[43779], 99.50th=[51643], 99.90th=[68682], 99.95th=[72877], |
| 3680 | | 99.99th=[78119] |
| 3681 | bw ( KiB/s): min= 532, max= 686, per=0.10%, avg=622.87, stdev=24.82, samples= 100 |
| 3682 | iops : min= 76, max= 98, avg=88.98, stdev= 3.54, samples= 100 |
| 3683 | lat (usec) : 250=0.04%, 500=64.11%, 750=4.81%, 1000=2.79% |
| 3684 | lat (msec) : 2=4.16%, 4=1.84%, 10=4.90%, 20=11.33%, 50=5.37% |
| 3685 | lat (msec) : 100=0.65% |
| 3686 | cpu : usr=0.27%, sys=0.18%, ctx=12072, majf=0, minf=21 |
| 3687 | IO depths : 1=85.0%, 2=13.1%, 4=1.8%, 8=0.1%, 16=0.0%, 32=0.0%, >=64=0.0% |
| 3688 | submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% |
| 3689 | complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% |
| 3690 | issued rwt: total=0,4450,0, short=0,0,0, dropped=0,0,0 |
| 3691 | latency : target=0, window=0, percentile=100.00%, depth=8 |
| 3692 | |
| 3693 | The job name (or first job's name when using :option:`group_reporting`) is printed, |
| 3694 | along with the group id, count of jobs being aggregated, last error id seen (which |
| 3695 | is 0 when there are no errors), pid/tid of that thread and the time the job/group |
| 3696 | completed. Below are the I/O statistics for each data direction performed (showing |
| 3697 | writes in the example above). In the order listed, they denote: |
| 3698 | |
| 3699 | **read/write/trim** |
| 3700 | The string before the colon shows the I/O direction the statistics |
| 3701 | are for. **IOPS** is the average I/Os performed per second. **BW** |
| 3702 | is the average bandwidth rate shown as: value in power of 2 format |
| 3703 | (value in power of 10 format). The last two values show: (**total |
| 3704 | I/O performed** in power of 2 format / **runtime** of that thread). |
| 3705 | |
| 3706 | **slat** |
| 3707 | Submission latency (**min** being the minimum, **max** being the |
| 3708 | maximum, **avg** being the average, **stdev** being the standard |
| 3709 | deviation). This is the time it took to submit the I/O. For |
| 3710 | sync I/O this row is not displayed as the slat is really the |
| 3711 | completion latency (since queue/complete is one operation there). |
| 3712 | This value can be in nanoseconds, microseconds or milliseconds --- |
| 3713 | fio will choose the most appropriate base and print that (in the |
| 3714 | example above nanoseconds was the best scale). Note: in :option:`--minimal` mode |
| 3715 | latencies are always expressed in microseconds. |
| 3716 | |
| 3717 | **clat** |
| 3718 | Completion latency. Same names as slat, this denotes the time from |
| 3719 | submission to completion of the I/O pieces. For sync I/O, clat will |
| 3720 | usually be equal (or very close) to 0, as the time from submit to |
| 3721 | complete is basically just CPU time (I/O has already been done, see slat |
| 3722 | explanation). |
| 3723 | |
| 3724 | **lat** |
| 3725 | Total latency. Same names as slat and clat, this denotes the time from |
| 3726 | when fio created the I/O unit to completion of the I/O operation. |
| 3727 | |
| 3728 | **bw** |
| 3729 | Bandwidth statistics based on samples. Same names as the xlat stats, |
| 3730 | but also includes the number of samples taken (**samples**) and an |
| 3731 | approximate percentage of total aggregate bandwidth this thread |
| 3732 | received in its group (**per**). This last value is only really |
| 3733 | useful if the threads in this group are on the same disk, since they |
| 3734 | are then competing for disk access. |
| 3735 | |
| 3736 | **iops** |
| 3737 | IOPS statistics based on samples. Same names as bw. |
| 3738 | |
| 3739 | **lat (nsec/usec/msec)** |
| 3740 | The distribution of I/O completion latencies. This is the time from when |
| 3741 | I/O leaves fio and when it gets completed. Unlike the separate |
| 3742 | read/write/trim sections above, the data here and in the remaining |
| 3743 | sections apply to all I/Os for the reporting group. 250=0.04% means that |
| 3744 | 0.04% of the I/Os completed in under 250us. 500=64.11% means that 64.11% |
| 3745 | of the I/Os required 250 to 499us for completion. |
| 3746 | |
| 3747 | **cpu** |
| 3748 | CPU usage. User and system time, along with the number of context |
| 3749 | switches this thread went through, usage of system and user time, and |
| 3750 | finally the number of major and minor page faults. The CPU utilization |
| 3751 | numbers are averages for the jobs in that reporting group, while the |
| 3752 | context and fault counters are summed. |
| 3753 | |
| 3754 | **IO depths** |
| 3755 | The distribution of I/O depths over the job lifetime. The numbers are |
| 3756 | divided into powers of 2 and each entry covers depths from that value |
| 3757 | up to those that are lower than the next entry -- e.g., 16= covers |
| 3758 | depths from 16 to 31. Note that the range covered by a depth |
| 3759 | distribution entry can be different to the range covered by the |
| 3760 | equivalent submit/complete distribution entry. |
| 3761 | |
| 3762 | **IO submit** |
| 3763 | How many pieces of I/O were submitting in a single submit call. Each |
| 3764 | entry denotes that amount and below, until the previous entry -- e.g., |
| 3765 | 16=100% means that we submitted anywhere between 9 to 16 I/Os per submit |
| 3766 | call. Note that the range covered by a submit distribution entry can |
| 3767 | be different to the range covered by the equivalent depth distribution |
| 3768 | entry. |
| 3769 | |
| 3770 | **IO complete** |
| 3771 | Like the above submit number, but for completions instead. |
| 3772 | |
| 3773 | **IO issued rwt** |
| 3774 | The number of read/write/trim requests issued, and how many of them were |
| 3775 | short or dropped. |
| 3776 | |
| 3777 | **IO latency** |
| 3778 | These values are for :option:`latency_target` and related options. When |
| 3779 | these options are engaged, this section describes the I/O depth required |
| 3780 | to meet the specified latency target. |
| 3781 | |
| 3782 | .. |
| 3783 | Example output was based on the following: |
| 3784 | TZ=UTC fio --ioengine=null --iodepth=2 --size=100M --numjobs=2 \ |
| 3785 | --rate_process=poisson --io_limit=32M --name=read --bs=128k \ |
| 3786 | --rate=11M --name=write --rw=write --bs=2k --rate=700k |
| 3787 | |
| 3788 | After each client has been listed, the group statistics are printed. They |
| 3789 | will look like this:: |
| 3790 | |
| 3791 | Run status group 0 (all jobs): |
| 3792 | READ: bw=20.9MiB/s (21.9MB/s), 10.4MiB/s-10.8MiB/s (10.9MB/s-11.3MB/s), io=64.0MiB (67.1MB), run=2973-3069msec |
| 3793 | WRITE: bw=1231KiB/s (1261kB/s), 616KiB/s-621KiB/s (630kB/s-636kB/s), io=64.0MiB (67.1MB), run=52747-53223msec |
| 3794 | |
| 3795 | For each data direction it prints: |
| 3796 | |
| 3797 | **bw** |
| 3798 | Aggregate bandwidth of threads in this group followed by the |
| 3799 | minimum and maximum bandwidth of all the threads in this group. |
| 3800 | Values outside of brackets are power-of-2 format and those |
| 3801 | within are the equivalent value in a power-of-10 format. |
| 3802 | **io** |
| 3803 | Aggregate I/O performed of all threads in this group. The |
| 3804 | format is the same as bw. |
| 3805 | **run** |
| 3806 | The smallest and longest runtimes of the threads in this group. |
| 3807 | |
| 3808 | And finally, the disk statistics are printed. This is Linux specific. They will look like this:: |
| 3809 | |
| 3810 | Disk stats (read/write): |
| 3811 | sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00% |
| 3812 | |
| 3813 | Each value is printed for both reads and writes, with reads first. The |
| 3814 | numbers denote: |
| 3815 | |
| 3816 | **ios** |
| 3817 | Number of I/Os performed by all groups. |
| 3818 | **merge** |
| 3819 | Number of merges performed by the I/O scheduler. |
| 3820 | **ticks** |
| 3821 | Number of ticks we kept the disk busy. |
| 3822 | **in_queue** |
| 3823 | Total time spent in the disk queue. |
| 3824 | **util** |
| 3825 | The disk utilization. A value of 100% means we kept the disk |
| 3826 | busy constantly, 50% would be a disk idling half of the time. |
| 3827 | |
| 3828 | It is also possible to get fio to dump the current output while it is running, |
| 3829 | without terminating the job. To do that, send fio the **USR1** signal. You can |
| 3830 | also get regularly timed dumps by using the :option:`--status-interval` |
| 3831 | parameter, or by creating a file in :file:`/tmp` named |
| 3832 | :file:`fio-dump-status`. If fio sees this file, it will unlink it and dump the |
| 3833 | current output status. |
| 3834 | |
| 3835 | |
| 3836 | Terse output |
| 3837 | ------------ |
| 3838 | |
| 3839 | For scripted usage where you typically want to generate tables or graphs of the |
| 3840 | results, fio can output the results in a semicolon separated format. The format |
| 3841 | is one long line of values, such as:: |
| 3842 | |
| 3843 | 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% |
| 3844 | A description of this job goes here. |
| 3845 | |
| 3846 | The job description (if provided) follows on a second line for terse v2. |
| 3847 | It appears on the same line for other terse versions. |
| 3848 | |
| 3849 | To enable terse output, use the :option:`--minimal` or |
| 3850 | :option:`--output-format`\=terse command line options. The |
| 3851 | first value is the version of the terse output format. If the output has to be |
| 3852 | changed for some reason, this number will be incremented by 1 to signify that |
| 3853 | change. |
| 3854 | |
| 3855 | Split up, the format is as follows (comments in brackets denote when a |
| 3856 | field was introduced or whether it's specific to some terse version): |
| 3857 | |
| 3858 | :: |
| 3859 | |
| 3860 | terse version, fio version [v3], jobname, groupid, error |
| 3861 | |
| 3862 | READ status:: |
| 3863 | |
| 3864 | Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec) |
| 3865 | Submission latency: min, max, mean, stdev (usec) |
| 3866 | Completion latency: min, max, mean, stdev (usec) |
| 3867 | Completion latency percentiles: 20 fields (see below) |
| 3868 | Total latency: min, max, mean, stdev (usec) |
| 3869 | Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5] |
| 3870 | IOPS [v5]: min, max, mean, stdev, number of samples |
| 3871 | |
| 3872 | WRITE status: |
| 3873 | |
| 3874 | :: |
| 3875 | |
| 3876 | Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec) |
| 3877 | Submission latency: min, max, mean, stdev (usec) |
| 3878 | Completion latency: min, max, mean, stdev (usec) |
| 3879 | Completion latency percentiles: 20 fields (see below) |
| 3880 | Total latency: min, max, mean, stdev (usec) |
| 3881 | Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5] |
| 3882 | IOPS [v5]: min, max, mean, stdev, number of samples |
| 3883 | |
| 3884 | TRIM status [all but version 3]: |
| 3885 | |
| 3886 | Fields are similar to READ/WRITE status. |
| 3887 | |
| 3888 | CPU usage:: |
| 3889 | |
| 3890 | user, system, context switches, major faults, minor faults |
| 3891 | |
| 3892 | I/O depths:: |
| 3893 | |
| 3894 | <=1, 2, 4, 8, 16, 32, >=64 |
| 3895 | |
| 3896 | I/O latencies microseconds:: |
| 3897 | |
| 3898 | <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000 |
| 3899 | |
| 3900 | I/O latencies milliseconds:: |
| 3901 | |
| 3902 | <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000 |
| 3903 | |
| 3904 | Disk utilization [v3]:: |
| 3905 | |
| 3906 | disk name, read ios, write ios, read merges, write merges, read ticks, write ticks, |
| 3907 | time spent in queue, disk utilization percentage |
| 3908 | |
| 3909 | Additional Info (dependent on continue_on_error, default off):: |
| 3910 | |
| 3911 | total # errors, first error code |
| 3912 | |
| 3913 | Additional Info (dependent on description being set):: |
| 3914 | |
| 3915 | Text description |
| 3916 | |
| 3917 | Completion latency percentiles can be a grouping of up to 20 sets, so for the |
| 3918 | terse output fio writes all of them. Each field will look like this:: |
| 3919 | |
| 3920 | 1.00%=6112 |
| 3921 | |
| 3922 | which is the Xth percentile, and the `usec` latency associated with it. |
| 3923 | |
| 3924 | For `Disk utilization`, all disks used by fio are shown. So for each disk there |
| 3925 | will be a disk utilization section. |
| 3926 | |
| 3927 | Below is a single line containing short names for each of the fields in the |
| 3928 | minimal output v3, separated by semicolons:: |
| 3929 | |
| 3930 | 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_min;read_clat_max;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_min;write_clat_max;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;cpu_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 |
| 3931 | |
| 3932 | In client/server mode terse output differs from what appears when jobs are run |
| 3933 | locally. Disk utilization data is omitted from the standard terse output and |
| 3934 | for v3 and later appears on its own separate line at the end of each terse |
| 3935 | reporting cycle. |
| 3936 | |
| 3937 | |
| 3938 | JSON output |
| 3939 | ------------ |
| 3940 | |
| 3941 | The `json` output format is intended to be both human readable and convenient |
| 3942 | for automated parsing. For the most part its sections mirror those of the |
| 3943 | `normal` output. The `runtime` value is reported in msec and the `bw` value is |
| 3944 | reported in 1024 bytes per second units. |
| 3945 | |
| 3946 | |
| 3947 | JSON+ output |
| 3948 | ------------ |
| 3949 | |
| 3950 | The `json+` output format is identical to the `json` output format except that it |
| 3951 | adds a full dump of the completion latency bins. Each `bins` object contains a |
| 3952 | set of (key, value) pairs where keys are latency durations and values count how |
| 3953 | many I/Os had completion latencies of the corresponding duration. For example, |
| 3954 | consider: |
| 3955 | |
| 3956 | "bins" : { "87552" : 1, "89600" : 1, "94720" : 1, "96768" : 1, "97792" : 1, "99840" : 1, "100864" : 2, "103936" : 6, "104960" : 534, "105984" : 5995, "107008" : 7529, ... } |
| 3957 | |
| 3958 | This data indicates that one I/O required 87,552ns to complete, two I/Os required |
| 3959 | 100,864ns to complete, and 7529 I/Os required 107,008ns to complete. |
| 3960 | |
| 3961 | Also included with fio is a Python script `fio_jsonplus_clat2csv` that takes |
| 3962 | json+ output and generates CSV-formatted latency data suitable for plotting. |
| 3963 | |
| 3964 | The latency durations actually represent the midpoints of latency intervals. |
| 3965 | For details refer to :file:`stat.h`. |
| 3966 | |
| 3967 | |
| 3968 | Trace file format |
| 3969 | ----------------- |
| 3970 | |
| 3971 | There are two trace file format that you can encounter. The older (v1) format is |
| 3972 | unsupported since version 1.20-rc3 (March 2008). It will still be described |
| 3973 | below in case that you get an old trace and want to understand it. |
| 3974 | |
| 3975 | In any case the trace is a simple text file with a single action per line. |
| 3976 | |
| 3977 | |
| 3978 | Trace file format v1 |
| 3979 | ~~~~~~~~~~~~~~~~~~~~ |
| 3980 | |
| 3981 | Each line represents a single I/O action in the following format:: |
| 3982 | |
| 3983 | rw, offset, length |
| 3984 | |
| 3985 | where `rw=0/1` for read/write, and the `offset` and `length` entries being in bytes. |
| 3986 | |
| 3987 | This format is not supported in fio versions >= 1.20-rc3. |
| 3988 | |
| 3989 | |
| 3990 | Trace file format v2 |
| 3991 | ~~~~~~~~~~~~~~~~~~~~ |
| 3992 | |
| 3993 | The second version of the trace file format was added in fio version 1.17. It |
| 3994 | allows to access more then one file per trace and has a bigger set of possible |
| 3995 | file actions. |
| 3996 | |
| 3997 | The first line of the trace file has to be:: |
| 3998 | |
| 3999 | fio version 2 iolog |
| 4000 | |
| 4001 | Following this can be lines in two different formats, which are described below. |
| 4002 | |
| 4003 | The file management format:: |
| 4004 | |
| 4005 | filename action |
| 4006 | |
| 4007 | The `filename` is given as an absolute path. The `action` can be one of these: |
| 4008 | |
| 4009 | **add** |
| 4010 | Add the given `filename` to the trace. |
| 4011 | **open** |
| 4012 | Open the file with the given `filename`. The `filename` has to have |
| 4013 | been added with the **add** action before. |
| 4014 | **close** |
| 4015 | Close the file with the given `filename`. The file has to have been |
| 4016 | opened before. |
| 4017 | |
| 4018 | |
| 4019 | The file I/O action format:: |
| 4020 | |
| 4021 | filename action offset length |
| 4022 | |
| 4023 | The `filename` is given as an absolute path, and has to have been added and |
| 4024 | opened before it can be used with this format. The `offset` and `length` are |
| 4025 | given in bytes. The `action` can be one of these: |
| 4026 | |
| 4027 | **wait** |
| 4028 | Wait for `offset` microseconds. Everything below 100 is discarded. |
| 4029 | The time is relative to the previous `wait` statement. |
| 4030 | **read** |
| 4031 | Read `length` bytes beginning from `offset`. |
| 4032 | **write** |
| 4033 | Write `length` bytes beginning from `offset`. |
| 4034 | **sync** |
| 4035 | :manpage:`fsync(2)` the file. |
| 4036 | **datasync** |
| 4037 | :manpage:`fdatasync(2)` the file. |
| 4038 | **trim** |
| 4039 | Trim the given file from the given `offset` for `length` bytes. |
| 4040 | |
| 4041 | |
| 4042 | I/O Replay - Merging Traces |
| 4043 | --------------------------- |
| 4044 | |
| 4045 | Colocation is a common practice used to get the most out of a machine. |
| 4046 | Knowing which workloads play nicely with each other and which ones don't is |
| 4047 | a much harder task. While fio can replay workloads concurrently via multiple |
| 4048 | jobs, it leaves some variability up to the scheduler making results harder to |
| 4049 | reproduce. Merging is a way to make the order of events consistent. |
| 4050 | |
| 4051 | Merging is integrated into I/O replay and done when a |
| 4052 | :option:`merge_blktrace_file` is specified. The list of files passed to |
| 4053 | :option:`read_iolog` go through the merge process and output a single file |
| 4054 | stored to the specified file. The output file is passed on as if it were the |
| 4055 | only file passed to :option:`read_iolog`. An example would look like:: |
| 4056 | |
| 4057 | $ fio --read_iolog="<file1>:<file2>" --merge_blktrace_file="<output_file>" |
| 4058 | |
| 4059 | Creating only the merged file can be done by passing the command line argument |
| 4060 | :option:`--merge-blktrace-only`. |
| 4061 | |
| 4062 | Scaling traces can be done to see the relative impact of any particular trace |
| 4063 | being slowed down or sped up. :option:`merge_blktrace_scalars` takes in a colon |
| 4064 | separated list of percentage scalars. It is index paired with the files passed |
| 4065 | to :option:`read_iolog`. |
| 4066 | |
| 4067 | With scaling, it may be desirable to match the running time of all traces. |
| 4068 | This can be done with :option:`merge_blktrace_iters`. It is index paired with |
| 4069 | :option:`read_iolog` just like :option:`merge_blktrace_scalars`. |
| 4070 | |
| 4071 | In an example, given two traces, A and B, each 60s long. If we want to see |
| 4072 | the impact of trace A issuing IOs twice as fast and repeat trace A over the |
| 4073 | runtime of trace B, the following can be done:: |
| 4074 | |
| 4075 | $ fio --read_iolog="<trace_a>:"<trace_b>" --merge_blktrace_file"<output_file>" --merge_blktrace_scalars="50:100" --merge_blktrace_iters="2:1" |
| 4076 | |
| 4077 | This runs trace A at 2x the speed twice for approximately the same runtime as |
| 4078 | a single run of trace B. |
| 4079 | |
| 4080 | |
| 4081 | CPU idleness profiling |
| 4082 | ---------------------- |
| 4083 | |
| 4084 | In some cases, we want to understand CPU overhead in a test. For example, we |
| 4085 | test patches for the specific goodness of whether they reduce CPU usage. |
| 4086 | Fio implements a balloon approach to create a thread per CPU that runs at idle |
| 4087 | priority, meaning that it only runs when nobody else needs the cpu. |
| 4088 | By measuring the amount of work completed by the thread, idleness of each CPU |
| 4089 | can be derived accordingly. |
| 4090 | |
| 4091 | An unit work is defined as touching a full page of unsigned characters. Mean and |
| 4092 | standard deviation of time to complete an unit work is reported in "unit work" |
| 4093 | section. Options can be chosen to report detailed percpu idleness or overall |
| 4094 | system idleness by aggregating percpu stats. |
| 4095 | |
| 4096 | |
| 4097 | Verification and triggers |
| 4098 | ------------------------- |
| 4099 | |
| 4100 | Fio is usually run in one of two ways, when data verification is done. The first |
| 4101 | is a normal write job of some sort with verify enabled. When the write phase has |
| 4102 | completed, fio switches to reads and verifies everything it wrote. The second |
| 4103 | model is running just the write phase, and then later on running the same job |
| 4104 | (but with reads instead of writes) to repeat the same I/O patterns and verify |
| 4105 | the contents. Both of these methods depend on the write phase being completed, |
| 4106 | as fio otherwise has no idea how much data was written. |
| 4107 | |
| 4108 | With verification triggers, fio supports dumping the current write state to |
| 4109 | local files. Then a subsequent read verify workload can load this state and know |
| 4110 | exactly where to stop. This is useful for testing cases where power is cut to a |
| 4111 | server in a managed fashion, for instance. |
| 4112 | |
| 4113 | A verification trigger consists of two things: |
| 4114 | |
| 4115 | 1) Storing the write state of each job. |
| 4116 | 2) Executing a trigger command. |
| 4117 | |
| 4118 | The write state is relatively small, on the order of hundreds of bytes to single |
| 4119 | kilobytes. It contains information on the number of completions done, the last X |
| 4120 | completions, etc. |
| 4121 | |
| 4122 | A trigger is invoked either through creation ('touch') of a specified file in |
| 4123 | the system, or through a timeout setting. If fio is run with |
| 4124 | :option:`--trigger-file`\= :file:`/tmp/trigger-file`, then it will continually |
| 4125 | check for the existence of :file:`/tmp/trigger-file`. When it sees this file, it |
| 4126 | will fire off the trigger (thus saving state, and executing the trigger |
| 4127 | command). |
| 4128 | |
| 4129 | For client/server runs, there's both a local and remote trigger. If fio is |
| 4130 | running as a server backend, it will send the job states back to the client for |
| 4131 | safe storage, then execute the remote trigger, if specified. If a local trigger |
| 4132 | is specified, the server will still send back the write state, but the client |
| 4133 | will then execute the trigger. |
| 4134 | |
| 4135 | Verification trigger example |
| 4136 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| 4137 | |
| 4138 | Let's say we want to run a powercut test on the remote Linux machine 'server'. |
| 4139 | Our write workload is in :file:`write-test.fio`. We want to cut power to 'server' at |
| 4140 | some point during the run, and we'll run this test from the safety or our local |
| 4141 | machine, 'localbox'. On the server, we'll start the fio backend normally:: |
| 4142 | |
| 4143 | server# fio --server |
| 4144 | |
| 4145 | and on the client, we'll fire off the workload:: |
| 4146 | |
| 4147 | localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger-remote="bash -c \"echo b > /proc/sysrq-triger\"" |
| 4148 | |
| 4149 | We set :file:`/tmp/my-trigger` as the trigger file, and we tell fio to execute:: |
| 4150 | |
| 4151 | echo b > /proc/sysrq-trigger |
| 4152 | |
| 4153 | on the server once it has received the trigger and sent us the write state. This |
| 4154 | will work, but it's not **really** cutting power to the server, it's merely |
| 4155 | abruptly rebooting it. If we have a remote way of cutting power to the server |
| 4156 | through IPMI or similar, we could do that through a local trigger command |
| 4157 | instead. Let's assume we have a script that does IPMI reboot of a given hostname, |
| 4158 | ipmi-reboot. On localbox, we could then have run fio with a local trigger |
| 4159 | instead:: |
| 4160 | |
| 4161 | localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger="ipmi-reboot server" |
| 4162 | |
| 4163 | For this case, fio would wait for the server to send us the write state, then |
| 4164 | execute ``ipmi-reboot server`` when that happened. |
| 4165 | |
| 4166 | Loading verify state |
| 4167 | ~~~~~~~~~~~~~~~~~~~~ |
| 4168 | |
| 4169 | To load stored write state, a read verification job file must contain the |
| 4170 | :option:`verify_state_load` option. If that is set, fio will load the previously |
| 4171 | stored state. For a local fio run this is done by loading the files directly, |
| 4172 | and on a client/server run, the server backend will ask the client to send the |
| 4173 | files over and load them from there. |
| 4174 | |
| 4175 | |
| 4176 | Log File Formats |
| 4177 | ---------------- |
| 4178 | |
| 4179 | Fio supports a variety of log file formats, for logging latencies, bandwidth, |
| 4180 | and IOPS. The logs share a common format, which looks like this: |
| 4181 | |
| 4182 | *time* (`msec`), *value*, *data direction*, *block size* (`bytes`), |
| 4183 | *offset* (`bytes`), *command priority* |
| 4184 | |
| 4185 | *Time* for the log entry is always in milliseconds. The *value* logged depends |
| 4186 | on the type of log, it will be one of the following: |
| 4187 | |
| 4188 | **Latency log** |
| 4189 | Value is latency in nsecs |
| 4190 | **Bandwidth log** |
| 4191 | Value is in KiB/sec |
| 4192 | **IOPS log** |
| 4193 | Value is IOPS |
| 4194 | |
| 4195 | *Data direction* is one of the following: |
| 4196 | |
| 4197 | **0** |
| 4198 | I/O is a READ |
| 4199 | **1** |
| 4200 | I/O is a WRITE |
| 4201 | **2** |
| 4202 | I/O is a TRIM |
| 4203 | |
| 4204 | The entry's *block size* is always in bytes. The *offset* is the position in bytes |
| 4205 | from the start of the file for that particular I/O. The logging of the offset can be |
| 4206 | toggled with :option:`log_offset`. |
| 4207 | |
| 4208 | *Command priority* is 0 for normal priority and 1 for high priority. This is controlled |
| 4209 | by the ioengine specific :option:`cmdprio_percentage`. |
| 4210 | |
| 4211 | Fio defaults to logging every individual I/O but when windowed logging is set |
| 4212 | through :option:`log_avg_msec`, either the average (by default) or the maximum |
| 4213 | (:option:`log_max_value` is set) *value* seen over the specified period of time |
| 4214 | is recorded. Each *data direction* seen within the window period will aggregate |
| 4215 | its values in a separate row. Further, when using windowed logging the *block |
| 4216 | size* and *offset* entries will always contain 0. |
| 4217 | |
| 4218 | |
| 4219 | Client/Server |
| 4220 | ------------- |
| 4221 | |
| 4222 | Normally fio is invoked as a stand-alone application on the machine where the |
| 4223 | I/O workload should be generated. However, the backend and frontend of fio can |
| 4224 | be run separately i.e., the fio server can generate an I/O workload on the "Device |
| 4225 | Under Test" while being controlled by a client on another machine. |
| 4226 | |
| 4227 | Start the server on the machine which has access to the storage DUT:: |
| 4228 | |
| 4229 | $ fio --server=args |
| 4230 | |
| 4231 | where `args` defines what fio listens to. The arguments are of the form |
| 4232 | ``type,hostname`` or ``IP,port``. *type* is either ``ip`` (or ip4) for TCP/IP |
| 4233 | v4, ``ip6`` for TCP/IP v6, or ``sock`` for a local unix domain socket. |
| 4234 | *hostname* is either a hostname or IP address, and *port* is the port to listen |
| 4235 | to (only valid for TCP/IP, not a local socket). Some examples: |
| 4236 | |
| 4237 | 1) ``fio --server`` |
| 4238 | |
| 4239 | Start a fio server, listening on all interfaces on the default port (8765). |
| 4240 | |
| 4241 | 2) ``fio --server=ip:hostname,4444`` |
| 4242 | |
| 4243 | Start a fio server, listening on IP belonging to hostname and on port 4444. |
| 4244 | |
| 4245 | 3) ``fio --server=ip6:::1,4444`` |
| 4246 | |
| 4247 | Start a fio server, listening on IPv6 localhost ::1 and on port 4444. |
| 4248 | |
| 4249 | 4) ``fio --server=,4444`` |
| 4250 | |
| 4251 | Start a fio server, listening on all interfaces on port 4444. |
| 4252 | |
| 4253 | 5) ``fio --server=1.2.3.4`` |
| 4254 | |
| 4255 | Start a fio server, listening on IP 1.2.3.4 on the default port. |
| 4256 | |
| 4257 | 6) ``fio --server=sock:/tmp/fio.sock`` |
| 4258 | |
| 4259 | Start a fio server, listening on the local socket :file:`/tmp/fio.sock`. |
| 4260 | |
| 4261 | Once a server is running, a "client" can connect to the fio server with:: |
| 4262 | |
| 4263 | fio <local-args> --client=<server> <remote-args> <job file(s)> |
| 4264 | |
| 4265 | where `local-args` are arguments for the client where it is running, `server` |
| 4266 | is the connect string, and `remote-args` and `job file(s)` are sent to the |
| 4267 | server. The `server` string follows the same format as it does on the server |
| 4268 | side, to allow IP/hostname/socket and port strings. |
| 4269 | |
| 4270 | Fio can connect to multiple servers this way:: |
| 4271 | |
| 4272 | fio --client=<server1> <job file(s)> --client=<server2> <job file(s)> |
| 4273 | |
| 4274 | If the job file is located on the fio server, then you can tell the server to |
| 4275 | load a local file as well. This is done by using :option:`--remote-config` :: |
| 4276 | |
| 4277 | fio --client=server --remote-config /path/to/file.fio |
| 4278 | |
| 4279 | Then fio will open this local (to the server) job file instead of being passed |
| 4280 | one from the client. |
| 4281 | |
| 4282 | If you have many servers (example: 100 VMs/containers), you can input a pathname |
| 4283 | of a file containing host IPs/names as the parameter value for the |
| 4284 | :option:`--client` option. For example, here is an example :file:`host.list` |
| 4285 | file containing 2 hostnames:: |
| 4286 | |
| 4287 | host1.your.dns.domain |
| 4288 | host2.your.dns.domain |
| 4289 | |
| 4290 | The fio command would then be:: |
| 4291 | |
| 4292 | fio --client=host.list <job file(s)> |
| 4293 | |
| 4294 | In this mode, you cannot input server-specific parameters or job files -- all |
| 4295 | servers receive the same job file. |
| 4296 | |
| 4297 | In order to let ``fio --client`` runs use a shared filesystem from multiple |
| 4298 | hosts, ``fio --client`` now prepends the IP address of the server to the |
| 4299 | filename. For example, if fio is using the directory :file:`/mnt/nfs/fio` and is |
| 4300 | writing filename :file:`fileio.tmp`, with a :option:`--client` `hostfile` |
| 4301 | containing two hostnames ``h1`` and ``h2`` with IP addresses 192.168.10.120 and |
| 4302 | 192.168.10.121, then fio will create two files:: |
| 4303 | |
| 4304 | /mnt/nfs/fio/192.168.10.120.fileio.tmp |
| 4305 | /mnt/nfs/fio/192.168.10.121.fileio.tmp |
| 4306 | |
| 4307 | Terse output in client/server mode will differ slightly from what is produced |
| 4308 | when fio is run in stand-alone mode. See the terse output section for details. |