Commit | Line | Data |
---|---|---|
f8b8f7da | 1 | .TH fio 1 "December 2014" "User Manual" |
d60e92d1 AC |
2 | .SH NAME |
3 | fio \- flexible I/O tester | |
4 | .SH SYNOPSIS | |
5 | .B fio | |
6 | [\fIoptions\fR] [\fIjobfile\fR]... | |
7 | .SH DESCRIPTION | |
8 | .B fio | |
9 | is a tool that will spawn a number of threads or processes doing a | |
10 | particular type of I/O action as specified by the user. | |
11 | The typical use of fio is to write a job file matching the I/O load | |
12 | one wants to simulate. | |
13 | .SH OPTIONS | |
14 | .TP | |
49da1240 JA |
15 | .BI \-\-debug \fR=\fPtype |
16 | Enable verbose tracing of various fio actions. May be `all' for all types | |
17 | or individual types separated by a comma (eg \-\-debug=io,file). `help' will | |
18 | list all available tracing options. | |
19 | .TP | |
d60e92d1 AC |
20 | .BI \-\-output \fR=\fPfilename |
21 | Write output to \fIfilename\fR. | |
22 | .TP | |
e28ee21d | 23 | .BI \-\-output-format \fR=\fPformat |
513e37ee VF |
24 | Set the reporting format to \fInormal\fR, \fIterse\fR, \fIjson\fR, or |
25 | \fIjson+\fR. Multiple formats can be selected, separate by a comma. \fIterse\fR | |
26 | is a CSV based format. \fIjson+\fR is like \fIjson\fR, except it adds a full | |
27 | dump of the latency buckets. | |
e28ee21d | 28 | .TP |
b2cecdc2 | 29 | .BI \-\-runtime \fR=\fPruntime |
30 | Limit run time to \fIruntime\fR seconds. | |
d60e92d1 | 31 | .TP |
d60e92d1 AC |
32 | .B \-\-bandwidth\-log |
33 | Generate per-job bandwidth logs. | |
34 | .TP | |
35 | .B \-\-minimal | |
d1429b5c | 36 | Print statistics in a terse, semicolon-delimited format. |
d60e92d1 | 37 | .TP |
f6a7df53 JA |
38 | .B \-\-append-terse |
39 | Print statistics in selected mode AND terse, semicolon-delimited format. | |
40 | Deprecated, use \-\-output-format instead to select multiple formats. | |
41 | .TP | |
49da1240 JA |
42 | .B \-\-version |
43 | Display version information and exit. | |
44 | .TP | |
065248bf | 45 | .BI \-\-terse\-version \fR=\fPversion |
4d658652 | 46 | Set terse version output format (Current version 3, or older version 2). |
49da1240 JA |
47 | .TP |
48 | .B \-\-help | |
49 | Display usage information and exit. | |
50 | .TP | |
fec0f21c JA |
51 | .B \-\-cpuclock-test |
52 | Perform test and validation of internal CPU clock | |
53 | .TP | |
54 | .BI \-\-crctest[\fR=\fPtest] | |
55 | Test the speed of the builtin checksumming functions. If no argument is given, | |
56 | all of them are tested. Or a comma separated list can be passed, in which | |
57 | case the given ones are tested. | |
58 | .TP | |
49da1240 JA |
59 | .BI \-\-cmdhelp \fR=\fPcommand |
60 | Print help information for \fIcommand\fR. May be `all' for all commands. | |
61 | .TP | |
de890a1e SL |
62 | .BI \-\-enghelp \fR=\fPioengine[,command] |
63 | List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR. | |
64 | .TP | |
d60e92d1 AC |
65 | .BI \-\-showcmd \fR=\fPjobfile |
66 | Convert \fIjobfile\fR to a set of command-line options. | |
67 | .TP | |
d60e92d1 AC |
68 | .BI \-\-eta \fR=\fPwhen |
69 | Specifies when real-time ETA estimate should be printed. \fIwhen\fR may | |
70 | be one of `always', `never' or `auto'. | |
71 | .TP | |
30b5d57f JA |
72 | .BI \-\-eta\-newline \fR=\fPtime |
73 | Force an ETA newline for every `time` period passed. | |
74 | .TP | |
75 | .BI \-\-status\-interval \fR=\fPtime | |
76 | Report full output status every `time` period passed. | |
77 | .TP | |
49da1240 JA |
78 | .BI \-\-readonly |
79 | Turn on safety read-only checks, preventing any attempted write. | |
80 | .TP | |
c0a5d35e | 81 | .BI \-\-section \fR=\fPsec |
cf145d90 | 82 | Only run section \fIsec\fR from job file. This option can be used multiple times to add more sections to run. |
c0a5d35e | 83 | .TP |
49da1240 JA |
84 | .BI \-\-alloc\-size \fR=\fPkb |
85 | Set the internal smalloc pool size to \fIkb\fP kilobytes. | |
d60e92d1 | 86 | .TP |
49da1240 JA |
87 | .BI \-\-warnings\-fatal |
88 | All fio parser warnings are fatal, causing fio to exit with an error. | |
9183788d | 89 | .TP |
49da1240 | 90 | .BI \-\-max\-jobs \fR=\fPnr |
57e118a2 | 91 | Set the maximum allowed number of jobs (threads/processes) to support. |
d60e92d1 | 92 | .TP |
49da1240 JA |
93 | .BI \-\-server \fR=\fPargs |
94 | Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section. | |
f57a9c59 | 95 | .TP |
49da1240 JA |
96 | .BI \-\-daemonize \fR=\fPpidfile |
97 | Background a fio server, writing the pid to the given pid file. | |
98 | .TP | |
99 | .BI \-\-client \fR=\fPhost | |
39b5f61e | 100 | Instead of running the jobs locally, send and run them on the given host or set of hosts. See client/server section. |
f2a2ce0e HL |
101 | .TP |
102 | .BI \-\-idle\-prof \fR=\fPoption | |
103 | Report cpu idleness on a system or percpu basis (\fIoption\fP=system,percpu) or run unit work calibration only (\fIoption\fP=calibrate). | |
d60e92d1 AC |
104 | .SH "JOB FILE FORMAT" |
105 | Job files are in `ini' format. They consist of one or more | |
106 | job definitions, which begin with a job name in square brackets and | |
107 | extend to the next job name. The job name can be any ASCII string | |
108 | except `global', which has a special meaning. Following the job name is | |
109 | a sequence of zero or more parameters, one per line, that define the | |
110 | behavior of the job. Any line starting with a `;' or `#' character is | |
d1429b5c | 111 | considered a comment and ignored. |
d9956b64 AC |
112 | .P |
113 | If \fIjobfile\fR is specified as `-', the job file will be read from | |
114 | standard input. | |
d60e92d1 AC |
115 | .SS "Global Section" |
116 | The global section contains default parameters for jobs specified in the | |
117 | job file. A job is only affected by global sections residing above it, | |
118 | and there may be any number of global sections. Specific job definitions | |
119 | may override any parameter set in global sections. | |
120 | .SH "JOB PARAMETERS" | |
121 | .SS Types | |
b470a02c SC |
122 | Some parameters may take arguments of a specific type. |
123 | Anywhere a numeric value is required, an arithmetic expression may be used, | |
d59aa780 JA |
124 | provided it is surrounded by parentheses. Supported operators are: |
125 | .RS | |
126 | .RS | |
127 | .TP | |
128 | .B addition (+) | |
129 | .TP | |
130 | .B subtraction (-) | |
131 | .TP | |
132 | .B multiplication (*) | |
133 | .TP | |
134 | .B division (/) | |
135 | .TP | |
136 | .B modulus (%) | |
137 | .TP | |
138 | .B exponentiation (^) | |
139 | .RE | |
140 | .RE | |
141 | .P | |
142 | For time values in expressions, units are microseconds by default. This is | |
143 | different than for time values not in expressions (not enclosed in | |
144 | parentheses). The types used are: | |
d60e92d1 AC |
145 | .TP |
146 | .I str | |
147 | String: a sequence of alphanumeric characters. | |
148 | .TP | |
149 | .I int | |
d60e92d1 | 150 | SI integer: a whole number, possibly containing a suffix denoting the base unit |
b09da8fa JA |
151 | of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting |
152 | kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5) | |
74454ce4 CE |
153 | respectively. If prefixed with '0x', the value is assumed to be base 16 |
154 | (hexadecimal). A suffix may include a trailing 'b', for instance 'kb' is | |
155 | identical to 'k'. You can specify a base 10 value by using 'KiB', 'MiB','GiB', | |
156 | etc. This is useful for disk drives where values are often given in base 10 | |
157 | values. Specifying '30GiB' will get you 30*1000^3 bytes. | |
158 | When specifying times the default suffix meaning changes, still denoting the | |
159 | base unit of the value, but accepted suffixes are 'D' (days), 'H' (hours), 'M' | |
0de5b26f JA |
160 | (minutes), 'S' Seconds, 'ms' (or msec) milli seconds, 'us' (or 'usec') micro |
161 | seconds. Time values without a unit specify seconds. | |
74454ce4 | 162 | The suffixes are not case sensitive. |
d60e92d1 AC |
163 | .TP |
164 | .I bool | |
165 | Boolean: a true or false value. `0' denotes false, `1' denotes true. | |
166 | .TP | |
167 | .I irange | |
168 | Integer range: a range of integers specified in the format | |
d1429b5c AC |
169 | \fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and |
170 | \fIupper\fR may contain a suffix as described above. If an option allows two | |
171 | sets of ranges, they are separated with a `,' or `/' character. For example: | |
172 | `8\-8k/8M\-4G'. | |
83349190 YH |
173 | .TP |
174 | .I float_list | |
175 | List of floating numbers: A list of floating numbers, separated by | |
cecbfd47 | 176 | a ':' character. |
d60e92d1 AC |
177 | .SS "Parameter List" |
178 | .TP | |
179 | .BI name \fR=\fPstr | |
d9956b64 | 180 | May be used to override the job name. On the command line, this parameter |
d60e92d1 AC |
181 | has the special purpose of signalling the start of a new job. |
182 | .TP | |
183 | .BI description \fR=\fPstr | |
184 | Human-readable description of the job. It is printed when the job is run, but | |
185 | otherwise has no special purpose. | |
186 | .TP | |
187 | .BI directory \fR=\fPstr | |
188 | Prefix filenames with this directory. Used to place files in a location other | |
189 | than `./'. | |
bcbfeefa CE |
190 | You can specify a number of directories by separating the names with a ':' |
191 | character. These directories will be assigned equally distributed to job clones | |
192 | creates with \fInumjobs\fR as long as they are using generated filenames. | |
193 | If specific \fIfilename(s)\fR are set fio will use the first listed directory, | |
194 | and thereby matching the \fIfilename\fR semantic which generates a file each | |
67445b63 JA |
195 | clone if not specified, but let all clones use the same if set. See |
196 | \fIfilename\fR for considerations regarding escaping certain characters on | |
197 | some platforms. | |
d60e92d1 AC |
198 | .TP |
199 | .BI filename \fR=\fPstr | |
200 | .B fio | |
201 | normally makes up a file name based on the job name, thread number, and file | |
d1429b5c | 202 | number. If you want to share files between threads in a job or several jobs, |
de890a1e SL |
203 | specify a \fIfilename\fR for each of them to override the default. |
204 | If the I/O engine is file-based, you can specify | |
d1429b5c AC |
205 | a number of files by separating the names with a `:' character. `\-' is a |
206 | reserved name, meaning stdin or stdout, depending on the read/write direction | |
67445b63 JA |
207 | set. On Windows, disk devices are accessed as \\.\PhysicalDrive0 for the first |
208 | device, \\.\PhysicalDrive1 for the second etc. Note: Windows and FreeBSD | |
209 | prevent write access to areas of the disk containing in-use data | |
210 | (e.g. filesystems). If the wanted filename does need to include a colon, then | |
4904acd5 JM |
211 | escape that with a '\\' character. For instance, if the filename is |
212 | "/dev/dsk/foo@3,0:c", then you would use filename="/dev/dsk/foo@3,0\\:c". | |
d60e92d1 | 213 | .TP |
de98bd30 | 214 | .BI filename_format \fR=\fPstr |
ce594fbe | 215 | If sharing multiple files between jobs, it is usually necessary to have |
de98bd30 JA |
216 | fio generate the exact names that you want. By default, fio will name a file |
217 | based on the default file format specification of | |
218 | \fBjobname.jobnumber.filenumber\fP. With this option, that can be | |
219 | customized. Fio will recognize and replace the following keywords in this | |
220 | string: | |
221 | .RS | |
222 | .RS | |
223 | .TP | |
224 | .B $jobname | |
225 | The name of the worker thread or process. | |
226 | .TP | |
227 | .B $jobnum | |
228 | The incremental number of the worker thread or process. | |
229 | .TP | |
230 | .B $filenum | |
231 | The incremental number of the file for that worker thread or process. | |
232 | .RE | |
233 | .P | |
234 | To have dependent jobs share a set of files, this option can be set to | |
235 | have fio generate filenames that are shared between the two. For instance, | |
236 | if \fBtestfiles.$filenum\fR is specified, file number 4 for any job will | |
237 | be named \fBtestfiles.4\fR. The default of \fB$jobname.$jobnum.$filenum\fR | |
238 | will be used if no other format specifier is given. | |
239 | .RE | |
240 | .P | |
241 | .TP | |
3ce9dcaf JA |
242 | .BI lockfile \fR=\fPstr |
243 | Fio defaults to not locking any files before it does IO to them. If a file or | |
244 | file descriptor is shared, fio can serialize IO to that file to make the end | |
245 | result consistent. This is usual for emulating real workloads that share files. | |
246 | The lock modes are: | |
247 | .RS | |
248 | .RS | |
249 | .TP | |
250 | .B none | |
251 | No locking. This is the default. | |
252 | .TP | |
253 | .B exclusive | |
cf145d90 | 254 | Only one thread or process may do IO at a time, excluding all others. |
3ce9dcaf JA |
255 | .TP |
256 | .B readwrite | |
257 | Read-write locking on the file. Many readers may access the file at the same | |
258 | time, but writes get exclusive access. | |
259 | .RE | |
ce594fbe | 260 | .RE |
3ce9dcaf | 261 | .P |
d60e92d1 AC |
262 | .BI opendir \fR=\fPstr |
263 | Recursively open any files below directory \fIstr\fR. | |
264 | .TP | |
265 | .BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr | |
266 | Type of I/O pattern. Accepted values are: | |
267 | .RS | |
268 | .RS | |
269 | .TP | |
270 | .B read | |
d1429b5c | 271 | Sequential reads. |
d60e92d1 AC |
272 | .TP |
273 | .B write | |
d1429b5c | 274 | Sequential writes. |
d60e92d1 | 275 | .TP |
fa769d44 SW |
276 | .B trim |
277 | Sequential trim (Linux block devices only). | |
278 | .TP | |
d60e92d1 | 279 | .B randread |
d1429b5c | 280 | Random reads. |
d60e92d1 AC |
281 | .TP |
282 | .B randwrite | |
d1429b5c | 283 | Random writes. |
d60e92d1 | 284 | .TP |
fa769d44 SW |
285 | .B randtrim |
286 | Random trim (Linux block devices only). | |
287 | .TP | |
10b023db | 288 | .B rw, readwrite |
d1429b5c | 289 | Mixed sequential reads and writes. |
d60e92d1 | 290 | .TP |
ff6bb260 | 291 | .B randrw |
d1429b5c | 292 | Mixed random reads and writes. |
82a90686 JA |
293 | .TP |
294 | .B trimwrite | |
295 | Trim and write mixed workload. Blocks will be trimmed first, then the same | |
296 | blocks will be written to. | |
d60e92d1 AC |
297 | .RE |
298 | .P | |
38dad62d JA |
299 | For mixed I/O, the default split is 50/50. For certain types of io the result |
300 | may still be skewed a bit, since the speed may be different. It is possible to | |
3b7fa9ec | 301 | specify a number of IO's to do before getting a new offset, this is done by |
38dad62d JA |
302 | appending a `:\fI<nr>\fR to the end of the string given. For a random read, it |
303 | would look like \fBrw=randread:8\fR for passing in an offset modifier with a | |
059b0802 JA |
304 | value of 8. If the postfix is used with a sequential IO pattern, then the value |
305 | specified will be added to the generated offset for each IO. For instance, | |
306 | using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO | |
307 | into sequential IO with holes. See the \fBrw_sequencer\fR option. | |
d60e92d1 AC |
308 | .RE |
309 | .TP | |
38dad62d JA |
310 | .BI rw_sequencer \fR=\fPstr |
311 | If an offset modifier is given by appending a number to the \fBrw=<str>\fR line, | |
312 | then this option controls how that number modifies the IO offset being | |
313 | generated. Accepted values are: | |
314 | .RS | |
315 | .RS | |
316 | .TP | |
317 | .B sequential | |
318 | Generate sequential offset | |
319 | .TP | |
320 | .B identical | |
321 | Generate the same offset | |
322 | .RE | |
323 | .P | |
324 | \fBsequential\fR is only useful for random IO, where fio would normally | |
325 | generate a new random offset for every IO. If you append eg 8 to randread, you | |
326 | would get a new random offset for every 8 IO's. The result would be a seek for | |
327 | only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify | |
328 | that. As sequential IO is already sequential, setting \fBsequential\fR for that | |
329 | would not result in any differences. \fBidentical\fR behaves in a similar | |
330 | fashion, except it sends the same offset 8 number of times before generating a | |
331 | new offset. | |
332 | .RE | |
333 | .P | |
334 | .TP | |
90fef2d1 JA |
335 | .BI kb_base \fR=\fPint |
336 | The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage | |
337 | manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious | |
5c9323fb | 338 | reasons. Allowed values are 1024 or 1000, with 1024 being the default. |
90fef2d1 | 339 | .TP |
771e58be JA |
340 | .BI unified_rw_reporting \fR=\fPbool |
341 | Fio normally reports statistics on a per data direction basis, meaning that | |
342 | read, write, and trim are accounted and reported separately. If this option is | |
cf145d90 | 343 | set fio sums the results and reports them as "mixed" instead. |
771e58be | 344 | .TP |
d60e92d1 | 345 | .BI randrepeat \fR=\fPbool |
56e2a5fc CE |
346 | Seed the random number generator used for random I/O patterns in a predictable |
347 | way so the pattern is repeatable across runs. Default: true. | |
348 | .TP | |
349 | .BI allrandrepeat \fR=\fPbool | |
350 | Seed all random number generators in a predictable way so results are | |
351 | repeatable across runs. Default: false. | |
d60e92d1 | 352 | .TP |
04778baf JA |
353 | .BI randseed \fR=\fPint |
354 | Seed the random number generators based on this seed value, to be able to | |
355 | control what sequence of output is being generated. If not set, the random | |
356 | sequence depends on the \fBrandrepeat\fR setting. | |
357 | .TP | |
a596f047 EG |
358 | .BI fallocate \fR=\fPstr |
359 | Whether pre-allocation is performed when laying down files. Accepted values | |
360 | are: | |
361 | .RS | |
362 | .RS | |
363 | .TP | |
364 | .B none | |
365 | Do not pre-allocate space. | |
366 | .TP | |
367 | .B posix | |
ccc2b328 | 368 | Pre-allocate via \fBposix_fallocate\fR\|(3). |
a596f047 EG |
369 | .TP |
370 | .B keep | |
ccc2b328 | 371 | Pre-allocate via \fBfallocate\fR\|(2) with FALLOC_FL_KEEP_SIZE set. |
a596f047 EG |
372 | .TP |
373 | .B 0 | |
374 | Backward-compatible alias for 'none'. | |
375 | .TP | |
376 | .B 1 | |
377 | Backward-compatible alias for 'posix'. | |
378 | .RE | |
379 | .P | |
380 | May not be available on all supported platforms. 'keep' is only | |
381 | available on Linux. If using ZFS on Solaris this must be set to 'none' | |
382 | because ZFS doesn't support it. Default: 'posix'. | |
383 | .RE | |
7bc8c2cf | 384 | .TP |
d60e92d1 | 385 | .BI fadvise_hint \fR=\fPbool |
cf145d90 | 386 | Use \fBposix_fadvise\fR\|(2) to advise the kernel what I/O patterns |
d1429b5c | 387 | are likely to be issued. Default: true. |
d60e92d1 | 388 | .TP |
37659335 JA |
389 | .BI fadvise_stream \fR=\fPint |
390 | Use \fBposix_fadvise\fR\|(2) to advise the kernel what stream ID the | |
391 | writes issued belong to. Only supported on Linux. Note, this option | |
392 | may change going forward. | |
393 | .TP | |
f7fa2653 | 394 | .BI size \fR=\fPint |
d60e92d1 | 395 | Total size of I/O for this job. \fBfio\fR will run until this many bytes have |
a4d3b4db JA |
396 | been transferred, unless limited by other options (\fBruntime\fR, for instance, |
397 | or increased/descreased by \fBio_size\fR). Unless \fBnrfiles\fR and | |
398 | \fBfilesize\fR options are given, this amount will be divided between the | |
399 | available files for the job. If not set, fio will use the full size of the | |
400 | given files or devices. If the files do not exist, size must be given. It is | |
401 | also possible to give size as a percentage between 1 and 100. If size=20% is | |
402 | given, fio will use 20% of the full size of the given files or devices. | |
403 | .TP | |
404 | .BI io_size \fR=\fPint "\fR,\fB io_limit \fR=\fPint | |
77731b29 JA |
405 | Normally fio operates within the region set by \fBsize\fR, which means that |
406 | the \fBsize\fR option sets both the region and size of IO to be performed. | |
407 | Sometimes that is not what you want. With this option, it is possible to | |
408 | define just the amount of IO that fio should do. For instance, if \fBsize\fR | |
409 | is set to 20G and \fBio_limit\fR is set to 5G, fio will perform IO within | |
a4d3b4db JA |
410 | the first 20G but exit when 5G have been done. The opposite is also |
411 | possible - if \fBsize\fR is set to 20G, and \fBio_size\fR is set to 40G, then | |
412 | fio will do 40G of IO within the 0..20G region. | |
d60e92d1 | 413 | .TP |
74586c1e | 414 | .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool |
3ce9dcaf JA |
415 | Sets size to something really large and waits for ENOSPC (no space left on |
416 | device) as the terminating condition. Only makes sense with sequential write. | |
417 | For a read workload, the mount point will be filled first then IO started on | |
4f12432e JA |
418 | the result. This option doesn't make sense if operating on a raw device node, |
419 | since the size of that is already known by the file system. Additionally, | |
420 | writing beyond end-of-device will not return ENOSPC there. | |
3ce9dcaf | 421 | .TP |
d60e92d1 AC |
422 | .BI filesize \fR=\fPirange |
423 | Individual file sizes. May be a range, in which case \fBfio\fR will select sizes | |
d1429b5c AC |
424 | for files at random within the given range, limited to \fBsize\fR in total (if |
425 | that is given). If \fBfilesize\fR is not specified, each created file is the | |
426 | same size. | |
d60e92d1 | 427 | .TP |
bedc9dc2 JA |
428 | .BI file_append \fR=\fPbool |
429 | Perform IO after the end of the file. Normally fio will operate within the | |
430 | size of a file. If this option is set, then fio will append to the file | |
431 | instead. This has identical behavior to setting \fRoffset\fP to the size | |
0aae4ce7 | 432 | of a file. This option is ignored on non-regular files. |
bedc9dc2 | 433 | .TP |
f7fa2653 | 434 | .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int] |
d9472271 JA |
435 | Block size for I/O units. Default: 4k. Values for reads, writes, and trims |
436 | can be specified separately in the format \fIread\fR,\fIwrite\fR,\fItrim\fR | |
437 | either of which may be empty to leave that value at its default. If a trailing | |
438 | comma isn't given, the remainder will inherit the last value set. | |
d60e92d1 | 439 | .TP |
9183788d | 440 | .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange] |
d1429b5c AC |
441 | Specify a range of I/O block sizes. The issued I/O unit will always be a |
442 | multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies | |
9183788d | 443 | to both reads and writes if only one range is given, but can be specified |
de8f6de9 | 444 | separately with a comma separating the values. Example: bsrange=1k-4k,2k-8k. |
9183788d JA |
445 | Also (see \fBblocksize\fR). |
446 | .TP | |
447 | .BI bssplit \fR=\fPstr | |
448 | This option allows even finer grained control of the block sizes issued, | |
449 | not just even splits between them. With this option, you can weight various | |
450 | block sizes for exact control of the issued IO for a job that has mixed | |
451 | block sizes. The format of the option is bssplit=blocksize/percentage, | |
5982a925 | 452 | optionally adding as many definitions as needed separated by a colon. |
9183788d | 453 | Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k |
c83cdd3e JA |
454 | blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate |
455 | splits to reads and writes. The format is identical to what the | |
456 | \fBbs\fR option accepts, the read and write parts are separated with a | |
457 | comma. | |
d60e92d1 AC |
458 | .TP |
459 | .B blocksize_unaligned\fR,\fP bs_unaligned | |
d1429b5c AC |
460 | If set, any size in \fBblocksize_range\fR may be used. This typically won't |
461 | work with direct I/O, as that normally requires sector alignment. | |
d60e92d1 | 462 | .TP |
2b7a01d0 | 463 | .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int] |
639ce0f3 MS |
464 | At what boundary to align random IO offsets. Defaults to the same as 'blocksize' |
465 | the minimum blocksize given. Minimum alignment is typically 512b | |
2b7a01d0 JA |
466 | for using direct IO, though it usually depends on the hardware block size. |
467 | This option is mutually exclusive with using a random map for files, so it | |
468 | will turn off that option. | |
43602667 | 469 | .TP |
6aca9b3d JA |
470 | .BI bs_is_seq_rand \fR=\fPbool |
471 | If this option is set, fio will use the normal read,write blocksize settings as | |
472 | sequential,random instead. Any random read or write will use the WRITE | |
473 | blocksize settings, and any sequential read or write will use the READ | |
474 | blocksize setting. | |
475 | .TP | |
d60e92d1 | 476 | .B zero_buffers |
cf145d90 | 477 | Initialize buffers with all zeros. Default: fill buffers with random data. |
d60e92d1 | 478 | .TP |
901bb994 JA |
479 | .B refill_buffers |
480 | If this option is given, fio will refill the IO buffers on every submit. The | |
481 | default is to only fill it at init time and reuse that data. Only makes sense | |
482 | if zero_buffers isn't specified, naturally. If data verification is enabled, | |
483 | refill_buffers is also automatically enabled. | |
484 | .TP | |
fd68418e JA |
485 | .BI scramble_buffers \fR=\fPbool |
486 | If \fBrefill_buffers\fR is too costly and the target is using data | |
487 | deduplication, then setting this option will slightly modify the IO buffer | |
488 | contents to defeat normal de-dupe attempts. This is not enough to defeat | |
489 | more clever block compression attempts, but it will stop naive dedupe | |
490 | of blocks. Default: true. | |
491 | .TP | |
c5751c62 JA |
492 | .BI buffer_compress_percentage \fR=\fPint |
493 | If this is set, then fio will attempt to provide IO buffer content (on WRITEs) | |
494 | that compress to the specified level. Fio does this by providing a mix of | |
d1af2894 JA |
495 | random data and a fixed pattern. The fixed pattern is either zeroes, or the |
496 | pattern specified by \fBbuffer_pattern\fR. If the pattern option is used, it | |
497 | might skew the compression ratio slightly. Note that this is per block size | |
498 | unit, for file/disk wide compression level that matches this setting. Note | |
499 | that this is per block size unit, for file/disk wide compression level that | |
500 | matches this setting, you'll also want to set refill_buffers. | |
c5751c62 JA |
501 | .TP |
502 | .BI buffer_compress_chunk \fR=\fPint | |
503 | See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how | |
504 | big the ranges of random data and zeroed data is. Without this set, fio will | |
505 | provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by | |
506 | the remaining zeroed. With this set to some chunk size smaller than the block | |
507 | size, fio can alternate random and zeroed data throughout the IO buffer. | |
508 | .TP | |
ce35b1ec | 509 | .BI buffer_pattern \fR=\fPstr |
cf145d90 CVB |
510 | If set, fio will fill the IO buffers with this pattern. If not set, the contents |
511 | of IO buffers is defined by the other options related to buffer contents. The | |
ce35b1ec | 512 | setting can be any pattern of bytes, and can be prefixed with 0x for hex |
02975b64 | 513 | values. It may also be a string, where the string must then be wrapped with |
2fa5a241 RP |
514 | "", e.g.: |
515 | .RS | |
516 | .RS | |
517 | \fBbuffer_pattern\fR="abcd" | |
518 | .RS | |
519 | or | |
520 | .RE | |
521 | \fBbuffer_pattern\fR=-12 | |
522 | .RS | |
523 | or | |
524 | .RE | |
525 | \fBbuffer_pattern\fR=0xdeadface | |
526 | .RE | |
527 | .LP | |
528 | Also you can combine everything together in any order: | |
529 | .LP | |
530 | .RS | |
531 | \fBbuffer_pattern\fR=0xdeadface"abcd"-12 | |
532 | .RE | |
533 | .RE | |
ce35b1ec | 534 | .TP |
5c94b008 JA |
535 | .BI dedupe_percentage \fR=\fPint |
536 | If set, fio will generate this percentage of identical buffers when writing. | |
537 | These buffers will be naturally dedupable. The contents of the buffers depend | |
538 | on what other buffer compression settings have been set. It's possible to have | |
539 | the individual buffers either fully compressible, or not at all. This option | |
540 | only controls the distribution of unique buffers. | |
541 | .TP | |
d60e92d1 AC |
542 | .BI nrfiles \fR=\fPint |
543 | Number of files to use for this job. Default: 1. | |
544 | .TP | |
545 | .BI openfiles \fR=\fPint | |
546 | Number of files to keep open at the same time. Default: \fBnrfiles\fR. | |
547 | .TP | |
548 | .BI file_service_type \fR=\fPstr | |
549 | Defines how files to service are selected. The following types are defined: | |
550 | .RS | |
551 | .RS | |
552 | .TP | |
553 | .B random | |
5c9323fb | 554 | Choose a file at random. |
d60e92d1 AC |
555 | .TP |
556 | .B roundrobin | |
cf145d90 | 557 | Round robin over opened files (default). |
5c9323fb | 558 | .TP |
6b7f6851 JA |
559 | .B sequential |
560 | Do each file in the set sequentially. | |
d60e92d1 AC |
561 | .RE |
562 | .P | |
cf145d90 | 563 | The number of I/Os to issue before switching to a new file can be specified by |
d60e92d1 AC |
564 | appending `:\fIint\fR' to the service type. |
565 | .RE | |
566 | .TP | |
567 | .BI ioengine \fR=\fPstr | |
568 | Defines how the job issues I/O. The following types are defined: | |
569 | .RS | |
570 | .RS | |
571 | .TP | |
572 | .B sync | |
ccc2b328 | 573 | Basic \fBread\fR\|(2) or \fBwrite\fR\|(2) I/O. \fBfseek\fR\|(2) is used to |
d60e92d1 AC |
574 | position the I/O location. |
575 | .TP | |
a31041ea | 576 | .B psync |
ccc2b328 | 577 | Basic \fBpread\fR\|(2) or \fBpwrite\fR\|(2) I/O. |
a31041ea | 578 | .TP |
9183788d | 579 | .B vsync |
ccc2b328 | 580 | Basic \fBreadv\fR\|(2) or \fBwritev\fR\|(2) I/O. Will emulate queuing by |
cecbfd47 | 581 | coalescing adjacent IOs into a single submission. |
9183788d | 582 | .TP |
a46c5e01 | 583 | .B pvsync |
ccc2b328 | 584 | Basic \fBpreadv\fR\|(2) or \fBpwritev\fR\|(2) I/O. |
a46c5e01 | 585 | .TP |
d60e92d1 | 586 | .B libaio |
de890a1e | 587 | Linux native asynchronous I/O. This ioengine defines engine specific options. |
d60e92d1 AC |
588 | .TP |
589 | .B posixaio | |
ccc2b328 | 590 | POSIX asynchronous I/O using \fBaio_read\fR\|(3) and \fBaio_write\fR\|(3). |
03e20d68 BC |
591 | .TP |
592 | .B solarisaio | |
593 | Solaris native asynchronous I/O. | |
594 | .TP | |
595 | .B windowsaio | |
596 | Windows native asynchronous I/O. | |
d60e92d1 AC |
597 | .TP |
598 | .B mmap | |
ccc2b328 SW |
599 | File is memory mapped with \fBmmap\fR\|(2) and data copied using |
600 | \fBmemcpy\fR\|(3). | |
d60e92d1 AC |
601 | .TP |
602 | .B splice | |
ccc2b328 | 603 | \fBsplice\fR\|(2) is used to transfer the data and \fBvmsplice\fR\|(2) to |
d1429b5c | 604 | transfer data from user-space to the kernel. |
d60e92d1 AC |
605 | .TP |
606 | .B syslet-rw | |
607 | Use the syslet system calls to make regular read/write asynchronous. | |
608 | .TP | |
609 | .B sg | |
610 | SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if | |
ccc2b328 SW |
611 | the target is an sg character device, we use \fBread\fR\|(2) and |
612 | \fBwrite\fR\|(2) for asynchronous I/O. | |
d60e92d1 AC |
613 | .TP |
614 | .B null | |
615 | Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR | |
616 | itself and for debugging and testing purposes. | |
617 | .TP | |
618 | .B net | |
de890a1e SL |
619 | Transfer over the network. The protocol to be used can be defined with the |
620 | \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR, | |
621 | \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified. | |
622 | This ioengine defines engine specific options. | |
d60e92d1 AC |
623 | .TP |
624 | .B netsplice | |
ccc2b328 | 625 | Like \fBnet\fR, but uses \fBsplice\fR\|(2) and \fBvmsplice\fR\|(2) to map data |
de890a1e | 626 | and send/receive. This ioengine defines engine specific options. |
d60e92d1 | 627 | .TP |
53aec0a4 | 628 | .B cpuio |
d60e92d1 AC |
629 | Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and |
630 | \fBcpucycles\fR parameters. | |
631 | .TP | |
632 | .B guasi | |
633 | The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface | |
cecbfd47 | 634 | approach to asynchronous I/O. |
d1429b5c AC |
635 | .br |
636 | See <http://www.xmailserver.org/guasi\-lib.html>. | |
d60e92d1 | 637 | .TP |
21b8aee8 | 638 | .B rdma |
85286c5c BVA |
639 | The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ) |
640 | and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols. | |
21b8aee8 | 641 | .TP |
d60e92d1 AC |
642 | .B external |
643 | Loads an external I/O engine object file. Append the engine filename as | |
644 | `:\fIenginepath\fR'. | |
d54fce84 DM |
645 | .TP |
646 | .B falloc | |
cecbfd47 | 647 | IO engine that does regular linux native fallocate call to simulate data |
d54fce84 DM |
648 | transfer as fio ioengine |
649 | .br | |
650 | DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,) | |
651 | .br | |
0981fd71 | 652 | DIR_WRITE does fallocate(,mode = 0) |
d54fce84 DM |
653 | .br |
654 | DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE) | |
655 | .TP | |
656 | .B e4defrag | |
657 | IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity | |
658 | request to DDIR_WRITE event | |
0d978694 DAG |
659 | .TP |
660 | .B rbd | |
ff6bb260 SL |
661 | IO engine supporting direct access to Ceph Rados Block Devices (RBD) via librbd |
662 | without the need to use the kernel rbd driver. This ioengine defines engine specific | |
0d978694 | 663 | options. |
a7c386f4 | 664 | .TP |
665 | .B gfapi | |
cc47f094 | 666 | Using Glusterfs libgfapi sync interface to direct access to Glusterfs volumes without |
667 | having to go through FUSE. This ioengine defines engine specific | |
668 | options. | |
669 | .TP | |
670 | .B gfapi_async | |
671 | Using Glusterfs libgfapi async interface to direct access to Glusterfs volumes without | |
a7c386f4 | 672 | having to go through FUSE. This ioengine defines engine specific |
673 | options. | |
1b10477b | 674 | .TP |
b74e419e MM |
675 | .B libhdfs |
676 | Read and write through Hadoop (HDFS). The \fBfilename\fR option is used to | |
677 | specify host,port of the hdfs name-node to connect. This engine interprets | |
678 | offsets a little differently. In HDFS, files once created cannot be modified. | |
679 | So random writes are not possible. To imitate this, libhdfs engine expects | |
680 | bunch of small files to be created over HDFS, and engine will randomly pick a | |
681 | file out of those files based on the offset generated by fio backend. (see the | |
682 | example job file to create such files, use rw=write option). Please note, you | |
683 | might want to set necessary environment variables to work with hdfs/libhdfs | |
684 | properly. | |
65fa28ca DE |
685 | .TP |
686 | .B mtd | |
687 | Read, write and erase an MTD character device (e.g., /dev/mtd0). Discards are | |
688 | treated as erases. Depending on the underlying device type, the I/O may have | |
689 | to go in a certain pattern, e.g., on NAND, writing sequentially to erase blocks | |
690 | and discarding before overwriting. The writetrim mode works well for this | |
691 | constraint. | |
d60e92d1 | 692 | .RE |
595e1734 | 693 | .P |
d60e92d1 AC |
694 | .RE |
695 | .TP | |
696 | .BI iodepth \fR=\fPint | |
8489dae4 SK |
697 | Number of I/O units to keep in flight against the file. Note that increasing |
698 | iodepth beyond 1 will not affect synchronous ioengines (except for small | |
cf145d90 | 699 | degress when verify_async is in use). Even async engines may impose OS |
ee72ca09 JA |
700 | restrictions causing the desired depth not to be achieved. This may happen on |
701 | Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is | |
702 | not async on that OS. Keep an eye on the IO depth distribution in the | |
703 | fio output to verify that the achieved depth is as expected. Default: 1. | |
d60e92d1 | 704 | .TP |
e63a0b2f RP |
705 | .BI iodepth_batch \fR=\fPint "\fR,\fP iodepth_batch_submit" \fR=\fPint |
706 | This defines how many pieces of IO to submit at once. It defaults to 1 | |
707 | which means that we submit each IO as soon as it is available, but can | |
708 | be raised to submit bigger batches of IO at the time. If it is set to 0 | |
709 | the \fBiodepth\fR value will be used. | |
d60e92d1 | 710 | .TP |
82407585 | 711 | .BI iodepth_batch_complete_min \fR=\fPint "\fR,\fP iodepth_batch_complete" \fR=\fPint |
3ce9dcaf JA |
712 | This defines how many pieces of IO to retrieve at once. It defaults to 1 which |
713 | means that we'll ask for a minimum of 1 IO in the retrieval process from the | |
714 | kernel. The IO retrieval will go on until we hit the limit set by | |
715 | \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for | |
716 | completed events before queuing more IO. This helps reduce IO latency, at the | |
717 | cost of more retrieval system calls. | |
718 | .TP | |
82407585 RP |
719 | .BI iodepth_batch_complete_max \fR=\fPint |
720 | This defines maximum pieces of IO to | |
721 | retrieve at once. This variable should be used along with | |
722 | \fBiodepth_batch_complete_min\fR=int variable, specifying the range | |
723 | of min and max amount of IO which should be retrieved. By default | |
724 | it is equal to \fBiodepth_batch_complete_min\fR value. | |
725 | ||
726 | Example #1: | |
727 | .RS | |
728 | .RS | |
729 | \fBiodepth_batch_complete_min\fR=1 | |
730 | .LP | |
731 | \fBiodepth_batch_complete_max\fR=<iodepth> | |
732 | .RE | |
733 | ||
734 | which means that we will retrieve at leat 1 IO and up to the | |
735 | whole submitted queue depth. If none of IO has been completed | |
736 | yet, we will wait. | |
737 | ||
738 | Example #2: | |
739 | .RS | |
740 | \fBiodepth_batch_complete_min\fR=0 | |
741 | .LP | |
742 | \fBiodepth_batch_complete_max\fR=<iodepth> | |
743 | .RE | |
744 | ||
745 | which means that we can retrieve up to the whole submitted | |
746 | queue depth, but if none of IO has been completed yet, we will | |
747 | NOT wait and immediately exit the system call. In this example | |
748 | we simply do polling. | |
749 | .RE | |
750 | .TP | |
d60e92d1 AC |
751 | .BI iodepth_low \fR=\fPint |
752 | Low watermark indicating when to start filling the queue again. Default: | |
ff6bb260 | 753 | \fBiodepth\fR. |
d60e92d1 | 754 | .TP |
1ad01bd1 JA |
755 | .BI io_submit_mode \fR=\fPstr |
756 | This option controls how fio submits the IO to the IO engine. The default is | |
757 | \fBinline\fR, which means that the fio job threads submit and reap IO directly. | |
758 | If set to \fBoffload\fR, the job threads will offload IO submission to a | |
759 | dedicated pool of IO threads. This requires some coordination and thus has a | |
760 | bit of extra overhead, especially for lower queue depth IO where it can | |
761 | increase latencies. The benefit is that fio can manage submission rates | |
762 | independently of the device completion rates. This avoids skewed latency | |
763 | reporting if IO gets back up on the device side (the coordinated omission | |
764 | problem). | |
765 | .TP | |
d60e92d1 AC |
766 | .BI direct \fR=\fPbool |
767 | If true, use non-buffered I/O (usually O_DIRECT). Default: false. | |
768 | .TP | |
d01612f3 CM |
769 | .BI atomic \fR=\fPbool |
770 | If value is true, attempt to use atomic direct IO. Atomic writes are guaranteed | |
771 | to be stable once acknowledged by the operating system. Only Linux supports | |
772 | O_ATOMIC right now. | |
773 | .TP | |
d60e92d1 AC |
774 | .BI buffered \fR=\fPbool |
775 | If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter. | |
776 | Default: true. | |
777 | .TP | |
f7fa2653 | 778 | .BI offset \fR=\fPint |
d60e92d1 AC |
779 | Offset in the file to start I/O. Data before the offset will not be touched. |
780 | .TP | |
591e9e06 JA |
781 | .BI offset_increment \fR=\fPint |
782 | If this is provided, then the real offset becomes the | |
69bdd6ba JH |
783 | offset + offset_increment * thread_number, where the thread number is a |
784 | counter that starts at 0 and is incremented for each sub-job (i.e. when | |
785 | numjobs option is specified). This option is useful if there are several jobs | |
786 | which are intended to operate on a file in parallel disjoint segments, with | |
787 | even spacing between the starting points. | |
591e9e06 | 788 | .TP |
ddf24e42 JA |
789 | .BI number_ios \fR=\fPint |
790 | Fio will normally perform IOs until it has exhausted the size of the region | |
791 | set by \fBsize\fR, or if it exhaust the allocated time (or hits an error | |
792 | condition). With this setting, the range/size can be set independently of | |
793 | the number of IOs to perform. When fio reaches this number, it will exit | |
be3fec7d JA |
794 | normally and report status. Note that this does not extend the amount |
795 | of IO that will be done, it will only stop fio if this condition is met | |
796 | before other end-of-job criteria. | |
ddf24e42 | 797 | .TP |
d60e92d1 | 798 | .BI fsync \fR=\fPint |
d1429b5c AC |
799 | How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If |
800 | 0, don't sync. Default: 0. | |
d60e92d1 | 801 | .TP |
5f9099ea JA |
802 | .BI fdatasync \fR=\fPint |
803 | Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the | |
804 | data parts of the file. Default: 0. | |
805 | .TP | |
fa769d44 SW |
806 | .BI write_barrier \fR=\fPint |
807 | Make every Nth write a barrier write. | |
808 | .TP | |
e76b1da4 | 809 | .BI sync_file_range \fR=\fPstr:int |
ccc2b328 SW |
810 | Use \fBsync_file_range\fR\|(2) for every \fRval\fP number of write operations. Fio will |
811 | track range of writes that have happened since the last \fBsync_file_range\fR\|(2) call. | |
e76b1da4 JA |
812 | \fRstr\fP can currently be one or more of: |
813 | .RS | |
814 | .TP | |
815 | .B wait_before | |
816 | SYNC_FILE_RANGE_WAIT_BEFORE | |
817 | .TP | |
818 | .B write | |
819 | SYNC_FILE_RANGE_WRITE | |
820 | .TP | |
821 | .B wait_after | |
822 | SYNC_FILE_RANGE_WRITE | |
823 | .TP | |
824 | .RE | |
825 | .P | |
826 | So if you do sync_file_range=wait_before,write:8, fio would use | |
827 | \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes. | |
ccc2b328 | 828 | Also see the \fBsync_file_range\fR\|(2) man page. This option is Linux specific. |
e76b1da4 | 829 | .TP |
d60e92d1 | 830 | .BI overwrite \fR=\fPbool |
d1429b5c | 831 | If writing, setup the file first and do overwrites. Default: false. |
d60e92d1 AC |
832 | .TP |
833 | .BI end_fsync \fR=\fPbool | |
dbd11ead | 834 | Sync file contents when a write stage has completed. Default: false. |
d60e92d1 AC |
835 | .TP |
836 | .BI fsync_on_close \fR=\fPbool | |
837 | If true, sync file contents on close. This differs from \fBend_fsync\fR in that | |
d1429b5c | 838 | it will happen on every close, not just at the end of the job. Default: false. |
d60e92d1 | 839 | .TP |
d60e92d1 AC |
840 | .BI rwmixread \fR=\fPint |
841 | Percentage of a mixed workload that should be reads. Default: 50. | |
842 | .TP | |
843 | .BI rwmixwrite \fR=\fPint | |
d1429b5c | 844 | Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and |
c35dd7a6 JA |
845 | \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two |
846 | overrides the first. This may interfere with a given rate setting, if fio is | |
847 | asked to limit reads or writes to a certain rate. If that is the case, then | |
848 | the distribution may be skewed. Default: 50. | |
d60e92d1 | 849 | .TP |
92d42d69 JA |
850 | .BI random_distribution \fR=\fPstr:float |
851 | By default, fio will use a completely uniform random distribution when asked | |
852 | to perform random IO. Sometimes it is useful to skew the distribution in | |
853 | specific ways, ensuring that some parts of the data is more hot than others. | |
854 | Fio includes the following distribution models: | |
855 | .RS | |
856 | .TP | |
857 | .B random | |
858 | Uniform random distribution | |
859 | .TP | |
860 | .B zipf | |
861 | Zipf distribution | |
862 | .TP | |
863 | .B pareto | |
864 | Pareto distribution | |
865 | .TP | |
866 | .RE | |
867 | .P | |
868 | When using a zipf or pareto distribution, an input value is also needed to | |
869 | define the access pattern. For zipf, this is the zipf theta. For pareto, | |
870 | it's the pareto power. Fio includes a test program, genzipf, that can be | |
871 | used visualize what the given input values will yield in terms of hit rates. | |
872 | If you wanted to use zipf with a theta of 1.2, you would use | |
873 | random_distribution=zipf:1.2 as the option. If a non-uniform model is used, | |
874 | fio will disable use of the random map. | |
875 | .TP | |
211c9b89 JA |
876 | .BI percentage_random \fR=\fPint |
877 | For a random workload, set how big a percentage should be random. This defaults | |
878 | to 100%, in which case the workload is fully random. It can be set from | |
879 | anywhere from 0 to 100. Setting it to 0 would make the workload fully | |
d9472271 JA |
880 | sequential. It is possible to set different values for reads, writes, and |
881 | trim. To do so, simply use a comma separated list. See \fBblocksize\fR. | |
211c9b89 | 882 | .TP |
d60e92d1 AC |
883 | .B norandommap |
884 | Normally \fBfio\fR will cover every block of the file when doing random I/O. If | |
885 | this parameter is given, a new offset will be chosen without looking at past | |
886 | I/O history. This parameter is mutually exclusive with \fBverify\fR. | |
887 | .TP | |
744492c9 | 888 | .BI softrandommap \fR=\fPbool |
3ce9dcaf JA |
889 | See \fBnorandommap\fR. If fio runs with the random block map enabled and it |
890 | fails to allocate the map, if this option is set it will continue without a | |
891 | random block map. As coverage will not be as complete as with random maps, this | |
892 | option is disabled by default. | |
893 | .TP | |
e8b1961d JA |
894 | .BI random_generator \fR=\fPstr |
895 | Fio supports the following engines for generating IO offsets for random IO: | |
896 | .RS | |
897 | .TP | |
898 | .B tausworthe | |
899 | Strong 2^88 cycle random number generator | |
900 | .TP | |
901 | .B lfsr | |
902 | Linear feedback shift register generator | |
903 | .TP | |
c3546b53 JA |
904 | .B tausworthe64 |
905 | Strong 64-bit 2^258 cycle random number generator | |
906 | .TP | |
e8b1961d JA |
907 | .RE |
908 | .P | |
909 | Tausworthe is a strong random number generator, but it requires tracking on the | |
910 | side if we want to ensure that blocks are only read or written once. LFSR | |
911 | guarantees that we never generate the same offset twice, and it's also less | |
912 | computationally expensive. It's not a true random generator, however, though | |
913 | for IO purposes it's typically good enough. LFSR only works with single block | |
914 | sizes, not with workloads that use multiple block sizes. If used with such a | |
915 | workload, fio may read or write some blocks multiple times. | |
916 | .TP | |
d60e92d1 | 917 | .BI nice \fR=\fPint |
ccc2b328 | 918 | Run job with given nice value. See \fBnice\fR\|(2). |
d60e92d1 AC |
919 | .TP |
920 | .BI prio \fR=\fPint | |
921 | Set I/O priority value of this job between 0 (highest) and 7 (lowest). See | |
ccc2b328 | 922 | \fBionice\fR\|(1). |
d60e92d1 AC |
923 | .TP |
924 | .BI prioclass \fR=\fPint | |
ccc2b328 | 925 | Set I/O priority class. See \fBionice\fR\|(1). |
d60e92d1 AC |
926 | .TP |
927 | .BI thinktime \fR=\fPint | |
928 | Stall job for given number of microseconds between issuing I/Os. | |
929 | .TP | |
930 | .BI thinktime_spin \fR=\fPint | |
931 | Pretend to spend CPU time for given number of microseconds, sleeping the rest | |
932 | of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set. | |
933 | .TP | |
934 | .BI thinktime_blocks \fR=\fPint | |
4d01ece6 JA |
935 | Only valid if thinktime is set - control how many blocks to issue, before |
936 | waiting \fBthinktime\fR microseconds. If not set, defaults to 1 which will | |
937 | make fio wait \fBthinktime\fR microseconds after every block. This | |
938 | effectively makes any queue depth setting redundant, since no more than 1 IO | |
939 | will be queued before we have to complete it and do our thinktime. In other | |
940 | words, this setting effectively caps the queue depth if the latter is larger. | |
d60e92d1 AC |
941 | Default: 1. |
942 | .TP | |
943 | .BI rate \fR=\fPint | |
c35dd7a6 JA |
944 | Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix |
945 | rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each, | |
946 | or you can specify read and writes separately. Using \fBrate\fR=1m,500k would | |
947 | limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes | |
948 | can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only | |
949 | limit writes (to 500KB/sec), the latter will only limit reads. | |
d60e92d1 | 950 | .TP |
6d428bcd | 951 | .BI rate_min \fR=\fPint |
d60e92d1 | 952 | Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth. |
c35dd7a6 JA |
953 | Failing to meet this requirement will cause the job to exit. The same format |
954 | as \fBrate\fR is used for read vs write separation. | |
d60e92d1 AC |
955 | .TP |
956 | .BI rate_iops \fR=\fPint | |
c35dd7a6 JA |
957 | Cap the bandwidth to this number of IOPS. Basically the same as rate, just |
958 | specified independently of bandwidth. The same format as \fBrate\fR is used for | |
de8f6de9 | 959 | read vs write separation. If \fBblocksize\fR is a range, the smallest block |
c35dd7a6 | 960 | size is used as the metric. |
d60e92d1 AC |
961 | .TP |
962 | .BI rate_iops_min \fR=\fPint | |
c35dd7a6 | 963 | If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR |
de8f6de9 | 964 | is used for read vs write separation. |
d60e92d1 | 965 | .TP |
6de65959 JA |
966 | .BI rate_process \fR=\fPstr |
967 | This option controls how fio manages rated IO submissions. The default is | |
968 | \fBlinear\fR, which submits IO in a linear fashion with fixed delays between | |
969 | IOs that gets adjusted based on IO completion rates. If this is set to | |
970 | \fBpoisson\fR, fio will submit IO based on a more real world random request | |
971 | flow, known as the Poisson process | |
5d02b083 JA |
972 | (https://en.wikipedia.org/wiki/Poisson_process). The lambda will be |
973 | 10^6 / IOPS for the given workload. | |
ff6bb260 | 974 | .TP |
6d428bcd JA |
975 | .BI rate_cycle \fR=\fPint |
976 | Average bandwidth for \fBrate\fR and \fBrate_min\fR over this number of | |
d60e92d1 AC |
977 | milliseconds. Default: 1000ms. |
978 | .TP | |
3e260a46 JA |
979 | .BI latency_target \fR=\fPint |
980 | If set, fio will attempt to find the max performance point that the given | |
981 | workload will run at while maintaining a latency below this target. The | |
982 | values is given in microseconds. See \fBlatency_window\fR and | |
983 | \fBlatency_percentile\fR. | |
984 | .TP | |
985 | .BI latency_window \fR=\fPint | |
986 | Used with \fBlatency_target\fR to specify the sample window that the job | |
987 | is run at varying queue depths to test the performance. The value is given | |
988 | in microseconds. | |
989 | .TP | |
990 | .BI latency_percentile \fR=\fPfloat | |
991 | The percentage of IOs that must fall within the criteria specified by | |
992 | \fBlatency_target\fR and \fBlatency_window\fR. If not set, this defaults | |
993 | to 100.0, meaning that all IOs must be equal or below to the value set | |
994 | by \fBlatency_target\fR. | |
995 | .TP | |
15501535 JA |
996 | .BI max_latency \fR=\fPint |
997 | If set, fio will exit the job if it exceeds this maximum latency. It will exit | |
998 | with an ETIME error. | |
999 | .TP | |
d60e92d1 AC |
1000 | .BI cpumask \fR=\fPint |
1001 | Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job | |
1002 | may run on. See \fBsched_setaffinity\fR\|(2). | |
1003 | .TP | |
1004 | .BI cpus_allowed \fR=\fPstr | |
1005 | Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers. | |
1006 | .TP | |
c2acfbac JA |
1007 | .BI cpus_allowed_policy \fR=\fPstr |
1008 | Set the policy of how fio distributes the CPUs specified by \fBcpus_allowed\fR | |
1009 | or \fBcpumask\fR. Two policies are supported: | |
1010 | .RS | |
1011 | .RS | |
1012 | .TP | |
1013 | .B shared | |
1014 | All jobs will share the CPU set specified. | |
1015 | .TP | |
1016 | .B split | |
1017 | Each job will get a unique CPU from the CPU set. | |
1018 | .RE | |
1019 | .P | |
1020 | \fBshared\fR is the default behaviour, if the option isn't specified. If | |
ada083cd JA |
1021 | \fBsplit\fR is specified, then fio will assign one cpu per job. If not enough |
1022 | CPUs are given for the jobs listed, then fio will roundrobin the CPUs in | |
1023 | the set. | |
c2acfbac JA |
1024 | .RE |
1025 | .P | |
1026 | .TP | |
d0b937ed | 1027 | .BI numa_cpu_nodes \fR=\fPstr |
cecbfd47 | 1028 | Set this job running on specified NUMA nodes' CPUs. The arguments allow |
d0b937ed YR |
1029 | comma delimited list of cpu numbers, A-B ranges, or 'all'. |
1030 | .TP | |
1031 | .BI numa_mem_policy \fR=\fPstr | |
1032 | Set this job's memory policy and corresponding NUMA nodes. Format of | |
cecbfd47 | 1033 | the arguments: |
d0b937ed YR |
1034 | .RS |
1035 | .TP | |
1036 | .B <mode>[:<nodelist>] | |
1037 | .TP | |
1038 | .B mode | |
1039 | is one of the following memory policy: | |
1040 | .TP | |
1041 | .B default, prefer, bind, interleave, local | |
1042 | .TP | |
1043 | .RE | |
1044 | For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is | |
1045 | needed to be specified. For \fBprefer\fR, only one node is | |
1046 | allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows | |
1047 | comma delimited list of numbers, A-B ranges, or 'all'. | |
1048 | .TP | |
23ed19b0 CE |
1049 | .BI startdelay \fR=\fPirange |
1050 | Delay start of job for the specified number of seconds. Supports all time | |
1051 | suffixes to allow specification of hours, minutes, seconds and | |
bd66aa2c | 1052 | milliseconds - seconds are the default if a unit is omitted. |
23ed19b0 CE |
1053 | Can be given as a range which causes each thread to choose randomly out of the |
1054 | range. | |
d60e92d1 AC |
1055 | .TP |
1056 | .BI runtime \fR=\fPint | |
1057 | Terminate processing after the specified number of seconds. | |
1058 | .TP | |
1059 | .B time_based | |
1060 | If given, run for the specified \fBruntime\fR duration even if the files are | |
1061 | completely read or written. The same workload will be repeated as many times | |
1062 | as \fBruntime\fR allows. | |
1063 | .TP | |
901bb994 JA |
1064 | .BI ramp_time \fR=\fPint |
1065 | If set, fio will run the specified workload for this amount of time before | |
1066 | logging any performance numbers. Useful for letting performance settle before | |
1067 | logging results, thus minimizing the runtime required for stable results. Note | |
c35dd7a6 JA |
1068 | that the \fBramp_time\fR is considered lead in time for a job, thus it will |
1069 | increase the total runtime if a special timeout or runtime is specified. | |
901bb994 | 1070 | .TP |
d60e92d1 AC |
1071 | .BI invalidate \fR=\fPbool |
1072 | Invalidate buffer-cache for the file prior to starting I/O. Default: true. | |
1073 | .TP | |
1074 | .BI sync \fR=\fPbool | |
1075 | Use synchronous I/O for buffered writes. For the majority of I/O engines, | |
d1429b5c | 1076 | this means using O_SYNC. Default: false. |
d60e92d1 AC |
1077 | .TP |
1078 | .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr | |
1079 | Allocation method for I/O unit buffer. Allowed values are: | |
1080 | .RS | |
1081 | .RS | |
1082 | .TP | |
1083 | .B malloc | |
ccc2b328 | 1084 | Allocate memory with \fBmalloc\fR\|(3). |
d60e92d1 AC |
1085 | .TP |
1086 | .B shm | |
ccc2b328 | 1087 | Use shared memory buffers allocated through \fBshmget\fR\|(2). |
d60e92d1 AC |
1088 | .TP |
1089 | .B shmhuge | |
1090 | Same as \fBshm\fR, but use huge pages as backing. | |
1091 | .TP | |
1092 | .B mmap | |
ccc2b328 | 1093 | Use \fBmmap\fR\|(2) for allocation. Uses anonymous memory unless a filename |
d60e92d1 AC |
1094 | is given after the option in the format `:\fIfile\fR'. |
1095 | .TP | |
1096 | .B mmaphuge | |
1097 | Same as \fBmmap\fR, but use huge files as backing. | |
09c782bb JA |
1098 | .TP |
1099 | .B mmapshared | |
1100 | Same as \fBmmap\fR, but use a MMAP_SHARED mapping. | |
d60e92d1 AC |
1101 | .RE |
1102 | .P | |
1103 | The amount of memory allocated is the maximum allowed \fBblocksize\fR for the | |
1104 | job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work, | |
1105 | the system must have free huge pages allocated. \fBmmaphuge\fR also needs to | |
2e266ba6 JA |
1106 | have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux, |
1107 | huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR | |
1108 | and the documentation for that. Normally you just need to echo an appropriate | |
1109 | number, eg echoing 8 will ensure that the OS has 8 huge pages ready for | |
1110 | use. | |
d60e92d1 AC |
1111 | .RE |
1112 | .TP | |
d392365e | 1113 | .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint |
cecbfd47 | 1114 | This indicates the memory alignment of the IO memory buffers. Note that the |
d529ee19 JA |
1115 | given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR |
1116 | the alignment of the following buffers are given by the \fBbs\fR used. In | |
1117 | other words, if using a \fBbs\fR that is a multiple of the page sized in the | |
1118 | system, all buffers will be aligned to this value. If using a \fBbs\fR that | |
1119 | is not page aligned, the alignment of subsequent IO memory buffers is the | |
1120 | sum of the \fBiomem_align\fR and \fBbs\fR used. | |
1121 | .TP | |
f7fa2653 | 1122 | .BI hugepage\-size \fR=\fPint |
d60e92d1 | 1123 | Defines the size of a huge page. Must be at least equal to the system setting. |
b22989b9 | 1124 | Should be a multiple of 1MB. Default: 4MB. |
d60e92d1 AC |
1125 | .TP |
1126 | .B exitall | |
1127 | Terminate all jobs when one finishes. Default: wait for each job to finish. | |
1128 | .TP | |
f9cafb12 JA |
1129 | .B exitall_on_error \fR=\fPbool |
1130 | Terminate all jobs if one job finishes in error. Default: wait for each job | |
1131 | to finish. | |
1132 | .TP | |
d60e92d1 AC |
1133 | .BI bwavgtime \fR=\fPint |
1134 | Average bandwidth calculations over the given time in milliseconds. Default: | |
1135 | 500ms. | |
1136 | .TP | |
c8eeb9df JA |
1137 | .BI iopsavgtime \fR=\fPint |
1138 | Average IOPS calculations over the given time in milliseconds. Default: | |
1139 | 500ms. | |
1140 | .TP | |
d60e92d1 | 1141 | .BI create_serialize \fR=\fPbool |
d1429b5c | 1142 | If true, serialize file creation for the jobs. Default: true. |
d60e92d1 AC |
1143 | .TP |
1144 | .BI create_fsync \fR=\fPbool | |
ccc2b328 | 1145 | \fBfsync\fR\|(2) data file after creation. Default: true. |
d60e92d1 | 1146 | .TP |
6b7f6851 JA |
1147 | .BI create_on_open \fR=\fPbool |
1148 | If true, the files are not created until they are opened for IO by the job. | |
1149 | .TP | |
25460cf6 JA |
1150 | .BI create_only \fR=\fPbool |
1151 | If true, fio will only run the setup phase of the job. If files need to be | |
1152 | laid out or updated on disk, only that will be done. The actual job contents | |
1153 | are not executed. | |
1154 | .TP | |
2378826d JA |
1155 | .BI allow_file_create \fR=\fPbool |
1156 | If true, fio is permitted to create files as part of its workload. This is | |
1157 | the default behavior. If this option is false, then fio will error out if the | |
1158 | files it needs to use don't already exist. Default: true. | |
1159 | .TP | |
e81ecca3 JA |
1160 | .BI allow_mounted_write \fR=\fPbool |
1161 | If this isn't set, fio will abort jobs that are destructive (eg that write) | |
1162 | to what appears to be a mounted device or partition. This should help catch | |
1163 | creating inadvertently destructive tests, not realizing that the test will | |
1164 | destroy data on the mounted file system. Default: false. | |
1165 | .TP | |
e9f48479 JA |
1166 | .BI pre_read \fR=\fPbool |
1167 | If this is given, files will be pre-read into memory before starting the given | |
1168 | IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is | |
9c0d2241 JA |
1169 | pointless to pre-read and then drop the cache. This will only work for IO |
1170 | engines that are seekable, since they allow you to read the same data | |
1171 | multiple times. Thus it will not work on eg network or splice IO. | |
e9f48479 | 1172 | .TP |
d60e92d1 AC |
1173 | .BI unlink \fR=\fPbool |
1174 | Unlink job files when done. Default: false. | |
1175 | .TP | |
1176 | .BI loops \fR=\fPint | |
1177 | Specifies the number of iterations (runs of the same workload) of this job. | |
1178 | Default: 1. | |
1179 | .TP | |
5e4c7118 JA |
1180 | .BI verify_only \fR=\fPbool |
1181 | Do not perform the specified workload, only verify data still matches previous | |
1182 | invocation of this workload. This option allows one to check data multiple | |
1183 | times at a later date without overwriting it. This option makes sense only for | |
1184 | workloads that write data, and does not support workloads with the | |
1185 | \fBtime_based\fR option set. | |
1186 | .TP | |
d60e92d1 AC |
1187 | .BI do_verify \fR=\fPbool |
1188 | Run the verify phase after a write phase. Only valid if \fBverify\fR is set. | |
1189 | Default: true. | |
1190 | .TP | |
1191 | .BI verify \fR=\fPstr | |
b638d82f RP |
1192 | Method of verifying file contents after each iteration of the job. Each |
1193 | verification method also implies verification of special header, which is | |
1194 | written to the beginning of each block. This header also includes meta | |
1195 | information, like offset of the block, block number, timestamp when block | |
1196 | was written, etc. \fBverify\fR=str can be combined with \fBverify_pattern\fR=str | |
1197 | option. The allowed values are: | |
d60e92d1 AC |
1198 | .RS |
1199 | .RS | |
1200 | .TP | |
844ea602 | 1201 | .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 xxhash |
0539d758 JA |
1202 | Store appropriate checksum in the header of each block. crc32c-intel is |
1203 | hardware accelerated SSE4.2 driven, falls back to regular crc32c if | |
1204 | not supported by the system. | |
d60e92d1 AC |
1205 | .TP |
1206 | .B meta | |
b638d82f RP |
1207 | This option is deprecated, since now meta information is included in generic |
1208 | verification header and meta verification happens by default. For detailed | |
1209 | information see the description of the \fBverify\fR=str setting. This option | |
1210 | is kept because of compatibility's sake with old configurations. Do not use it. | |
d60e92d1 | 1211 | .TP |
59245381 JA |
1212 | .B pattern |
1213 | Verify a strict pattern. Normally fio includes a header with some basic | |
1214 | information and checksumming, but if this option is set, only the | |
1215 | specific pattern set with \fBverify_pattern\fR is verified. | |
1216 | .TP | |
d60e92d1 AC |
1217 | .B null |
1218 | Pretend to verify. Used for testing internals. | |
1219 | .RE | |
b892dc08 JA |
1220 | |
1221 | This option can be used for repeated burn-in tests of a system to make sure | |
1222 | that the written data is also correctly read back. If the data direction given | |
1223 | is a read or random read, fio will assume that it should verify a previously | |
1224 | written file. If the data direction includes any form of write, the verify will | |
1225 | be of the newly written data. | |
d60e92d1 AC |
1226 | .RE |
1227 | .TP | |
5c9323fb | 1228 | .BI verifysort \fR=\fPbool |
d60e92d1 AC |
1229 | If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to |
1230 | read them back in a sorted manner. Default: true. | |
1231 | .TP | |
fa769d44 SW |
1232 | .BI verifysort_nr \fR=\fPint |
1233 | Pre-load and sort verify blocks for a read workload. | |
1234 | .TP | |
f7fa2653 | 1235 | .BI verify_offset \fR=\fPint |
d60e92d1 | 1236 | Swap the verification header with data somewhere else in the block before |
d1429b5c | 1237 | writing. It is swapped back before verifying. |
d60e92d1 | 1238 | .TP |
f7fa2653 | 1239 | .BI verify_interval \fR=\fPint |
d60e92d1 AC |
1240 | Write the verification header for this number of bytes, which should divide |
1241 | \fBblocksize\fR. Default: \fBblocksize\fR. | |
1242 | .TP | |
996093bb JA |
1243 | .BI verify_pattern \fR=\fPstr |
1244 | If set, fio will fill the io buffers with this pattern. Fio defaults to filling | |
1245 | with totally random bytes, but sometimes it's interesting to fill with a known | |
1246 | pattern for io verification purposes. Depending on the width of the pattern, | |
1247 | fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a | |
1248 | decimal or a hex number). The verify_pattern if larger than a 32-bit quantity | |
1249 | has to be a hex number that starts with either "0x" or "0X". Use with | |
b638d82f RP |
1250 | \fBverify\fP=str. Also, verify_pattern supports %o format, which means that for |
1251 | each block offset will be written and then verifyied back, e.g.: | |
2fa5a241 RP |
1252 | .RS |
1253 | .RS | |
1254 | \fBverify_pattern\fR=%o | |
1255 | .RE | |
1256 | Or use combination of everything: | |
1257 | .LP | |
1258 | .RS | |
1259 | \fBverify_pattern\fR=0xff%o"abcd"-21 | |
1260 | .RE | |
1261 | .RE | |
996093bb | 1262 | .TP |
d60e92d1 AC |
1263 | .BI verify_fatal \fR=\fPbool |
1264 | If true, exit the job on the first observed verification failure. Default: | |
1265 | false. | |
1266 | .TP | |
b463e936 JA |
1267 | .BI verify_dump \fR=\fPbool |
1268 | If set, dump the contents of both the original data block and the data block we | |
1269 | read off disk to files. This allows later analysis to inspect just what kind of | |
ef71e317 | 1270 | data corruption occurred. Off by default. |
b463e936 | 1271 | .TP |
e8462bd8 JA |
1272 | .BI verify_async \fR=\fPint |
1273 | Fio will normally verify IO inline from the submitting thread. This option | |
1274 | takes an integer describing how many async offload threads to create for IO | |
1275 | verification instead, causing fio to offload the duty of verifying IO contents | |
c85c324c JA |
1276 | to one or more separate threads. If using this offload option, even sync IO |
1277 | engines can benefit from using an \fBiodepth\fR setting higher than 1, as it | |
1278 | allows them to have IO in flight while verifies are running. | |
e8462bd8 JA |
1279 | .TP |
1280 | .BI verify_async_cpus \fR=\fPstr | |
1281 | Tell fio to set the given CPU affinity on the async IO verification threads. | |
1282 | See \fBcpus_allowed\fP for the format used. | |
1283 | .TP | |
6f87418f JA |
1284 | .BI verify_backlog \fR=\fPint |
1285 | Fio will normally verify the written contents of a job that utilizes verify | |
1286 | once that job has completed. In other words, everything is written then | |
1287 | everything is read back and verified. You may want to verify continually | |
1288 | instead for a variety of reasons. Fio stores the meta data associated with an | |
1289 | IO block in memory, so for large verify workloads, quite a bit of memory would | |
092f707f DN |
1290 | be used up holding this meta data. If this option is enabled, fio will write |
1291 | only N blocks before verifying these blocks. | |
6f87418f JA |
1292 | .TP |
1293 | .BI verify_backlog_batch \fR=\fPint | |
1294 | Control how many blocks fio will verify if verify_backlog is set. If not set, | |
1295 | will default to the value of \fBverify_backlog\fR (meaning the entire queue is | |
ff6bb260 SL |
1296 | read back and verified). If \fBverify_backlog_batch\fR is less than |
1297 | \fBverify_backlog\fR then not all blocks will be verified, if | |
092f707f DN |
1298 | \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks |
1299 | will be verified more than once. | |
6f87418f | 1300 | .TP |
fa769d44 SW |
1301 | .BI trim_percentage \fR=\fPint |
1302 | Number of verify blocks to discard/trim. | |
1303 | .TP | |
1304 | .BI trim_verify_zero \fR=\fPbool | |
1305 | Verify that trim/discarded blocks are returned as zeroes. | |
1306 | .TP | |
1307 | .BI trim_backlog \fR=\fPint | |
1308 | Trim after this number of blocks are written. | |
1309 | .TP | |
1310 | .BI trim_backlog_batch \fR=\fPint | |
1311 | Trim this number of IO blocks. | |
1312 | .TP | |
1313 | .BI experimental_verify \fR=\fPbool | |
1314 | Enable experimental verification. | |
1315 | .TP | |
ca09be4b JA |
1316 | .BI verify_state_save \fR=\fPbool |
1317 | When a job exits during the write phase of a verify workload, save its | |
1318 | current state. This allows fio to replay up until that point, if the | |
1319 | verify state is loaded for the verify read phase. | |
1320 | .TP | |
1321 | .BI verify_state_load \fR=\fPbool | |
1322 | If a verify termination trigger was used, fio stores the current write | |
1323 | state of each thread. This can be used at verification time so that fio | |
1324 | knows how far it should verify. Without this information, fio will run | |
1325 | a full verification pass, according to the settings in the job file used. | |
1326 | .TP | |
d392365e | 1327 | .B stonewall "\fR,\fP wait_for_previous" |
5982a925 | 1328 | Wait for preceding jobs in the job file to exit before starting this one. |
d60e92d1 AC |
1329 | \fBstonewall\fR implies \fBnew_group\fR. |
1330 | .TP | |
1331 | .B new_group | |
1332 | Start a new reporting group. If not given, all jobs in a file will be part | |
1333 | of the same reporting group, unless separated by a stonewall. | |
1334 | .TP | |
1335 | .BI numjobs \fR=\fPint | |
ff6bb260 | 1336 | Number of clones (processes/threads performing the same workload) of this job. |
d60e92d1 AC |
1337 | Default: 1. |
1338 | .TP | |
1339 | .B group_reporting | |
1340 | If set, display per-group reports instead of per-job when \fBnumjobs\fR is | |
1341 | specified. | |
1342 | .TP | |
1343 | .B thread | |
1344 | Use threads created with \fBpthread_create\fR\|(3) instead of processes created | |
1345 | with \fBfork\fR\|(2). | |
1346 | .TP | |
f7fa2653 | 1347 | .BI zonesize \fR=\fPint |
d60e92d1 AC |
1348 | Divide file into zones of the specified size in bytes. See \fBzoneskip\fR. |
1349 | .TP | |
fa769d44 SW |
1350 | .BI zonerange \fR=\fPint |
1351 | Give size of an IO zone. See \fBzoneskip\fR. | |
1352 | .TP | |
f7fa2653 | 1353 | .BI zoneskip \fR=\fPint |
d1429b5c | 1354 | Skip the specified number of bytes when \fBzonesize\fR bytes of data have been |
d60e92d1 AC |
1355 | read. |
1356 | .TP | |
1357 | .BI write_iolog \fR=\fPstr | |
5b42a488 SH |
1358 | Write the issued I/O patterns to the specified file. Specify a separate file |
1359 | for each job, otherwise the iologs will be interspersed and the file may be | |
1360 | corrupt. | |
d60e92d1 AC |
1361 | .TP |
1362 | .BI read_iolog \fR=\fPstr | |
1363 | Replay the I/O patterns contained in the specified file generated by | |
1364 | \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file. | |
1365 | .TP | |
64bbb865 DN |
1366 | .BI replay_no_stall \fR=\fPint |
1367 | While replaying I/O patterns using \fBread_iolog\fR the default behavior | |
1368 | attempts to respect timing information between I/Os. Enabling | |
1369 | \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while | |
1370 | still respecting ordering. | |
1371 | .TP | |
d1c46c04 DN |
1372 | .BI replay_redirect \fR=\fPstr |
1373 | While replaying I/O patterns using \fBread_iolog\fR the default behavior | |
1374 | is to replay the IOPS onto the major/minor device that each IOP was recorded | |
1375 | from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the | |
1376 | single specified device regardless of the device it was recorded from. | |
1377 | .TP | |
0c63576e JA |
1378 | .BI replay_align \fR=\fPint |
1379 | Force alignment of IO offsets and lengths in a trace to this power of 2 value. | |
1380 | .TP | |
1381 | .BI replay_scale \fR=\fPint | |
1382 | Scale sector offsets down by this factor when replaying traces. | |
1383 | .TP | |
3a5db920 JA |
1384 | .BI per_job_logs \fR=\fPbool |
1385 | If set, this generates bw/clat/iops log with per file private filenames. If | |
1386 | not set, jobs with identical names will share the log filename. Default: true. | |
1387 | .TP | |
836bad52 | 1388 | .BI write_bw_log \fR=\fPstr |
901bb994 JA |
1389 | If given, write a bandwidth log of the jobs in this job file. Can be used to |
1390 | store data of the bandwidth of the jobs in their lifetime. The included | |
1391 | fio_generate_plots script uses gnuplot to turn these text files into nice | |
26b26fca | 1392 | graphs. See \fBwrite_lat_log\fR for behaviour of given filename. For this |
8ad3b3dd | 1393 | option, the postfix is _bw.x.log, where x is the index of the job (1..N, |
3a5db920 JA |
1394 | where N is the number of jobs). If \fBper_job_logs\fR is false, then the |
1395 | filename will not include the job index. | |
d60e92d1 | 1396 | .TP |
836bad52 | 1397 | .BI write_lat_log \fR=\fPstr |
901bb994 | 1398 | Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no |
8ad3b3dd JA |
1399 | filename is given with this option, the default filename of |
1400 | "jobname_type.x.log" is used, where x is the index of the job (1..N, where | |
1401 | N is the number of jobs). Even if the filename is given, fio will still | |
3a5db920 JA |
1402 | append the type of log. If \fBper_job_logs\fR is false, then the filename will |
1403 | not include the job index. | |
901bb994 | 1404 | .TP |
c8eeb9df JA |
1405 | .BI write_iops_log \fR=\fPstr |
1406 | Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this | |
8ad3b3dd JA |
1407 | option, the default filename of "jobname_type.x.log" is used, where x is the |
1408 | index of the job (1..N, where N is the number of jobs). Even if the filename | |
3a5db920 JA |
1409 | is given, fio will still append the type of log. If \fBper_job_logs\fR is false, |
1410 | then the filename will not include the job index. | |
c8eeb9df | 1411 | .TP |
b8bc8cba JA |
1412 | .BI log_avg_msec \fR=\fPint |
1413 | By default, fio will log an entry in the iops, latency, or bw log for every | |
1414 | IO that completes. When writing to the disk log, that can quickly grow to a | |
1415 | very large size. Setting this option makes fio average the each log entry | |
1416 | over the specified period of time, reducing the resolution of the log. | |
1417 | Defaults to 0. | |
1418 | .TP | |
ae588852 JA |
1419 | .BI log_offset \fR=\fPbool |
1420 | If this is set, the iolog options will include the byte offset for the IO | |
1421 | entry as well as the other data values. | |
1422 | .TP | |
aee2ab67 JA |
1423 | .BI log_compression \fR=\fPint |
1424 | If this is set, fio will compress the IO logs as it goes, to keep the memory | |
1425 | footprint lower. When a log reaches the specified size, that chunk is removed | |
1426 | and compressed in the background. Given that IO logs are fairly highly | |
1427 | compressible, this yields a nice memory savings for longer runs. The downside | |
1428 | is that the compression will consume some background CPU cycles, so it may | |
1429 | impact the run. This, however, is also true if the logging ends up consuming | |
1430 | most of the system memory. So pick your poison. The IO logs are saved | |
1431 | normally at the end of a run, by decompressing the chunks and storing them | |
1432 | in the specified log file. This feature depends on the availability of zlib. | |
1433 | .TP | |
c08f9fe2 JA |
1434 | .BI log_compression_cpus \fR=\fPstr |
1435 | Define the set of CPUs that are allowed to handle online log compression | |
1436 | for the IO jobs. This can provide better isolation between performance | |
1437 | sensitive jobs, and background compression work. | |
1438 | .TP | |
b26317c9 | 1439 | .BI log_store_compressed \fR=\fPbool |
c08f9fe2 JA |
1440 | If set, fio will store the log files in a compressed format. They can be |
1441 | decompressed with fio, using the \fB\-\-inflate-log\fR command line parameter. | |
1442 | The files will be stored with a \fB\.fz\fR suffix. | |
b26317c9 | 1443 | .TP |
66347cfa DE |
1444 | .BI block_error_percentiles \fR=\fPbool |
1445 | If set, record errors in trim block-sized units from writes and trims and output | |
1446 | a histogram of how many trims it took to get to errors, and what kind of error | |
1447 | was encountered. | |
1448 | .TP | |
836bad52 | 1449 | .BI disable_lat \fR=\fPbool |
02af0988 | 1450 | Disable measurements of total latency numbers. Useful only for cutting |
ccc2b328 | 1451 | back the number of calls to \fBgettimeofday\fR\|(2), as that does impact performance at |
901bb994 JA |
1452 | really high IOPS rates. Note that to really get rid of a large amount of these |
1453 | calls, this option must be used with disable_slat and disable_bw as well. | |
1454 | .TP | |
836bad52 | 1455 | .BI disable_clat \fR=\fPbool |
c95f9daf | 1456 | Disable measurements of completion latency numbers. See \fBdisable_lat\fR. |
02af0988 | 1457 | .TP |
836bad52 | 1458 | .BI disable_slat \fR=\fPbool |
02af0988 | 1459 | Disable measurements of submission latency numbers. See \fBdisable_lat\fR. |
901bb994 | 1460 | .TP |
836bad52 | 1461 | .BI disable_bw_measurement \fR=\fPbool |
02af0988 | 1462 | Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR. |
d60e92d1 | 1463 | .TP |
f7fa2653 | 1464 | .BI lockmem \fR=\fPint |
d60e92d1 | 1465 | Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to |
81c6b6cd | 1466 | simulate a smaller amount of memory. The amount specified is per worker. |
d60e92d1 AC |
1467 | .TP |
1468 | .BI exec_prerun \fR=\fPstr | |
1469 | Before running the job, execute the specified command with \fBsystem\fR\|(3). | |
ce486495 EV |
1470 | .RS |
1471 | Output is redirected in a file called \fBjobname.prerun.txt\fR | |
1472 | .RE | |
d60e92d1 AC |
1473 | .TP |
1474 | .BI exec_postrun \fR=\fPstr | |
1475 | Same as \fBexec_prerun\fR, but the command is executed after the job completes. | |
ce486495 EV |
1476 | .RS |
1477 | Output is redirected in a file called \fBjobname.postrun.txt\fR | |
1478 | .RE | |
d60e92d1 AC |
1479 | .TP |
1480 | .BI ioscheduler \fR=\fPstr | |
1481 | Attempt to switch the device hosting the file to the specified I/O scheduler. | |
1482 | .TP | |
d60e92d1 | 1483 | .BI disk_util \fR=\fPbool |
d1429b5c | 1484 | Generate disk utilization statistics if the platform supports it. Default: true. |
901bb994 | 1485 | .TP |
23893646 JA |
1486 | .BI clocksource \fR=\fPstr |
1487 | Use the given clocksource as the base of timing. The supported options are: | |
1488 | .RS | |
1489 | .TP | |
1490 | .B gettimeofday | |
ccc2b328 | 1491 | \fBgettimeofday\fR\|(2) |
23893646 JA |
1492 | .TP |
1493 | .B clock_gettime | |
ccc2b328 | 1494 | \fBclock_gettime\fR\|(2) |
23893646 JA |
1495 | .TP |
1496 | .B cpu | |
1497 | Internal CPU clock source | |
1498 | .TP | |
1499 | .RE | |
1500 | .P | |
1501 | \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast | |
1502 | (and fio is heavy on time calls). Fio will automatically use this clocksource | |
1503 | if it's supported and considered reliable on the system it is running on, | |
1504 | unless another clocksource is specifically set. For x86/x86-64 CPUs, this | |
1505 | means supporting TSC Invariant. | |
1506 | .TP | |
901bb994 | 1507 | .BI gtod_reduce \fR=\fPbool |
ccc2b328 | 1508 | Enable all of the \fBgettimeofday\fR\|(2) reducing options (disable_clat, disable_slat, |
901bb994 | 1509 | disable_bw) plus reduce precision of the timeout somewhat to really shrink the |
ccc2b328 | 1510 | \fBgettimeofday\fR\|(2) call count. With this option enabled, we only do about 0.4% of |
901bb994 JA |
1511 | the gtod() calls we would have done if all time keeping was enabled. |
1512 | .TP | |
1513 | .BI gtod_cpu \fR=\fPint | |
1514 | Sometimes it's cheaper to dedicate a single thread of execution to just getting | |
1515 | the current time. Fio (and databases, for instance) are very intensive on | |
ccc2b328 | 1516 | \fBgettimeofday\fR\|(2) calls. With this option, you can set one CPU aside for doing |
901bb994 JA |
1517 | nothing but logging current time to a shared memory location. Then the other |
1518 | threads/processes that run IO workloads need only copy that segment, instead of | |
ccc2b328 | 1519 | entering the kernel with a \fBgettimeofday\fR\|(2) call. The CPU set aside for doing |
901bb994 JA |
1520 | these time calls will be excluded from other uses. Fio will manually clear it |
1521 | from the CPU mask of other jobs. | |
f2bba182 | 1522 | .TP |
8b28bd41 DM |
1523 | .BI ignore_error \fR=\fPstr |
1524 | Sometimes you want to ignore some errors during test in that case you can specify | |
1525 | error list for each error type. | |
1526 | .br | |
1527 | ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST | |
1528 | .br | |
1529 | errors for given error type is separated with ':'. | |
1530 | Error may be symbol ('ENOSPC', 'ENOMEM') or an integer. | |
1531 | .br | |
1532 | Example: ignore_error=EAGAIN,ENOSPC:122 . | |
ff6bb260 SL |
1533 | .br |
1534 | This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE. | |
8b28bd41 DM |
1535 | .TP |
1536 | .BI error_dump \fR=\fPbool | |
1537 | If set dump every error even if it is non fatal, true by default. If disabled | |
1538 | only fatal error will be dumped | |
1539 | .TP | |
fa769d44 SW |
1540 | .BI profile \fR=\fPstr |
1541 | Select a specific builtin performance test. | |
1542 | .TP | |
a696fa2a JA |
1543 | .BI cgroup \fR=\fPstr |
1544 | Add job to this control group. If it doesn't exist, it will be created. | |
6adb38a1 JA |
1545 | The system must have a mounted cgroup blkio mount point for this to work. If |
1546 | your system doesn't have it mounted, you can do so with: | |
1547 | ||
5982a925 | 1548 | # mount \-t cgroup \-o blkio none /cgroup |
a696fa2a JA |
1549 | .TP |
1550 | .BI cgroup_weight \fR=\fPint | |
1551 | Set the weight of the cgroup to this value. See the documentation that comes | |
1552 | with the kernel, allowed values are in the range of 100..1000. | |
e0b0d892 | 1553 | .TP |
7de87099 VG |
1554 | .BI cgroup_nodelete \fR=\fPbool |
1555 | Normally fio will delete the cgroups it has created after the job completion. | |
1556 | To override this behavior and to leave cgroups around after the job completion, | |
1557 | set cgroup_nodelete=1. This can be useful if one wants to inspect various | |
1558 | cgroup files after job completion. Default: false | |
1559 | .TP | |
e0b0d892 JA |
1560 | .BI uid \fR=\fPint |
1561 | Instead of running as the invoking user, set the user ID to this value before | |
1562 | the thread/process does any work. | |
1563 | .TP | |
1564 | .BI gid \fR=\fPint | |
1565 | Set group ID, see \fBuid\fR. | |
83349190 | 1566 | .TP |
fa769d44 SW |
1567 | .BI unit_base \fR=\fPint |
1568 | Base unit for reporting. Allowed values are: | |
1569 | .RS | |
1570 | .TP | |
1571 | .B 0 | |
1572 | Use auto-detection (default). | |
1573 | .TP | |
1574 | .B 8 | |
1575 | Byte based. | |
1576 | .TP | |
1577 | .B 1 | |
1578 | Bit based. | |
1579 | .RE | |
1580 | .P | |
1581 | .TP | |
9e684a49 DE |
1582 | .BI flow_id \fR=\fPint |
1583 | The ID of the flow. If not specified, it defaults to being a global flow. See | |
1584 | \fBflow\fR. | |
1585 | .TP | |
1586 | .BI flow \fR=\fPint | |
1587 | Weight in token-based flow control. If this value is used, then there is a | |
1588 | \fBflow counter\fR which is used to regulate the proportion of activity between | |
1589 | two or more jobs. fio attempts to keep this flow counter near zero. The | |
1590 | \fBflow\fR parameter stands for how much should be added or subtracted to the | |
1591 | flow counter on each iteration of the main I/O loop. That is, if one job has | |
1592 | \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly | |
1593 | 1:8 ratio in how much one runs vs the other. | |
1594 | .TP | |
1595 | .BI flow_watermark \fR=\fPint | |
1596 | The maximum value that the absolute value of the flow counter is allowed to | |
1597 | reach before the job must wait for a lower value of the counter. | |
1598 | .TP | |
1599 | .BI flow_sleep \fR=\fPint | |
1600 | The period of time, in microseconds, to wait after the flow watermark has been | |
1601 | exceeded before retrying operations | |
1602 | .TP | |
83349190 YH |
1603 | .BI clat_percentiles \fR=\fPbool |
1604 | Enable the reporting of percentiles of completion latencies. | |
1605 | .TP | |
1606 | .BI percentile_list \fR=\fPfloat_list | |
66347cfa DE |
1607 | Overwrite the default list of percentiles for completion latencies and the |
1608 | block error histogram. Each number is a floating number in the range (0,100], | |
1609 | and the maximum length of the list is 20. Use ':' to separate the | |
3eb07285 | 1610 | numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to |
83349190 YH |
1611 | report the values of completion latency below which 99.5% and 99.9% of |
1612 | the observed latencies fell, respectively. | |
de890a1e SL |
1613 | .SS "Ioengine Parameters List" |
1614 | Some parameters are only valid when a specific ioengine is in use. These are | |
1615 | used identically to normal parameters, with the caveat that when used on the | |
cf145d90 | 1616 | command line, they must come after the ioengine. |
de890a1e | 1617 | .TP |
e4585935 JA |
1618 | .BI (cpu)cpuload \fR=\fPint |
1619 | Attempt to use the specified percentage of CPU cycles. | |
1620 | .TP | |
1621 | .BI (cpu)cpuchunks \fR=\fPint | |
1622 | Split the load into cycles of the given time. In microseconds. | |
1623 | .TP | |
046395d7 JA |
1624 | .BI (cpu)exit_on_io_done \fR=\fPbool |
1625 | Detect when IO threads are done, then exit. | |
1626 | .TP | |
de890a1e SL |
1627 | .BI (libaio)userspace_reap |
1628 | Normally, with the libaio engine in use, fio will use | |
1629 | the io_getevents system call to reap newly returned events. | |
1630 | With this flag turned on, the AIO ring will be read directly | |
1631 | from user-space to reap events. The reaping mode is only | |
1632 | enabled when polling for a minimum of 0 events (eg when | |
1633 | iodepth_batch_complete=0). | |
1634 | .TP | |
1635 | .BI (net,netsplice)hostname \fR=\fPstr | |
1636 | The host name or IP address to use for TCP or UDP based IO. | |
1637 | If the job is a TCP listener or UDP reader, the hostname is not | |
b511c9aa | 1638 | used and must be omitted unless it is a valid UDP multicast address. |
de890a1e SL |
1639 | .TP |
1640 | .BI (net,netsplice)port \fR=\fPint | |
6315af9d JA |
1641 | The TCP or UDP port to bind to or connect to. If this is used with |
1642 | \fBnumjobs\fR to spawn multiple instances of the same job type, then | |
1643 | this will be the starting port number since fio will use a range of ports. | |
de890a1e | 1644 | .TP |
b93b6a2e SB |
1645 | .BI (net,netsplice)interface \fR=\fPstr |
1646 | The IP address of the network interface used to send or receive UDP multicast | |
1647 | packets. | |
1648 | .TP | |
d3a623de SB |
1649 | .BI (net,netsplice)ttl \fR=\fPint |
1650 | Time-to-live value for outgoing UDP multicast packets. Default: 1 | |
1651 | .TP | |
1d360ffb JA |
1652 | .BI (net,netsplice)nodelay \fR=\fPbool |
1653 | Set TCP_NODELAY on TCP connections. | |
1654 | .TP | |
de890a1e SL |
1655 | .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr |
1656 | The network protocol to use. Accepted values are: | |
1657 | .RS | |
1658 | .RS | |
1659 | .TP | |
1660 | .B tcp | |
1661 | Transmission control protocol | |
1662 | .TP | |
49ccb8c1 JA |
1663 | .B tcpv6 |
1664 | Transmission control protocol V6 | |
1665 | .TP | |
de890a1e | 1666 | .B udp |
f5cc3d0e | 1667 | User datagram protocol |
de890a1e | 1668 | .TP |
49ccb8c1 JA |
1669 | .B udpv6 |
1670 | User datagram protocol V6 | |
1671 | .TP | |
de890a1e SL |
1672 | .B unix |
1673 | UNIX domain socket | |
1674 | .RE | |
1675 | .P | |
1676 | When the protocol is TCP or UDP, the port must also be given, | |
1677 | as well as the hostname if the job is a TCP listener or UDP | |
1678 | reader. For unix sockets, the normal filename option should be | |
1679 | used and the port is invalid. | |
1680 | .RE | |
1681 | .TP | |
1682 | .BI (net,netsplice)listen | |
1683 | For TCP network connections, tell fio to listen for incoming | |
1684 | connections rather than initiating an outgoing connection. The | |
1685 | hostname must be omitted if this option is used. | |
d54fce84 | 1686 | .TP |
7aeb1e94 | 1687 | .BI (net, pingpong) \fR=\fPbool |
cecbfd47 | 1688 | Normally a network writer will just continue writing data, and a network reader |
cf145d90 | 1689 | will just consume packets. If pingpong=1 is set, a writer will send its normal |
7aeb1e94 JA |
1690 | payload to the reader, then wait for the reader to send the same payload back. |
1691 | This allows fio to measure network latencies. The submission and completion | |
1692 | latencies then measure local time spent sending or receiving, and the | |
1693 | completion latency measures how long it took for the other end to receive and | |
b511c9aa SB |
1694 | send back. For UDP multicast traffic pingpong=1 should only be set for a single |
1695 | reader when multiple readers are listening to the same address. | |
7aeb1e94 | 1696 | .TP |
1008602c JA |
1697 | .BI (net, window_size) \fR=\fPint |
1698 | Set the desired socket buffer size for the connection. | |
1699 | .TP | |
e5f34d95 JA |
1700 | .BI (net, mss) \fR=\fPint |
1701 | Set the TCP maximum segment size (TCP_MAXSEG). | |
1702 | .TP | |
d54fce84 DM |
1703 | .BI (e4defrag,donorname) \fR=\fPstr |
1704 | File will be used as a block donor (swap extents between files) | |
1705 | .TP | |
1706 | .BI (e4defrag,inplace) \fR=\fPint | |
ff6bb260 | 1707 | Configure donor file block allocation strategy |
d54fce84 DM |
1708 | .RS |
1709 | .BI 0(default) : | |
1710 | Preallocate donor's file on init | |
1711 | .TP | |
1712 | .BI 1: | |
cecbfd47 | 1713 | allocate space immediately inside defragment event, and free right after event |
d54fce84 | 1714 | .RE |
0d978694 DAG |
1715 | .TP |
1716 | .BI (rbd)rbdname \fR=\fPstr | |
1717 | Specifies the name of the RBD. | |
1718 | .TP | |
1719 | .BI (rbd)pool \fR=\fPstr | |
1720 | Specifies the name of the Ceph pool containing the RBD. | |
1721 | .TP | |
1722 | .BI (rbd)clientname \fR=\fPstr | |
1723 | Specifies the username (without the 'client.' prefix) used to access the Ceph cluster. | |
65fa28ca DE |
1724 | .TP |
1725 | .BI (mtd)skipbad \fR=\fPbool | |
1726 | Skip operations against known bad blocks. | |
d60e92d1 | 1727 | .SH OUTPUT |
d1429b5c AC |
1728 | While running, \fBfio\fR will display the status of the created jobs. For |
1729 | example: | |
d60e92d1 | 1730 | .RS |
d1429b5c | 1731 | .P |
d60e92d1 AC |
1732 | Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s] |
1733 | .RE | |
1734 | .P | |
d1429b5c AC |
1735 | The characters in the first set of brackets denote the current status of each |
1736 | threads. The possible values are: | |
1737 | .P | |
1738 | .PD 0 | |
d60e92d1 AC |
1739 | .RS |
1740 | .TP | |
1741 | .B P | |
1742 | Setup but not started. | |
1743 | .TP | |
1744 | .B C | |
1745 | Thread created. | |
1746 | .TP | |
1747 | .B I | |
1748 | Initialized, waiting. | |
1749 | .TP | |
1750 | .B R | |
1751 | Running, doing sequential reads. | |
1752 | .TP | |
1753 | .B r | |
1754 | Running, doing random reads. | |
1755 | .TP | |
1756 | .B W | |
1757 | Running, doing sequential writes. | |
1758 | .TP | |
1759 | .B w | |
1760 | Running, doing random writes. | |
1761 | .TP | |
1762 | .B M | |
1763 | Running, doing mixed sequential reads/writes. | |
1764 | .TP | |
1765 | .B m | |
1766 | Running, doing mixed random reads/writes. | |
1767 | .TP | |
1768 | .B F | |
1769 | Running, currently waiting for \fBfsync\fR\|(2). | |
1770 | .TP | |
1771 | .B V | |
1772 | Running, verifying written data. | |
1773 | .TP | |
1774 | .B E | |
1775 | Exited, not reaped by main thread. | |
1776 | .TP | |
1777 | .B \- | |
1778 | Exited, thread reaped. | |
1779 | .RE | |
d1429b5c | 1780 | .PD |
d60e92d1 AC |
1781 | .P |
1782 | The second set of brackets shows the estimated completion percentage of | |
1783 | the current group. The third set shows the read and write I/O rate, | |
1784 | respectively. Finally, the estimated run time of the job is displayed. | |
1785 | .P | |
1786 | When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data | |
1787 | for each thread, each group of threads, and each disk, in that order. | |
1788 | .P | |
1789 | Per-thread statistics first show the threads client number, group-id, and | |
1790 | error code. The remaining figures are as follows: | |
1791 | .RS | |
d60e92d1 AC |
1792 | .TP |
1793 | .B io | |
1794 | Number of megabytes of I/O performed. | |
1795 | .TP | |
1796 | .B bw | |
1797 | Average data rate (bandwidth). | |
1798 | .TP | |
1799 | .B runt | |
1800 | Threads run time. | |
1801 | .TP | |
1802 | .B slat | |
1803 | Submission latency minimum, maximum, average and standard deviation. This is | |
1804 | the time it took to submit the I/O. | |
1805 | .TP | |
1806 | .B clat | |
1807 | Completion latency minimum, maximum, average and standard deviation. This | |
1808 | is the time between submission and completion. | |
1809 | .TP | |
1810 | .B bw | |
1811 | Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average | |
1812 | and standard deviation. | |
1813 | .TP | |
1814 | .B cpu | |
1815 | CPU usage statistics. Includes user and system time, number of context switches | |
1816 | this thread went through and number of major and minor page faults. | |
1817 | .TP | |
1818 | .B IO depths | |
1819 | Distribution of I/O depths. Each depth includes everything less than (or equal) | |
1820 | to it, but greater than the previous depth. | |
1821 | .TP | |
1822 | .B IO issued | |
1823 | Number of read/write requests issued, and number of short read/write requests. | |
1824 | .TP | |
1825 | .B IO latencies | |
1826 | Distribution of I/O completion latencies. The numbers follow the same pattern | |
1827 | as \fBIO depths\fR. | |
1828 | .RE | |
d60e92d1 AC |
1829 | .P |
1830 | The group statistics show: | |
d1429b5c | 1831 | .PD 0 |
d60e92d1 AC |
1832 | .RS |
1833 | .TP | |
1834 | .B io | |
1835 | Number of megabytes I/O performed. | |
1836 | .TP | |
1837 | .B aggrb | |
1838 | Aggregate bandwidth of threads in the group. | |
1839 | .TP | |
1840 | .B minb | |
1841 | Minimum average bandwidth a thread saw. | |
1842 | .TP | |
1843 | .B maxb | |
1844 | Maximum average bandwidth a thread saw. | |
1845 | .TP | |
1846 | .B mint | |
d1429b5c | 1847 | Shortest runtime of threads in the group. |
d60e92d1 AC |
1848 | .TP |
1849 | .B maxt | |
1850 | Longest runtime of threads in the group. | |
1851 | .RE | |
d1429b5c | 1852 | .PD |
d60e92d1 AC |
1853 | .P |
1854 | Finally, disk statistics are printed with reads first: | |
d1429b5c | 1855 | .PD 0 |
d60e92d1 AC |
1856 | .RS |
1857 | .TP | |
1858 | .B ios | |
1859 | Number of I/Os performed by all groups. | |
1860 | .TP | |
1861 | .B merge | |
1862 | Number of merges in the I/O scheduler. | |
1863 | .TP | |
1864 | .B ticks | |
1865 | Number of ticks we kept the disk busy. | |
1866 | .TP | |
1867 | .B io_queue | |
1868 | Total time spent in the disk queue. | |
1869 | .TP | |
1870 | .B util | |
1871 | Disk utilization. | |
1872 | .RE | |
d1429b5c | 1873 | .PD |
8423bd11 JA |
1874 | .P |
1875 | It is also possible to get fio to dump the current output while it is | |
1876 | running, without terminating the job. To do that, send fio the \fBUSR1\fR | |
1877 | signal. | |
d60e92d1 | 1878 | .SH TERSE OUTPUT |
2b8c71b0 CE |
1879 | If the \fB\-\-minimal\fR / \fB\-\-append-terse\fR options are given, the |
1880 | results will be printed/appended in a semicolon-delimited format suitable for | |
1881 | scripted use. | |
1882 | A job description (if provided) follows on a new line. Note that the first | |
525c2bfa JA |
1883 | number in the line is the version number. If the output has to be changed |
1884 | for some reason, this number will be incremented by 1 to signify that | |
1885 | change. The fields are: | |
d60e92d1 AC |
1886 | .P |
1887 | .RS | |
5e726d0a | 1888 | .B terse version, fio version, jobname, groupid, error |
d60e92d1 AC |
1889 | .P |
1890 | Read status: | |
1891 | .RS | |
312b4af2 | 1892 | .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP |
d60e92d1 AC |
1893 | .P |
1894 | Submission latency: | |
1895 | .RS | |
1896 | .B min, max, mean, standard deviation | |
1897 | .RE | |
1898 | Completion latency: | |
1899 | .RS | |
1900 | .B min, max, mean, standard deviation | |
1901 | .RE | |
1db92cb6 JA |
1902 | Completion latency percentiles (20 fields): |
1903 | .RS | |
1904 | .B Xth percentile=usec | |
1905 | .RE | |
525c2bfa JA |
1906 | Total latency: |
1907 | .RS | |
1908 | .B min, max, mean, standard deviation | |
1909 | .RE | |
d60e92d1 AC |
1910 | Bandwidth: |
1911 | .RS | |
1912 | .B min, max, aggregate percentage of total, mean, standard deviation | |
1913 | .RE | |
1914 | .RE | |
1915 | .P | |
1916 | Write status: | |
1917 | .RS | |
312b4af2 | 1918 | .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP |
d60e92d1 AC |
1919 | .P |
1920 | Submission latency: | |
1921 | .RS | |
1922 | .B min, max, mean, standard deviation | |
1923 | .RE | |
1924 | Completion latency: | |
1925 | .RS | |
1926 | .B min, max, mean, standard deviation | |
1927 | .RE | |
1db92cb6 JA |
1928 | Completion latency percentiles (20 fields): |
1929 | .RS | |
1930 | .B Xth percentile=usec | |
1931 | .RE | |
525c2bfa JA |
1932 | Total latency: |
1933 | .RS | |
1934 | .B min, max, mean, standard deviation | |
1935 | .RE | |
d60e92d1 AC |
1936 | Bandwidth: |
1937 | .RS | |
1938 | .B min, max, aggregate percentage of total, mean, standard deviation | |
1939 | .RE | |
1940 | .RE | |
1941 | .P | |
d1429b5c | 1942 | CPU usage: |
d60e92d1 | 1943 | .RS |
bd2626f0 | 1944 | .B user, system, context switches, major page faults, minor page faults |
d60e92d1 AC |
1945 | .RE |
1946 | .P | |
1947 | IO depth distribution: | |
1948 | .RS | |
1949 | .B <=1, 2, 4, 8, 16, 32, >=64 | |
1950 | .RE | |
1951 | .P | |
562c2d2f | 1952 | IO latency distribution: |
d60e92d1 | 1953 | .RS |
562c2d2f DN |
1954 | Microseconds: |
1955 | .RS | |
1956 | .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000 | |
1957 | .RE | |
1958 | Milliseconds: | |
1959 | .RS | |
1960 | .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000 | |
1961 | .RE | |
1962 | .RE | |
1963 | .P | |
f2f788dd JA |
1964 | Disk utilization (1 for each disk used): |
1965 | .RS | |
1966 | .B name, read ios, write ios, read merges, write merges, read ticks, write ticks, read in-queue time, write in-queue time, disk utilization percentage | |
1967 | .RE | |
1968 | .P | |
5982a925 | 1969 | Error Info (dependent on continue_on_error, default off): |
562c2d2f | 1970 | .RS |
ff6bb260 | 1971 | .B total # errors, first error code |
d60e92d1 AC |
1972 | .RE |
1973 | .P | |
562c2d2f | 1974 | .B text description (if provided in config - appears on newline) |
d60e92d1 | 1975 | .RE |
49da1240 JA |
1976 | .SH CLIENT / SERVER |
1977 | Normally you would run fio as a stand-alone application on the machine | |
1978 | where the IO workload should be generated. However, it is also possible to | |
1979 | run the frontend and backend of fio separately. This makes it possible to | |
1980 | have a fio server running on the machine(s) where the IO workload should | |
1981 | be running, while controlling it from another machine. | |
1982 | ||
1983 | To start the server, you would do: | |
1984 | ||
1985 | \fBfio \-\-server=args\fR | |
1986 | ||
1987 | on that machine, where args defines what fio listens to. The arguments | |
811826be | 1988 | are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4) |
20c67f10 MS |
1989 | for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain |
1990 | socket. 'hostname' is either a hostname or IP address, and 'port' is the port to | |
811826be | 1991 | listen to (only valid for TCP/IP, not a local socket). Some examples: |
49da1240 | 1992 | |
e01e9745 | 1993 | 1) fio \-\-server |
49da1240 JA |
1994 | |
1995 | Start a fio server, listening on all interfaces on the default port (8765). | |
1996 | ||
e01e9745 | 1997 | 2) fio \-\-server=ip:hostname,4444 |
49da1240 JA |
1998 | |
1999 | Start a fio server, listening on IP belonging to hostname and on port 4444. | |
2000 | ||
e01e9745 | 2001 | 3) fio \-\-server=ip6:::1,4444 |
811826be JA |
2002 | |
2003 | Start a fio server, listening on IPv6 localhost ::1 and on port 4444. | |
2004 | ||
e01e9745 | 2005 | 4) fio \-\-server=,4444 |
49da1240 JA |
2006 | |
2007 | Start a fio server, listening on all interfaces on port 4444. | |
2008 | ||
e01e9745 | 2009 | 5) fio \-\-server=1.2.3.4 |
49da1240 JA |
2010 | |
2011 | Start a fio server, listening on IP 1.2.3.4 on the default port. | |
2012 | ||
e01e9745 | 2013 | 6) fio \-\-server=sock:/tmp/fio.sock |
49da1240 JA |
2014 | |
2015 | Start a fio server, listening on the local socket /tmp/fio.sock. | |
2016 | ||
2017 | When a server is running, you can connect to it from a client. The client | |
2018 | is run with: | |
2019 | ||
e01e9745 | 2020 | fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)> |
49da1240 | 2021 | |
e01e9745 MS |
2022 | where \-\-local-args are arguments that are local to the client where it is |
2023 | running, 'server' is the connect string, and \-\-remote-args and <job file(s)> | |
49da1240 JA |
2024 | are sent to the server. The 'server' string follows the same format as it |
2025 | does on the server side, to allow IP/hostname/socket and port strings. | |
2026 | You can connect to multiple clients as well, to do that you could run: | |
2027 | ||
e01e9745 | 2028 | fio \-\-client=server2 \-\-client=server2 <job file(s)> |
323255cc JA |
2029 | |
2030 | If the job file is located on the fio server, then you can tell the server | |
2031 | to load a local file as well. This is done by using \-\-remote-config: | |
2032 | ||
2033 | fio \-\-client=server \-\-remote-config /path/to/file.fio | |
2034 | ||
39b5f61e | 2035 | Then fio will open this local (to the server) job file instead |
323255cc | 2036 | of being passed one from the client. |
39b5f61e | 2037 | |
ff6bb260 | 2038 | If you have many servers (example: 100 VMs/containers), you can input a pathname |
39b5f61e BE |
2039 | of a file containing host IPs/names as the parameter value for the \-\-client option. |
2040 | For example, here is an example "host.list" file containing 2 hostnames: | |
2041 | ||
2042 | host1.your.dns.domain | |
2043 | .br | |
2044 | host2.your.dns.domain | |
2045 | ||
2046 | The fio command would then be: | |
2047 | ||
2048 | fio \-\-client=host.list <job file> | |
2049 | ||
2050 | In this mode, you cannot input server-specific parameters or job files, and all | |
2051 | servers receive the same job file. | |
2052 | ||
2053 | In order to enable fio \-\-client runs utilizing a shared filesystem from multiple hosts, | |
ff6bb260 SL |
2054 | fio \-\-client now prepends the IP address of the server to the filename. For example, |
2055 | if fio is using directory /mnt/nfs/fio and is writing filename fileio.tmp, | |
39b5f61e BE |
2056 | with a \-\-client hostfile |
2057 | containing two hostnames h1 and h2 with IP addresses 192.168.10.120 and 192.168.10.121, then | |
2058 | fio will create two files: | |
2059 | ||
2060 | /mnt/nfs/fio/192.168.10.120.fileio.tmp | |
2061 | .br | |
2062 | /mnt/nfs/fio/192.168.10.121.fileio.tmp | |
2063 | ||
d60e92d1 | 2064 | .SH AUTHORS |
49da1240 | 2065 | |
d60e92d1 | 2066 | .B fio |
aa58d252 | 2067 | was written by Jens Axboe <jens.axboe@oracle.com>, |
f8b8f7da | 2068 | now Jens Axboe <axboe@fb.com>. |
d1429b5c AC |
2069 | .br |
2070 | This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based | |
d60e92d1 AC |
2071 | on documentation by Jens Axboe. |
2072 | .SH "REPORTING BUGS" | |
482900c9 | 2073 | Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>. |
d1429b5c | 2074 | See \fBREADME\fR. |
d60e92d1 | 2075 | .SH "SEE ALSO" |
d1429b5c AC |
2076 | For further documentation see \fBHOWTO\fR and \fBREADME\fR. |
2077 | .br | |
2078 | Sample jobfiles are available in the \fBexamples\fR directory. |