Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1 | .TH fio 1 "September 2007" "User Manual" |
| 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 |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 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 |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 20 | .BI \-\-output \fR=\fPfilename |
| 21 | Write output to \fIfilename\fR. |
| 22 | .TP |
liang xie | b2cecdc | 2012-08-31 08:22:42 -0700 | [diff] [blame] | 23 | .BI \-\-runtime \fR=\fPruntime |
| 24 | Limit run time to \fIruntime\fR seconds. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 25 | .TP |
| 26 | .B \-\-latency\-log |
| 27 | Generate per-job latency logs. |
| 28 | .TP |
| 29 | .B \-\-bandwidth\-log |
| 30 | Generate per-job bandwidth logs. |
| 31 | .TP |
| 32 | .B \-\-minimal |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 33 | Print statistics in a terse, semicolon-delimited format. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 34 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 35 | .B \-\-version |
| 36 | Display version information and exit. |
| 37 | .TP |
Jens Axboe | 065248b | 2011-10-13 20:51:05 +0200 | [diff] [blame] | 38 | .BI \-\-terse\-version \fR=\fPversion |
Jens Axboe | 4d65865 | 2011-10-17 15:05:47 +0200 | [diff] [blame] | 39 | Set terse version output format (Current version 3, or older version 2). |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 40 | .TP |
| 41 | .B \-\-help |
| 42 | Display usage information and exit. |
| 43 | .TP |
| 44 | .BI \-\-cmdhelp \fR=\fPcommand |
| 45 | Print help information for \fIcommand\fR. May be `all' for all commands. |
| 46 | .TP |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 47 | .BI \-\-enghelp \fR=\fPioengine[,command] |
| 48 | List all commands defined by \fIioengine\fR, or print help for \fIcommand\fR defined by \fIioengine\fR. |
| 49 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 50 | .BI \-\-showcmd \fR=\fPjobfile |
| 51 | Convert \fIjobfile\fR to a set of command-line options. |
| 52 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 53 | .BI \-\-eta \fR=\fPwhen |
| 54 | Specifies when real-time ETA estimate should be printed. \fIwhen\fR may |
| 55 | be one of `always', `never' or `auto'. |
| 56 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 57 | .BI \-\-readonly |
| 58 | Turn on safety read-only checks, preventing any attempted write. |
| 59 | .TP |
Aaron Carroll | c0a5d35 | 2008-02-26 23:10:39 +0100 | [diff] [blame] | 60 | .BI \-\-section \fR=\fPsec |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 61 | Only run section \fIsec\fR from job file. Multiple of these options can be given, adding more sections to run. |
Aaron Carroll | c0a5d35 | 2008-02-26 23:10:39 +0100 | [diff] [blame] | 62 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 63 | .BI \-\-alloc\-size \fR=\fPkb |
| 64 | Set the internal smalloc pool size to \fIkb\fP kilobytes. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 65 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 66 | .BI \-\-warnings\-fatal |
| 67 | All fio parser warnings are fatal, causing fio to exit with an error. |
Jens Axboe | 9183788 | 2008-02-05 12:02:07 +0100 | [diff] [blame] | 68 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 69 | .BI \-\-max\-jobs \fR=\fPnr |
Martin Steigerwald | 57e118a | 2012-05-07 17:06:13 +0200 | [diff] [blame] | 70 | Set the maximum allowed number of jobs (threads/processes) to support. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 71 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 72 | .BI \-\-server \fR=\fPargs |
| 73 | Start a backend server, with \fIargs\fP specifying what to listen to. See client/server section. |
Jens Axboe | f57a9c5 | 2011-09-09 21:01:37 +0200 | [diff] [blame] | 74 | .TP |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 75 | .BI \-\-daemonize \fR=\fPpidfile |
| 76 | Background a fio server, writing the pid to the given pid file. |
| 77 | .TP |
| 78 | .BI \-\-client \fR=\fPhost |
| 79 | Instead of running the jobs locally, send and run them on the given host. |
Huadong Liu | f2a2ce0 | 2013-01-30 13:22:24 +0100 | [diff] [blame] | 80 | .TP |
| 81 | .BI \-\-idle\-prof \fR=\fPoption |
| 82 | Report cpu idleness on a system or percpu basis (\fIoption\fP=system,percpu) or run unit work calibration only (\fIoption\fP=calibrate). |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 83 | .SH "JOB FILE FORMAT" |
| 84 | Job files are in `ini' format. They consist of one or more |
| 85 | job definitions, which begin with a job name in square brackets and |
| 86 | extend to the next job name. The job name can be any ASCII string |
| 87 | except `global', which has a special meaning. Following the job name is |
| 88 | a sequence of zero or more parameters, one per line, that define the |
| 89 | behavior of the job. Any line starting with a `;' or `#' character is |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 90 | considered a comment and ignored. |
Aaron Carroll | d9956b6 | 2007-11-16 12:12:45 +0100 | [diff] [blame] | 91 | .P |
| 92 | If \fIjobfile\fR is specified as `-', the job file will be read from |
| 93 | standard input. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 94 | .SS "Global Section" |
| 95 | The global section contains default parameters for jobs specified in the |
| 96 | job file. A job is only affected by global sections residing above it, |
| 97 | and there may be any number of global sections. Specific job definitions |
| 98 | may override any parameter set in global sections. |
| 99 | .SH "JOB PARAMETERS" |
| 100 | .SS Types |
| 101 | Some parameters may take arguments of a specific type. The types used are: |
| 102 | .TP |
| 103 | .I str |
| 104 | String: a sequence of alphanumeric characters. |
| 105 | .TP |
| 106 | .I int |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 107 | SI integer: a whole number, possibly containing a suffix denoting the base unit |
Jens Axboe | b09da8f | 2009-07-17 23:16:17 +0200 | [diff] [blame] | 108 | of the value. Accepted suffixes are `k', 'M', 'G', 'T', and 'P', denoting |
| 109 | kilo (1024), mega (1024^2), giga (1024^3), tera (1024^4), and peta (1024^5) |
| 110 | respectively. The suffix is not case sensitive. If prefixed with '0x', the |
Martin Steigerwald | 5982a92 | 2011-06-27 16:07:24 +0200 | [diff] [blame] | 111 | value is assumed to be base 16 (hexadecimal). A suffix may include a trailing 'b', |
| 112 | for instance 'kb' is identical to 'k'. You can specify a base 10 value |
Jens Axboe | 57fc29f | 2010-06-23 22:24:07 +0200 | [diff] [blame] | 113 | by using 'KiB', 'MiB', 'GiB', etc. This is useful for disk drives where |
| 114 | values are often given in base 10 values. Specifying '30GiB' will get you |
| 115 | 30*1000^3 bytes. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 116 | .TP |
| 117 | .I bool |
| 118 | Boolean: a true or false value. `0' denotes false, `1' denotes true. |
| 119 | .TP |
| 120 | .I irange |
| 121 | Integer range: a range of integers specified in the format |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 122 | \fIlower\fR:\fIupper\fR or \fIlower\fR\-\fIupper\fR. \fIlower\fR and |
| 123 | \fIupper\fR may contain a suffix as described above. If an option allows two |
| 124 | sets of ranges, they are separated with a `,' or `/' character. For example: |
| 125 | `8\-8k/8M\-4G'. |
Yu-ju Hong | 8334919 | 2011-08-13 00:53:44 +0200 | [diff] [blame] | 126 | .TP |
| 127 | .I float_list |
| 128 | List of floating numbers: A list of floating numbers, separated by |
| 129 | a ':' charcater. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 130 | .SS "Parameter List" |
| 131 | .TP |
| 132 | .BI name \fR=\fPstr |
Aaron Carroll | d9956b6 | 2007-11-16 12:12:45 +0100 | [diff] [blame] | 133 | May be used to override the job name. On the command line, this parameter |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 134 | has the special purpose of signalling the start of a new job. |
| 135 | .TP |
| 136 | .BI description \fR=\fPstr |
| 137 | Human-readable description of the job. It is printed when the job is run, but |
| 138 | otherwise has no special purpose. |
| 139 | .TP |
| 140 | .BI directory \fR=\fPstr |
| 141 | Prefix filenames with this directory. Used to place files in a location other |
| 142 | than `./'. |
| 143 | .TP |
| 144 | .BI filename \fR=\fPstr |
| 145 | .B fio |
| 146 | normally makes up a file name based on the job name, thread number, and file |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 147 | number. If you want to share files between threads in a job or several jobs, |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 148 | specify a \fIfilename\fR for each of them to override the default. |
| 149 | If the I/O engine is file-based, you can specify |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 150 | a number of files by separating the names with a `:' character. `\-' is a |
| 151 | reserved name, meaning stdin or stdout, depending on the read/write direction |
| 152 | set. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 153 | .TP |
Jens Axboe | de98bd3 | 2013-04-05 11:09:20 +0200 | [diff] [blame] | 154 | .BI filename_format \fR=\fPstr |
Jens Axboe | ce594fb | 2013-04-05 16:32:33 +0200 | [diff] [blame] | 155 | If sharing multiple files between jobs, it is usually necessary to have |
Jens Axboe | de98bd3 | 2013-04-05 11:09:20 +0200 | [diff] [blame] | 156 | fio generate the exact names that you want. By default, fio will name a file |
| 157 | based on the default file format specification of |
| 158 | \fBjobname.jobnumber.filenumber\fP. With this option, that can be |
| 159 | customized. Fio will recognize and replace the following keywords in this |
| 160 | string: |
| 161 | .RS |
| 162 | .RS |
| 163 | .TP |
| 164 | .B $jobname |
| 165 | The name of the worker thread or process. |
| 166 | .TP |
| 167 | .B $jobnum |
| 168 | The incremental number of the worker thread or process. |
| 169 | .TP |
| 170 | .B $filenum |
| 171 | The incremental number of the file for that worker thread or process. |
| 172 | .RE |
| 173 | .P |
| 174 | To have dependent jobs share a set of files, this option can be set to |
| 175 | have fio generate filenames that are shared between the two. For instance, |
| 176 | if \fBtestfiles.$filenum\fR is specified, file number 4 for any job will |
| 177 | be named \fBtestfiles.4\fR. The default of \fB$jobname.$jobnum.$filenum\fR |
| 178 | will be used if no other format specifier is given. |
| 179 | .RE |
| 180 | .P |
| 181 | .TP |
Jens Axboe | 3ce9dca | 2009-06-10 08:55:21 +0200 | [diff] [blame] | 182 | .BI lockfile \fR=\fPstr |
| 183 | Fio defaults to not locking any files before it does IO to them. If a file or |
| 184 | file descriptor is shared, fio can serialize IO to that file to make the end |
| 185 | result consistent. This is usual for emulating real workloads that share files. |
| 186 | The lock modes are: |
| 187 | .RS |
| 188 | .RS |
| 189 | .TP |
| 190 | .B none |
| 191 | No locking. This is the default. |
| 192 | .TP |
| 193 | .B exclusive |
| 194 | Only one thread or process may do IO at the time, excluding all others. |
| 195 | .TP |
| 196 | .B readwrite |
| 197 | Read-write locking on the file. Many readers may access the file at the same |
| 198 | time, but writes get exclusive access. |
| 199 | .RE |
Jens Axboe | ce594fb | 2013-04-05 16:32:33 +0200 | [diff] [blame] | 200 | .RE |
Jens Axboe | 3ce9dca | 2009-06-10 08:55:21 +0200 | [diff] [blame] | 201 | .P |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 202 | .BI opendir \fR=\fPstr |
| 203 | Recursively open any files below directory \fIstr\fR. |
| 204 | .TP |
| 205 | .BI readwrite \fR=\fPstr "\fR,\fP rw" \fR=\fPstr |
| 206 | Type of I/O pattern. Accepted values are: |
| 207 | .RS |
| 208 | .RS |
| 209 | .TP |
| 210 | .B read |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 211 | Sequential reads. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 212 | .TP |
| 213 | .B write |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 214 | Sequential writes. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 215 | .TP |
| 216 | .B randread |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 217 | Random reads. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 218 | .TP |
| 219 | .B randwrite |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 220 | Random writes. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 221 | .TP |
Jens Axboe | 10b023d | 2012-03-23 13:40:06 +0100 | [diff] [blame] | 222 | .B rw, readwrite |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 223 | Mixed sequential reads and writes. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 224 | .TP |
| 225 | .B randrw |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 226 | Mixed random reads and writes. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 227 | .RE |
| 228 | .P |
Jens Axboe | 38dad62 | 2010-07-20 14:46:00 -0600 | [diff] [blame] | 229 | For mixed I/O, the default split is 50/50. For certain types of io the result |
| 230 | may still be skewed a bit, since the speed may be different. It is possible to |
Jens Axboe | 3b7fa9e | 2012-04-26 19:39:47 +0200 | [diff] [blame] | 231 | specify a number of IO's to do before getting a new offset, this is done by |
Jens Axboe | 38dad62 | 2010-07-20 14:46:00 -0600 | [diff] [blame] | 232 | appending a `:\fI<nr>\fR to the end of the string given. For a random read, it |
| 233 | would look like \fBrw=randread:8\fR for passing in an offset modifier with a |
Jens Axboe | 059b080 | 2011-08-25 09:09:37 +0200 | [diff] [blame] | 234 | value of 8. If the postfix is used with a sequential IO pattern, then the value |
| 235 | specified will be added to the generated offset for each IO. For instance, |
| 236 | using \fBrw=write:4k\fR will skip 4k for every write. It turns sequential IO |
| 237 | into sequential IO with holes. See the \fBrw_sequencer\fR option. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 238 | .RE |
| 239 | .TP |
Jens Axboe | 38dad62 | 2010-07-20 14:46:00 -0600 | [diff] [blame] | 240 | .BI rw_sequencer \fR=\fPstr |
| 241 | If an offset modifier is given by appending a number to the \fBrw=<str>\fR line, |
| 242 | then this option controls how that number modifies the IO offset being |
| 243 | generated. Accepted values are: |
| 244 | .RS |
| 245 | .RS |
| 246 | .TP |
| 247 | .B sequential |
| 248 | Generate sequential offset |
| 249 | .TP |
| 250 | .B identical |
| 251 | Generate the same offset |
| 252 | .RE |
| 253 | .P |
| 254 | \fBsequential\fR is only useful for random IO, where fio would normally |
| 255 | generate a new random offset for every IO. If you append eg 8 to randread, you |
| 256 | would get a new random offset for every 8 IO's. The result would be a seek for |
| 257 | only every 8 IO's, instead of for every IO. Use \fBrw=randread:8\fR to specify |
| 258 | that. As sequential IO is already sequential, setting \fBsequential\fR for that |
| 259 | would not result in any differences. \fBidentical\fR behaves in a similar |
| 260 | fashion, except it sends the same offset 8 number of times before generating a |
| 261 | new offset. |
| 262 | .RE |
| 263 | .P |
| 264 | .TP |
Jens Axboe | 90fef2d | 2009-07-17 22:33:32 +0200 | [diff] [blame] | 265 | .BI kb_base \fR=\fPint |
| 266 | The base unit for a kilobyte. The defacto base is 2^10, 1024. Storage |
| 267 | manufacturers like to use 10^3 or 1000 as a base ten unit instead, for obvious |
| 268 | reasons. Allow values are 1024 or 1000, with 1024 being the default. |
| 269 | .TP |
Jens Axboe | 771e58b | 2013-01-30 12:56:23 +0100 | [diff] [blame] | 270 | .BI unified_rw_reporting \fR=\fPbool |
| 271 | Fio normally reports statistics on a per data direction basis, meaning that |
| 272 | read, write, and trim are accounted and reported separately. If this option is |
| 273 | set, the fio will sum the results and report them as "mixed" instead. |
| 274 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 275 | .BI randrepeat \fR=\fPbool |
| 276 | Seed the random number generator in a predictable way so results are repeatable |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 277 | across runs. Default: true. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 278 | .TP |
Jens Axboe | 2615cc4 | 2011-03-28 09:35:09 +0200 | [diff] [blame] | 279 | .BI use_os_rand \fR=\fPbool |
| 280 | Fio can either use the random generator supplied by the OS to generator random |
| 281 | offsets, or it can use it's own internal generator (based on Tausworthe). |
| 282 | Default is to use the internal generator, which is often of better quality and |
| 283 | faster. Default: false. |
| 284 | .TP |
Eric Gouriou | a596f04 | 2011-06-17 09:11:45 +0200 | [diff] [blame] | 285 | .BI fallocate \fR=\fPstr |
| 286 | Whether pre-allocation is performed when laying down files. Accepted values |
| 287 | are: |
| 288 | .RS |
| 289 | .RS |
| 290 | .TP |
| 291 | .B none |
| 292 | Do not pre-allocate space. |
| 293 | .TP |
| 294 | .B posix |
| 295 | Pre-allocate via posix_fallocate(). |
| 296 | .TP |
| 297 | .B keep |
| 298 | Pre-allocate via fallocate() with FALLOC_FL_KEEP_SIZE set. |
| 299 | .TP |
| 300 | .B 0 |
| 301 | Backward-compatible alias for 'none'. |
| 302 | .TP |
| 303 | .B 1 |
| 304 | Backward-compatible alias for 'posix'. |
| 305 | .RE |
| 306 | .P |
| 307 | May not be available on all supported platforms. 'keep' is only |
| 308 | available on Linux. If using ZFS on Solaris this must be set to 'none' |
| 309 | because ZFS doesn't support it. Default: 'posix'. |
| 310 | .RE |
Jens Axboe | 7bc8c2c | 2010-01-28 11:31:31 +0100 | [diff] [blame] | 311 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 312 | .BI fadvise_hint \fR=\fPbool |
Zhu Yanhai | 23a7b04 | 2012-01-02 14:32:43 +0100 | [diff] [blame] | 313 | Use of \fIposix_fadvise\fR\|(2) to advise the kernel what I/O patterns |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 314 | are likely to be issued. Default: true. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 315 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 316 | .BI size \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 317 | Total size of I/O for this job. \fBfio\fR will run until this many bytes have |
| 318 | been transfered, unless limited by other options (\fBruntime\fR, for instance). |
Jens Axboe | d7c8be0 | 2010-11-25 08:21:39 +0100 | [diff] [blame] | 319 | Unless \fBnrfiles\fR and \fBfilesize\fR options are given, this amount will be |
Jens Axboe | d666726 | 2010-06-25 11:32:48 +0200 | [diff] [blame] | 320 | divided between the available files for the job. If not set, fio will use the |
| 321 | full size of the given files or devices. If the the files do not exist, size |
Jens Axboe | 7bb5910 | 2011-07-12 19:47:03 +0200 | [diff] [blame] | 322 | must be given. It is also possible to give size as a percentage between 1 and |
| 323 | 100. If size=20% is given, fio will use 20% of the full size of the given files |
| 324 | or devices. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 325 | .TP |
Jens Axboe | 74586c1 | 2011-01-20 10:16:03 -0700 | [diff] [blame] | 326 | .BI fill_device \fR=\fPbool "\fR,\fB fill_fs" \fR=\fPbool |
Jens Axboe | 3ce9dca | 2009-06-10 08:55:21 +0200 | [diff] [blame] | 327 | Sets size to something really large and waits for ENOSPC (no space left on |
| 328 | device) as the terminating condition. Only makes sense with sequential write. |
| 329 | For a read workload, the mount point will be filled first then IO started on |
Jens Axboe | 4f12432 | 2011-01-19 15:35:26 -0700 | [diff] [blame] | 330 | the result. This option doesn't make sense if operating on a raw device node, |
| 331 | since the size of that is already known by the file system. Additionally, |
| 332 | writing beyond end-of-device will not return ENOSPC there. |
Jens Axboe | 3ce9dca | 2009-06-10 08:55:21 +0200 | [diff] [blame] | 333 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 334 | .BI filesize \fR=\fPirange |
| 335 | Individual file sizes. May be a range, in which case \fBfio\fR will select sizes |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 336 | for files at random within the given range, limited to \fBsize\fR in total (if |
| 337 | that is given). If \fBfilesize\fR is not specified, each created file is the |
| 338 | same size. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 339 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 340 | .BI blocksize \fR=\fPint[,int] "\fR,\fB bs" \fR=\fPint[,int] |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 341 | Block size for I/O units. Default: 4k. Values for reads and writes can be |
Jens Axboe | 656ebab | 2010-04-13 10:39:14 +0200 | [diff] [blame] | 342 | specified separately in the format \fIread\fR,\fIwrite\fR, either of |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 343 | which may be empty to leave that value at its default. |
| 344 | .TP |
Jens Axboe | 9183788 | 2008-02-05 12:02:07 +0100 | [diff] [blame] | 345 | .BI blocksize_range \fR=\fPirange[,irange] "\fR,\fB bsrange" \fR=\fPirange[,irange] |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 346 | Specify a range of I/O block sizes. The issued I/O unit will always be a |
| 347 | multiple of the minimum size, unless \fBblocksize_unaligned\fR is set. Applies |
Jens Axboe | 9183788 | 2008-02-05 12:02:07 +0100 | [diff] [blame] | 348 | to both reads and writes if only one range is given, but can be specified |
Jens Axboe | 656ebab | 2010-04-13 10:39:14 +0200 | [diff] [blame] | 349 | separately with a comma seperating the values. Example: bsrange=1k-4k,2k-8k. |
Jens Axboe | 9183788 | 2008-02-05 12:02:07 +0100 | [diff] [blame] | 350 | Also (see \fBblocksize\fR). |
| 351 | .TP |
| 352 | .BI bssplit \fR=\fPstr |
| 353 | This option allows even finer grained control of the block sizes issued, |
| 354 | not just even splits between them. With this option, you can weight various |
| 355 | block sizes for exact control of the issued IO for a job that has mixed |
| 356 | block sizes. The format of the option is bssplit=blocksize/percentage, |
Martin Steigerwald | 5982a92 | 2011-06-27 16:07:24 +0200 | [diff] [blame] | 357 | optionally adding as many definitions as needed separated by a colon. |
Jens Axboe | 9183788 | 2008-02-05 12:02:07 +0100 | [diff] [blame] | 358 | Example: bssplit=4k/10:64k/50:32k/40 would issue 50% 64k blocks, 10% 4k |
Jens Axboe | c83cdd3 | 2009-04-24 14:23:59 +0200 | [diff] [blame] | 359 | blocks and 40% 32k blocks. \fBbssplit\fR also supports giving separate |
| 360 | splits to reads and writes. The format is identical to what the |
| 361 | \fBbs\fR option accepts, the read and write parts are separated with a |
| 362 | comma. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 363 | .TP |
| 364 | .B blocksize_unaligned\fR,\fP bs_unaligned |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 365 | If set, any size in \fBblocksize_range\fR may be used. This typically won't |
| 366 | work with direct I/O, as that normally requires sector alignment. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 367 | .TP |
Jens Axboe | 2b7a01d | 2009-03-11 11:00:13 +0100 | [diff] [blame] | 368 | .BI blockalign \fR=\fPint[,int] "\fR,\fB ba" \fR=\fPint[,int] |
Martin Steigerwald | 639ce0f | 2009-05-20 11:33:49 +0200 | [diff] [blame] | 369 | At what boundary to align random IO offsets. Defaults to the same as 'blocksize' |
| 370 | the minimum blocksize given. Minimum alignment is typically 512b |
Jens Axboe | 2b7a01d | 2009-03-11 11:00:13 +0100 | [diff] [blame] | 371 | for using direct IO, though it usually depends on the hardware block size. |
| 372 | This option is mutually exclusive with using a random map for files, so it |
| 373 | will turn off that option. |
Jens Axboe | 4360266 | 2009-03-14 20:08:47 +0100 | [diff] [blame] | 374 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 375 | .B zero_buffers |
| 376 | Initialise buffers with all zeros. Default: fill buffers with random data. |
| 377 | .TP |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 378 | .B refill_buffers |
| 379 | If this option is given, fio will refill the IO buffers on every submit. The |
| 380 | default is to only fill it at init time and reuse that data. Only makes sense |
| 381 | if zero_buffers isn't specified, naturally. If data verification is enabled, |
| 382 | refill_buffers is also automatically enabled. |
| 383 | .TP |
Jens Axboe | fd68418 | 2011-09-19 09:24:44 +0200 | [diff] [blame] | 384 | .BI scramble_buffers \fR=\fPbool |
| 385 | If \fBrefill_buffers\fR is too costly and the target is using data |
| 386 | deduplication, then setting this option will slightly modify the IO buffer |
| 387 | contents to defeat normal de-dupe attempts. This is not enough to defeat |
| 388 | more clever block compression attempts, but it will stop naive dedupe |
| 389 | of blocks. Default: true. |
| 390 | .TP |
Jens Axboe | c5751c6 | 2012-03-15 15:02:56 +0100 | [diff] [blame] | 391 | .BI buffer_compress_percentage \fR=\fPint |
| 392 | If this is set, then fio will attempt to provide IO buffer content (on WRITEs) |
| 393 | that compress to the specified level. Fio does this by providing a mix of |
| 394 | random data and zeroes. Note that this is per block size unit, for file/disk |
| 395 | wide compression level that matches this setting, you'll also want to set |
| 396 | \fBrefill_buffers\fR. |
| 397 | .TP |
| 398 | .BI buffer_compress_chunk \fR=\fPint |
| 399 | See \fBbuffer_compress_percentage\fR. This setting allows fio to manage how |
| 400 | big the ranges of random data and zeroed data is. Without this set, fio will |
| 401 | provide \fBbuffer_compress_percentage\fR of blocksize random data, followed by |
| 402 | the remaining zeroed. With this set to some chunk size smaller than the block |
| 403 | size, fio can alternate random and zeroed data throughout the IO buffer. |
| 404 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 405 | .BI nrfiles \fR=\fPint |
| 406 | Number of files to use for this job. Default: 1. |
| 407 | .TP |
| 408 | .BI openfiles \fR=\fPint |
| 409 | Number of files to keep open at the same time. Default: \fBnrfiles\fR. |
| 410 | .TP |
| 411 | .BI file_service_type \fR=\fPstr |
| 412 | Defines how files to service are selected. The following types are defined: |
| 413 | .RS |
| 414 | .RS |
| 415 | .TP |
| 416 | .B random |
| 417 | Choose a file at random |
| 418 | .TP |
| 419 | .B roundrobin |
| 420 | Round robin over open files (default). |
Jens Axboe | 6b7f685 | 2009-03-09 14:22:56 +0100 | [diff] [blame] | 421 | .B sequential |
| 422 | Do each file in the set sequentially. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 423 | .RE |
| 424 | .P |
| 425 | The number of I/Os to issue before switching a new file can be specified by |
| 426 | appending `:\fIint\fR' to the service type. |
| 427 | .RE |
| 428 | .TP |
| 429 | .BI ioengine \fR=\fPstr |
| 430 | Defines how the job issues I/O. The following types are defined: |
| 431 | .RS |
| 432 | .RS |
| 433 | .TP |
| 434 | .B sync |
| 435 | Basic \fIread\fR\|(2) or \fIwrite\fR\|(2) I/O. \fIfseek\fR\|(2) is used to |
| 436 | position the I/O location. |
| 437 | .TP |
gurudas pai | a31041e | 2007-10-23 15:12:30 +0200 | [diff] [blame] | 438 | .B psync |
| 439 | Basic \fIpread\fR\|(2) or \fIpwrite\fR\|(2) I/O. |
| 440 | .TP |
Jens Axboe | 9183788 | 2008-02-05 12:02:07 +0100 | [diff] [blame] | 441 | .B vsync |
| 442 | Basic \fIreadv\fR\|(2) or \fIwritev\fR\|(2) I/O. Will emulate queuing by |
| 443 | coalescing adjacents IOs into a single submission. |
| 444 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 445 | .B libaio |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 446 | Linux native asynchronous I/O. This ioengine defines engine specific options. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 447 | .TP |
| 448 | .B posixaio |
Bruce Cran | 03e20d6 | 2011-01-02 20:14:54 +0100 | [diff] [blame] | 449 | POSIX asynchronous I/O using \fIaio_read\fR\|(3) and \fIaio_write\fR\|(3). |
| 450 | .TP |
| 451 | .B solarisaio |
| 452 | Solaris native asynchronous I/O. |
| 453 | .TP |
| 454 | .B windowsaio |
| 455 | Windows native asynchronous I/O. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 456 | .TP |
| 457 | .B mmap |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 458 | File is memory mapped with \fImmap\fR\|(2) and data copied using |
| 459 | \fImemcpy\fR\|(3). |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 460 | .TP |
| 461 | .B splice |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 462 | \fIsplice\fR\|(2) is used to transfer the data and \fIvmsplice\fR\|(2) to |
| 463 | transfer data from user-space to the kernel. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 464 | .TP |
| 465 | .B syslet-rw |
| 466 | Use the syslet system calls to make regular read/write asynchronous. |
| 467 | .TP |
| 468 | .B sg |
| 469 | SCSI generic sg v3 I/O. May be either synchronous using the SG_IO ioctl, or if |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 470 | the target is an sg character device, we use \fIread\fR\|(2) and |
| 471 | \fIwrite\fR\|(2) for asynchronous I/O. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 472 | .TP |
| 473 | .B null |
| 474 | Doesn't transfer any data, just pretends to. Mainly used to exercise \fBfio\fR |
| 475 | itself and for debugging and testing purposes. |
| 476 | .TP |
| 477 | .B net |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 478 | Transfer over the network. The protocol to be used can be defined with the |
| 479 | \fBprotocol\fR parameter. Depending on the protocol, \fBfilename\fR, |
| 480 | \fBhostname\fR, \fBport\fR, or \fBlisten\fR must be specified. |
| 481 | This ioengine defines engine specific options. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 482 | .TP |
| 483 | .B netsplice |
| 484 | Like \fBnet\fR, but uses \fIsplice\fR\|(2) and \fIvmsplice\fR\|(2) to map data |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 485 | and send/receive. This ioengine defines engine specific options. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 486 | .TP |
gurudas pai | 53aec0a | 2007-10-05 13:20:18 +0200 | [diff] [blame] | 487 | .B cpuio |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 488 | Doesn't transfer any data, but burns CPU cycles according to \fBcpuload\fR and |
| 489 | \fBcpucycles\fR parameters. |
| 490 | .TP |
| 491 | .B guasi |
| 492 | The GUASI I/O engine is the Generic Userspace Asynchronous Syscall Interface |
| 493 | approach to asycnronous I/O. |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 494 | .br |
| 495 | See <http://www.xmailserver.org/guasi\-lib.html>. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 496 | .TP |
ren yufei | 21b8aee | 2011-08-01 10:01:57 +0200 | [diff] [blame] | 497 | .B rdma |
Bart Van Assche | 85286c5 | 2011-08-07 21:50:51 +0200 | [diff] [blame] | 498 | The RDMA I/O engine supports both RDMA memory semantics (RDMA_WRITE/RDMA_READ) |
| 499 | and channel semantics (Send/Recv) for the InfiniBand, RoCE and iWARP protocols. |
ren yufei | 21b8aee | 2011-08-01 10:01:57 +0200 | [diff] [blame] | 500 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 501 | .B external |
| 502 | Loads an external I/O engine object file. Append the engine filename as |
| 503 | `:\fIenginepath\fR'. |
Dmitry Monakhov | d54fce8 | 2012-09-20 15:37:17 +0400 | [diff] [blame] | 504 | .TP |
| 505 | .B falloc |
| 506 | IO engine that does regular linux native fallocate callt to simulate data |
| 507 | transfer as fio ioengine |
| 508 | .br |
| 509 | DDIR_READ does fallocate(,mode = FALLOC_FL_KEEP_SIZE,) |
| 510 | .br |
Jens Axboe | 0981fd7 | 2012-09-20 19:23:02 +0200 | [diff] [blame] | 511 | DIR_WRITE does fallocate(,mode = 0) |
Dmitry Monakhov | d54fce8 | 2012-09-20 15:37:17 +0400 | [diff] [blame] | 512 | .br |
| 513 | DDIR_TRIM does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE) |
| 514 | .TP |
| 515 | .B e4defrag |
| 516 | IO engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate defragment activity |
| 517 | request to DDIR_WRITE event |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 518 | .RE |
Jens Axboe | 595e173 | 2012-12-05 21:15:01 +0100 | [diff] [blame] | 519 | .P |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 520 | .RE |
| 521 | .TP |
| 522 | .BI iodepth \fR=\fPint |
Sebastian Kayser | 8489dae | 2010-12-01 22:28:47 +0100 | [diff] [blame] | 523 | Number of I/O units to keep in flight against the file. Note that increasing |
| 524 | iodepth beyond 1 will not affect synchronous ioengines (except for small |
Jens Axboe | ee72ca0 | 2010-12-02 20:05:37 +0100 | [diff] [blame] | 525 | degress when verify_async is in use). Even async engines my impose OS |
| 526 | restrictions causing the desired depth not to be achieved. This may happen on |
| 527 | Linux when using libaio and not setting \fBdirect\fR=1, since buffered IO is |
| 528 | not async on that OS. Keep an eye on the IO depth distribution in the |
| 529 | fio output to verify that the achieved depth is as expected. Default: 1. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 530 | .TP |
| 531 | .BI iodepth_batch \fR=\fPint |
| 532 | Number of I/Os to submit at once. Default: \fBiodepth\fR. |
| 533 | .TP |
Jens Axboe | 3ce9dca | 2009-06-10 08:55:21 +0200 | [diff] [blame] | 534 | .BI iodepth_batch_complete \fR=\fPint |
| 535 | This defines how many pieces of IO to retrieve at once. It defaults to 1 which |
| 536 | means that we'll ask for a minimum of 1 IO in the retrieval process from the |
| 537 | kernel. The IO retrieval will go on until we hit the limit set by |
| 538 | \fBiodepth_low\fR. If this variable is set to 0, then fio will always check for |
| 539 | completed events before queuing more IO. This helps reduce IO latency, at the |
| 540 | cost of more retrieval system calls. |
| 541 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 542 | .BI iodepth_low \fR=\fPint |
| 543 | Low watermark indicating when to start filling the queue again. Default: |
| 544 | \fBiodepth\fR. |
| 545 | .TP |
| 546 | .BI direct \fR=\fPbool |
| 547 | If true, use non-buffered I/O (usually O_DIRECT). Default: false. |
| 548 | .TP |
| 549 | .BI buffered \fR=\fPbool |
| 550 | If true, use buffered I/O. This is the opposite of the \fBdirect\fR parameter. |
| 551 | Default: true. |
| 552 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 553 | .BI offset \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 554 | Offset in the file to start I/O. Data before the offset will not be touched. |
| 555 | .TP |
Jens Axboe | 591e9e0 | 2012-03-15 14:50:58 +0100 | [diff] [blame] | 556 | .BI offset_increment \fR=\fPint |
| 557 | If this is provided, then the real offset becomes the |
| 558 | offset + offset_increment * thread_number, where the thread number is a counter |
| 559 | that starts at 0 and is incremented for each job. This option is useful if |
| 560 | there are several jobs which are intended to operate on a file in parallel in |
| 561 | disjoint segments, with even spacing between the starting points. |
| 562 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 563 | .BI fsync \fR=\fPint |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 564 | How many I/Os to perform before issuing an \fBfsync\fR\|(2) of dirty data. If |
| 565 | 0, don't sync. Default: 0. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 566 | .TP |
Jens Axboe | 5f9099e | 2009-06-16 22:40:26 +0200 | [diff] [blame] | 567 | .BI fdatasync \fR=\fPint |
| 568 | Like \fBfsync\fR, but uses \fBfdatasync\fR\|(2) instead to only sync the |
| 569 | data parts of the file. Default: 0. |
| 570 | .TP |
Jens Axboe | e76b1da | 2010-03-09 20:49:54 +0100 | [diff] [blame] | 571 | .BI sync_file_range \fR=\fPstr:int |
| 572 | Use sync_file_range() for every \fRval\fP number of write operations. Fio will |
| 573 | track range of writes that have happened since the last sync_file_range() call. |
| 574 | \fRstr\fP can currently be one or more of: |
| 575 | .RS |
| 576 | .TP |
| 577 | .B wait_before |
| 578 | SYNC_FILE_RANGE_WAIT_BEFORE |
| 579 | .TP |
| 580 | .B write |
| 581 | SYNC_FILE_RANGE_WRITE |
| 582 | .TP |
| 583 | .B wait_after |
| 584 | SYNC_FILE_RANGE_WRITE |
| 585 | .TP |
| 586 | .RE |
| 587 | .P |
| 588 | So if you do sync_file_range=wait_before,write:8, fio would use |
| 589 | \fBSYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE\fP for every 8 writes. |
| 590 | Also see the sync_file_range(2) man page. This option is Linux specific. |
| 591 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 592 | .BI overwrite \fR=\fPbool |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 593 | If writing, setup the file first and do overwrites. Default: false. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 594 | .TP |
| 595 | .BI end_fsync \fR=\fPbool |
Jens Axboe | dbd11ea | 2013-01-13 17:16:46 +0100 | [diff] [blame] | 596 | Sync file contents when a write stage has completed. Default: false. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 597 | .TP |
| 598 | .BI fsync_on_close \fR=\fPbool |
| 599 | If true, sync file contents on close. This differs from \fBend_fsync\fR in that |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 600 | it will happen on every close, not just at the end of the job. Default: false. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 601 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 602 | .BI rwmixread \fR=\fPint |
| 603 | Percentage of a mixed workload that should be reads. Default: 50. |
| 604 | .TP |
| 605 | .BI rwmixwrite \fR=\fPint |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 606 | Percentage of a mixed workload that should be writes. If \fBrwmixread\fR and |
Jens Axboe | c35dd7a | 2009-06-10 08:39:16 +0200 | [diff] [blame] | 607 | \fBrwmixwrite\fR are given and do not sum to 100%, the latter of the two |
| 608 | overrides the first. This may interfere with a given rate setting, if fio is |
| 609 | asked to limit reads or writes to a certain rate. If that is the case, then |
| 610 | the distribution may be skewed. Default: 50. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 611 | .TP |
Jens Axboe | 92d42d6 | 2012-11-15 15:38:32 -0700 | [diff] [blame] | 612 | .BI random_distribution \fR=\fPstr:float |
| 613 | By default, fio will use a completely uniform random distribution when asked |
| 614 | to perform random IO. Sometimes it is useful to skew the distribution in |
| 615 | specific ways, ensuring that some parts of the data is more hot than others. |
| 616 | Fio includes the following distribution models: |
| 617 | .RS |
| 618 | .TP |
| 619 | .B random |
| 620 | Uniform random distribution |
| 621 | .TP |
| 622 | .B zipf |
| 623 | Zipf distribution |
| 624 | .TP |
| 625 | .B pareto |
| 626 | Pareto distribution |
| 627 | .TP |
| 628 | .RE |
| 629 | .P |
| 630 | When using a zipf or pareto distribution, an input value is also needed to |
| 631 | define the access pattern. For zipf, this is the zipf theta. For pareto, |
| 632 | it's the pareto power. Fio includes a test program, genzipf, that can be |
| 633 | used visualize what the given input values will yield in terms of hit rates. |
| 634 | If you wanted to use zipf with a theta of 1.2, you would use |
| 635 | random_distribution=zipf:1.2 as the option. If a non-uniform model is used, |
| 636 | fio will disable use of the random map. |
| 637 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 638 | .B norandommap |
| 639 | Normally \fBfio\fR will cover every block of the file when doing random I/O. If |
| 640 | this parameter is given, a new offset will be chosen without looking at past |
| 641 | I/O history. This parameter is mutually exclusive with \fBverify\fR. |
| 642 | .TP |
Jens Axboe | 744492c | 2011-08-08 09:47:13 +0200 | [diff] [blame] | 643 | .BI softrandommap \fR=\fPbool |
Jens Axboe | 3ce9dca | 2009-06-10 08:55:21 +0200 | [diff] [blame] | 644 | See \fBnorandommap\fR. If fio runs with the random block map enabled and it |
| 645 | fails to allocate the map, if this option is set it will continue without a |
| 646 | random block map. As coverage will not be as complete as with random maps, this |
| 647 | option is disabled by default. |
| 648 | .TP |
Jens Axboe | e8b1961 | 2012-12-05 10:28:08 +0100 | [diff] [blame] | 649 | .BI random_generator \fR=\fPstr |
| 650 | Fio supports the following engines for generating IO offsets for random IO: |
| 651 | .RS |
| 652 | .TP |
| 653 | .B tausworthe |
| 654 | Strong 2^88 cycle random number generator |
| 655 | .TP |
| 656 | .B lfsr |
| 657 | Linear feedback shift register generator |
| 658 | .TP |
| 659 | .RE |
| 660 | .P |
| 661 | Tausworthe is a strong random number generator, but it requires tracking on the |
| 662 | side if we want to ensure that blocks are only read or written once. LFSR |
| 663 | guarantees that we never generate the same offset twice, and it's also less |
| 664 | computationally expensive. It's not a true random generator, however, though |
| 665 | for IO purposes it's typically good enough. LFSR only works with single block |
| 666 | sizes, not with workloads that use multiple block sizes. If used with such a |
| 667 | workload, fio may read or write some blocks multiple times. |
| 668 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 669 | .BI nice \fR=\fPint |
| 670 | Run job with given nice value. See \fInice\fR\|(2). |
| 671 | .TP |
| 672 | .BI prio \fR=\fPint |
| 673 | Set I/O priority value of this job between 0 (highest) and 7 (lowest). See |
| 674 | \fIionice\fR\|(1). |
| 675 | .TP |
| 676 | .BI prioclass \fR=\fPint |
| 677 | Set I/O priority class. See \fIionice\fR\|(1). |
| 678 | .TP |
| 679 | .BI thinktime \fR=\fPint |
| 680 | Stall job for given number of microseconds between issuing I/Os. |
| 681 | .TP |
| 682 | .BI thinktime_spin \fR=\fPint |
| 683 | Pretend to spend CPU time for given number of microseconds, sleeping the rest |
| 684 | of the time specified by \fBthinktime\fR. Only valid if \fBthinktime\fR is set. |
| 685 | .TP |
| 686 | .BI thinktime_blocks \fR=\fPint |
| 687 | Number of blocks to issue before waiting \fBthinktime\fR microseconds. |
| 688 | Default: 1. |
| 689 | .TP |
| 690 | .BI rate \fR=\fPint |
Jens Axboe | c35dd7a | 2009-06-10 08:39:16 +0200 | [diff] [blame] | 691 | Cap bandwidth used by this job. The number is in bytes/sec, the normal postfix |
| 692 | rules apply. You can use \fBrate\fR=500k to limit reads and writes to 500k each, |
| 693 | or you can specify read and writes separately. Using \fBrate\fR=1m,500k would |
| 694 | limit reads to 1MB/sec and writes to 500KB/sec. Capping only reads or writes |
| 695 | can be done with \fBrate\fR=,500k or \fBrate\fR=500k,. The former will only |
| 696 | limit writes (to 500KB/sec), the latter will only limit reads. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 697 | .TP |
| 698 | .BI ratemin \fR=\fPint |
| 699 | Tell \fBfio\fR to do whatever it can to maintain at least the given bandwidth. |
Jens Axboe | c35dd7a | 2009-06-10 08:39:16 +0200 | [diff] [blame] | 700 | Failing to meet this requirement will cause the job to exit. The same format |
| 701 | as \fBrate\fR is used for read vs write separation. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 702 | .TP |
| 703 | .BI rate_iops \fR=\fPint |
Jens Axboe | c35dd7a | 2009-06-10 08:39:16 +0200 | [diff] [blame] | 704 | Cap the bandwidth to this number of IOPS. Basically the same as rate, just |
| 705 | specified independently of bandwidth. The same format as \fBrate\fR is used for |
| 706 | read vs write seperation. If \fBblocksize\fR is a range, the smallest block |
| 707 | size is used as the metric. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 708 | .TP |
| 709 | .BI rate_iops_min \fR=\fPint |
Jens Axboe | c35dd7a | 2009-06-10 08:39:16 +0200 | [diff] [blame] | 710 | If this rate of I/O is not met, the job will exit. The same format as \fBrate\fR |
| 711 | is used for read vs write seperation. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 712 | .TP |
| 713 | .BI ratecycle \fR=\fPint |
| 714 | Average bandwidth for \fBrate\fR and \fBratemin\fR over this number of |
| 715 | milliseconds. Default: 1000ms. |
| 716 | .TP |
Jens Axboe | 1550153 | 2012-10-24 16:37:45 +0200 | [diff] [blame] | 717 | .BI max_latency \fR=\fPint |
| 718 | If set, fio will exit the job if it exceeds this maximum latency. It will exit |
| 719 | with an ETIME error. |
| 720 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 721 | .BI cpumask \fR=\fPint |
| 722 | Set CPU affinity for this job. \fIint\fR is a bitmask of allowed CPUs the job |
| 723 | may run on. See \fBsched_setaffinity\fR\|(2). |
| 724 | .TP |
| 725 | .BI cpus_allowed \fR=\fPstr |
| 726 | Same as \fBcpumask\fR, but allows a comma-delimited list of CPU numbers. |
| 727 | .TP |
Yufei Ren | d0b937e | 2012-10-19 23:11:52 -0400 | [diff] [blame] | 728 | .BI numa_cpu_nodes \fR=\fPstr |
| 729 | Set this job running on spcified NUMA nodes' CPUs. The arguments allow |
| 730 | comma delimited list of cpu numbers, A-B ranges, or 'all'. |
| 731 | .TP |
| 732 | .BI numa_mem_policy \fR=\fPstr |
| 733 | Set this job's memory policy and corresponding NUMA nodes. Format of |
| 734 | the argements: |
| 735 | .RS |
| 736 | .TP |
| 737 | .B <mode>[:<nodelist>] |
| 738 | .TP |
| 739 | .B mode |
| 740 | is one of the following memory policy: |
| 741 | .TP |
| 742 | .B default, prefer, bind, interleave, local |
| 743 | .TP |
| 744 | .RE |
| 745 | For \fBdefault\fR and \fBlocal\fR memory policy, no \fBnodelist\fR is |
| 746 | needed to be specified. For \fBprefer\fR, only one node is |
| 747 | allowed. For \fBbind\fR and \fBinterleave\fR, \fBnodelist\fR allows |
| 748 | comma delimited list of numbers, A-B ranges, or 'all'. |
| 749 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 750 | .BI startdelay \fR=\fPint |
| 751 | Delay start of job for the specified number of seconds. |
| 752 | .TP |
| 753 | .BI runtime \fR=\fPint |
| 754 | Terminate processing after the specified number of seconds. |
| 755 | .TP |
| 756 | .B time_based |
| 757 | If given, run for the specified \fBruntime\fR duration even if the files are |
| 758 | completely read or written. The same workload will be repeated as many times |
| 759 | as \fBruntime\fR allows. |
| 760 | .TP |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 761 | .BI ramp_time \fR=\fPint |
| 762 | If set, fio will run the specified workload for this amount of time before |
| 763 | logging any performance numbers. Useful for letting performance settle before |
| 764 | logging results, thus minimizing the runtime required for stable results. Note |
Jens Axboe | c35dd7a | 2009-06-10 08:39:16 +0200 | [diff] [blame] | 765 | that the \fBramp_time\fR is considered lead in time for a job, thus it will |
| 766 | increase the total runtime if a special timeout or runtime is specified. |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 767 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 768 | .BI invalidate \fR=\fPbool |
| 769 | Invalidate buffer-cache for the file prior to starting I/O. Default: true. |
| 770 | .TP |
| 771 | .BI sync \fR=\fPbool |
| 772 | Use synchronous I/O for buffered writes. For the majority of I/O engines, |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 773 | this means using O_SYNC. Default: false. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 774 | .TP |
| 775 | .BI iomem \fR=\fPstr "\fR,\fP mem" \fR=\fPstr |
| 776 | Allocation method for I/O unit buffer. Allowed values are: |
| 777 | .RS |
| 778 | .RS |
| 779 | .TP |
| 780 | .B malloc |
| 781 | Allocate memory with \fImalloc\fR\|(3). |
| 782 | .TP |
| 783 | .B shm |
| 784 | Use shared memory buffers allocated through \fIshmget\fR\|(2). |
| 785 | .TP |
| 786 | .B shmhuge |
| 787 | Same as \fBshm\fR, but use huge pages as backing. |
| 788 | .TP |
| 789 | .B mmap |
| 790 | Use \fImmap\fR\|(2) for allocation. Uses anonymous memory unless a filename |
| 791 | is given after the option in the format `:\fIfile\fR'. |
| 792 | .TP |
| 793 | .B mmaphuge |
| 794 | Same as \fBmmap\fR, but use huge files as backing. |
| 795 | .RE |
| 796 | .P |
| 797 | The amount of memory allocated is the maximum allowed \fBblocksize\fR for the |
| 798 | job multiplied by \fBiodepth\fR. For \fBshmhuge\fR or \fBmmaphuge\fR to work, |
| 799 | the system must have free huge pages allocated. \fBmmaphuge\fR also needs to |
Jens Axboe | 2e266ba | 2009-09-14 08:56:53 +0200 | [diff] [blame] | 800 | have hugetlbfs mounted, and \fIfile\fR must point there. At least on Linux, |
| 801 | huge pages must be manually allocated. See \fB/proc/sys/vm/nr_hugehages\fR |
| 802 | and the documentation for that. Normally you just need to echo an appropriate |
| 803 | number, eg echoing 8 will ensure that the OS has 8 huge pages ready for |
| 804 | use. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 805 | .RE |
| 806 | .TP |
Jens Axboe | d392365 | 2011-08-03 12:38:39 +0200 | [diff] [blame] | 807 | .BI iomem_align \fR=\fPint "\fR,\fP mem_align" \fR=\fPint |
Jens Axboe | d529ee1 | 2009-07-01 10:33:03 +0200 | [diff] [blame] | 808 | This indiciates the memory alignment of the IO memory buffers. Note that the |
| 809 | given alignment is applied to the first IO unit buffer, if using \fBiodepth\fR |
| 810 | the alignment of the following buffers are given by the \fBbs\fR used. In |
| 811 | other words, if using a \fBbs\fR that is a multiple of the page sized in the |
| 812 | system, all buffers will be aligned to this value. If using a \fBbs\fR that |
| 813 | is not page aligned, the alignment of subsequent IO memory buffers is the |
| 814 | sum of the \fBiomem_align\fR and \fBbs\fR used. |
| 815 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 816 | .BI hugepage\-size \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 817 | Defines the size of a huge page. Must be at least equal to the system setting. |
Jens Axboe | b22989b | 2009-07-17 22:29:23 +0200 | [diff] [blame] | 818 | Should be a multiple of 1MB. Default: 4MB. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 819 | .TP |
| 820 | .B exitall |
| 821 | Terminate all jobs when one finishes. Default: wait for each job to finish. |
| 822 | .TP |
| 823 | .BI bwavgtime \fR=\fPint |
| 824 | Average bandwidth calculations over the given time in milliseconds. Default: |
| 825 | 500ms. |
| 826 | .TP |
Jens Axboe | c8eeb9d | 2011-10-05 14:02:22 +0200 | [diff] [blame] | 827 | .BI iopsavgtime \fR=\fPint |
| 828 | Average IOPS calculations over the given time in milliseconds. Default: |
| 829 | 500ms. |
| 830 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 831 | .BI create_serialize \fR=\fPbool |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 832 | If true, serialize file creation for the jobs. Default: true. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 833 | .TP |
| 834 | .BI create_fsync \fR=\fPbool |
| 835 | \fIfsync\fR\|(2) data file after creation. Default: true. |
| 836 | .TP |
Jens Axboe | 6b7f685 | 2009-03-09 14:22:56 +0100 | [diff] [blame] | 837 | .BI create_on_open \fR=\fPbool |
| 838 | If true, the files are not created until they are opened for IO by the job. |
| 839 | .TP |
Jens Axboe | 25460cf | 2012-05-02 13:58:02 +0200 | [diff] [blame] | 840 | .BI create_only \fR=\fPbool |
| 841 | If true, fio will only run the setup phase of the job. If files need to be |
| 842 | laid out or updated on disk, only that will be done. The actual job contents |
| 843 | are not executed. |
| 844 | .TP |
Jens Axboe | e9f4847 | 2009-06-03 12:14:08 +0200 | [diff] [blame] | 845 | .BI pre_read \fR=\fPbool |
| 846 | If this is given, files will be pre-read into memory before starting the given |
| 847 | IO operation. This will also clear the \fR \fBinvalidate\fR flag, since it is |
Jens Axboe | 9c0d224 | 2009-07-01 12:26:28 +0200 | [diff] [blame] | 848 | pointless to pre-read and then drop the cache. This will only work for IO |
| 849 | engines that are seekable, since they allow you to read the same data |
| 850 | multiple times. Thus it will not work on eg network or splice IO. |
Jens Axboe | e9f4847 | 2009-06-03 12:14:08 +0200 | [diff] [blame] | 851 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 852 | .BI unlink \fR=\fPbool |
| 853 | Unlink job files when done. Default: false. |
| 854 | .TP |
| 855 | .BI loops \fR=\fPint |
| 856 | Specifies the number of iterations (runs of the same workload) of this job. |
| 857 | Default: 1. |
| 858 | .TP |
| 859 | .BI do_verify \fR=\fPbool |
| 860 | Run the verify phase after a write phase. Only valid if \fBverify\fR is set. |
| 861 | Default: true. |
| 862 | .TP |
| 863 | .BI verify \fR=\fPstr |
| 864 | Method of verifying file contents after each iteration of the job. Allowed |
| 865 | values are: |
| 866 | .RS |
| 867 | .RS |
| 868 | .TP |
Jens Axboe | b892dc0 | 2009-09-05 20:37:35 +0200 | [diff] [blame] | 869 | .B md5 crc16 crc32 crc32c crc32c-intel crc64 crc7 sha256 sha512 sha1 |
Jens Axboe | 0539d75 | 2010-06-21 15:22:56 +0200 | [diff] [blame] | 870 | Store appropriate checksum in the header of each block. crc32c-intel is |
| 871 | hardware accelerated SSE4.2 driven, falls back to regular crc32c if |
| 872 | not supported by the system. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 873 | .TP |
| 874 | .B meta |
| 875 | Write extra information about each I/O (timestamp, block number, etc.). The |
Jens Axboe | 996093b | 2010-06-24 08:37:13 +0200 | [diff] [blame] | 876 | block number is verified. See \fBverify_pattern\fR as well. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 877 | .TP |
| 878 | .B null |
| 879 | Pretend to verify. Used for testing internals. |
| 880 | .RE |
Jens Axboe | b892dc0 | 2009-09-05 20:37:35 +0200 | [diff] [blame] | 881 | |
| 882 | This option can be used for repeated burn-in tests of a system to make sure |
| 883 | that the written data is also correctly read back. If the data direction given |
| 884 | is a read or random read, fio will assume that it should verify a previously |
| 885 | written file. If the data direction includes any form of write, the verify will |
| 886 | be of the newly written data. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 887 | .RE |
| 888 | .TP |
| 889 | .BI verify_sort \fR=\fPbool |
| 890 | If true, written verify blocks are sorted if \fBfio\fR deems it to be faster to |
| 891 | read them back in a sorted manner. Default: true. |
| 892 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 893 | .BI verify_offset \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 894 | Swap the verification header with data somewhere else in the block before |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 895 | writing. It is swapped back before verifying. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 896 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 897 | .BI verify_interval \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 898 | Write the verification header for this number of bytes, which should divide |
| 899 | \fBblocksize\fR. Default: \fBblocksize\fR. |
| 900 | .TP |
Jens Axboe | 996093b | 2010-06-24 08:37:13 +0200 | [diff] [blame] | 901 | .BI verify_pattern \fR=\fPstr |
| 902 | If set, fio will fill the io buffers with this pattern. Fio defaults to filling |
| 903 | with totally random bytes, but sometimes it's interesting to fill with a known |
| 904 | pattern for io verification purposes. Depending on the width of the pattern, |
| 905 | fio will fill 1/2/3/4 bytes of the buffer at the time(it can be either a |
| 906 | decimal or a hex number). The verify_pattern if larger than a 32-bit quantity |
| 907 | has to be a hex number that starts with either "0x" or "0X". Use with |
| 908 | \fBverify\fP=meta. |
| 909 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 910 | .BI verify_fatal \fR=\fPbool |
| 911 | If true, exit the job on the first observed verification failure. Default: |
| 912 | false. |
| 913 | .TP |
Jens Axboe | b463e93 | 2011-01-12 09:03:23 +0100 | [diff] [blame] | 914 | .BI verify_dump \fR=\fPbool |
| 915 | If set, dump the contents of both the original data block and the data block we |
| 916 | read off disk to files. This allows later analysis to inspect just what kind of |
Jens Axboe | ef71e31 | 2011-10-25 22:43:36 +0200 | [diff] [blame] | 917 | data corruption occurred. Off by default. |
Jens Axboe | b463e93 | 2011-01-12 09:03:23 +0100 | [diff] [blame] | 918 | .TP |
Jens Axboe | e8462bd | 2009-07-06 12:59:04 +0200 | [diff] [blame] | 919 | .BI verify_async \fR=\fPint |
| 920 | Fio will normally verify IO inline from the submitting thread. This option |
| 921 | takes an integer describing how many async offload threads to create for IO |
| 922 | verification instead, causing fio to offload the duty of verifying IO contents |
Jens Axboe | c85c324 | 2009-07-06 14:12:57 +0200 | [diff] [blame] | 923 | to one or more separate threads. If using this offload option, even sync IO |
| 924 | engines can benefit from using an \fBiodepth\fR setting higher than 1, as it |
| 925 | allows them to have IO in flight while verifies are running. |
Jens Axboe | e8462bd | 2009-07-06 12:59:04 +0200 | [diff] [blame] | 926 | .TP |
| 927 | .BI verify_async_cpus \fR=\fPstr |
| 928 | Tell fio to set the given CPU affinity on the async IO verification threads. |
| 929 | See \fBcpus_allowed\fP for the format used. |
| 930 | .TP |
Jens Axboe | 6f87418 | 2010-06-21 12:53:26 +0200 | [diff] [blame] | 931 | .BI verify_backlog \fR=\fPint |
| 932 | Fio will normally verify the written contents of a job that utilizes verify |
| 933 | once that job has completed. In other words, everything is written then |
| 934 | everything is read back and verified. You may want to verify continually |
| 935 | instead for a variety of reasons. Fio stores the meta data associated with an |
| 936 | IO block in memory, so for large verify workloads, quite a bit of memory would |
David Nellans | 092f707 | 2010-10-26 08:08:42 -0600 | [diff] [blame] | 937 | be used up holding this meta data. If this option is enabled, fio will write |
| 938 | only N blocks before verifying these blocks. |
Jens Axboe | 6f87418 | 2010-06-21 12:53:26 +0200 | [diff] [blame] | 939 | .TP |
| 940 | .BI verify_backlog_batch \fR=\fPint |
| 941 | Control how many blocks fio will verify if verify_backlog is set. If not set, |
| 942 | will default to the value of \fBverify_backlog\fR (meaning the entire queue is |
David Nellans | 092f707 | 2010-10-26 08:08:42 -0600 | [diff] [blame] | 943 | read back and verified). If \fBverify_backlog_batch\fR is less than |
| 944 | \fBverify_backlog\fR then not all blocks will be verified, if |
| 945 | \fBverify_backlog_batch\fR is larger than \fBverify_backlog\fR, some blocks |
| 946 | will be verified more than once. |
Jens Axboe | 6f87418 | 2010-06-21 12:53:26 +0200 | [diff] [blame] | 947 | .TP |
Jens Axboe | d392365 | 2011-08-03 12:38:39 +0200 | [diff] [blame] | 948 | .B stonewall "\fR,\fP wait_for_previous" |
Martin Steigerwald | 5982a92 | 2011-06-27 16:07:24 +0200 | [diff] [blame] | 949 | Wait for preceding jobs in the job file to exit before starting this one. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 950 | \fBstonewall\fR implies \fBnew_group\fR. |
| 951 | .TP |
| 952 | .B new_group |
| 953 | Start a new reporting group. If not given, all jobs in a file will be part |
| 954 | of the same reporting group, unless separated by a stonewall. |
| 955 | .TP |
| 956 | .BI numjobs \fR=\fPint |
| 957 | Number of clones (processes/threads performing the same workload) of this job. |
| 958 | Default: 1. |
| 959 | .TP |
| 960 | .B group_reporting |
| 961 | If set, display per-group reports instead of per-job when \fBnumjobs\fR is |
| 962 | specified. |
| 963 | .TP |
| 964 | .B thread |
| 965 | Use threads created with \fBpthread_create\fR\|(3) instead of processes created |
| 966 | with \fBfork\fR\|(2). |
| 967 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 968 | .BI zonesize \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 969 | Divide file into zones of the specified size in bytes. See \fBzoneskip\fR. |
| 970 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 971 | .BI zoneskip \fR=\fPint |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 972 | Skip the specified number of bytes when \fBzonesize\fR bytes of data have been |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 973 | read. |
| 974 | .TP |
| 975 | .BI write_iolog \fR=\fPstr |
Stefan Hajnoczi | 5b42a48 | 2011-01-08 20:28:41 +0100 | [diff] [blame] | 976 | Write the issued I/O patterns to the specified file. Specify a separate file |
| 977 | for each job, otherwise the iologs will be interspersed and the file may be |
| 978 | corrupt. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 979 | .TP |
| 980 | .BI read_iolog \fR=\fPstr |
| 981 | Replay the I/O patterns contained in the specified file generated by |
| 982 | \fBwrite_iolog\fR, or may be a \fBblktrace\fR binary file. |
| 983 | .TP |
David Nellans | 64bbb86 | 2010-08-24 22:13:30 +0200 | [diff] [blame] | 984 | .BI replay_no_stall \fR=\fPint |
| 985 | While replaying I/O patterns using \fBread_iolog\fR the default behavior |
| 986 | attempts to respect timing information between I/Os. Enabling |
| 987 | \fBreplay_no_stall\fR causes I/Os to be replayed as fast as possible while |
| 988 | still respecting ordering. |
| 989 | .TP |
David Nellans | d1c46c0 | 2010-08-31 21:20:47 +0200 | [diff] [blame] | 990 | .BI replay_redirect \fR=\fPstr |
| 991 | While replaying I/O patterns using \fBread_iolog\fR the default behavior |
| 992 | is to replay the IOPS onto the major/minor device that each IOP was recorded |
| 993 | from. Setting \fBreplay_redirect\fR causes all IOPS to be replayed onto the |
| 994 | single specified device regardless of the device it was recorded from. |
| 995 | .TP |
Steven Noonan | 836bad5 | 2011-09-14 09:21:33 +0200 | [diff] [blame] | 996 | .BI write_bw_log \fR=\fPstr |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 997 | If given, write a bandwidth log of the jobs in this job file. Can be used to |
| 998 | store data of the bandwidth of the jobs in their lifetime. The included |
| 999 | fio_generate_plots script uses gnuplot to turn these text files into nice |
| 1000 | graphs. See \fBwrite_log_log\fR for behaviour of given filename. For this |
| 1001 | option, the postfix is _bw.log. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1002 | .TP |
Steven Noonan | 836bad5 | 2011-09-14 09:21:33 +0200 | [diff] [blame] | 1003 | .BI write_lat_log \fR=\fPstr |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 1004 | Same as \fBwrite_bw_log\fR, but writes I/O completion latencies. If no |
| 1005 | filename is given with this option, the default filename of "jobname_type.log" |
| 1006 | is used. Even if the filename is given, fio will still append the type of log. |
| 1007 | .TP |
Jens Axboe | c8eeb9d | 2011-10-05 14:02:22 +0200 | [diff] [blame] | 1008 | .BI write_iops_log \fR=\fPstr |
| 1009 | Same as \fBwrite_bw_log\fR, but writes IOPS. If no filename is given with this |
| 1010 | option, the default filename of "jobname_type.log" is used. Even if the |
| 1011 | filename is given, fio will still append the type of log. |
| 1012 | .TP |
Jens Axboe | b8bc8cb | 2011-12-01 09:04:31 +0100 | [diff] [blame] | 1013 | .BI log_avg_msec \fR=\fPint |
| 1014 | By default, fio will log an entry in the iops, latency, or bw log for every |
| 1015 | IO that completes. When writing to the disk log, that can quickly grow to a |
| 1016 | very large size. Setting this option makes fio average the each log entry |
| 1017 | over the specified period of time, reducing the resolution of the log. |
| 1018 | Defaults to 0. |
| 1019 | .TP |
Steven Noonan | 836bad5 | 2011-09-14 09:21:33 +0200 | [diff] [blame] | 1020 | .BI disable_lat \fR=\fPbool |
Jens Axboe | 02af098 | 2010-06-24 09:59:34 +0200 | [diff] [blame] | 1021 | Disable measurements of total latency numbers. Useful only for cutting |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 1022 | back the number of calls to gettimeofday, as that does impact performance at |
| 1023 | really high IOPS rates. Note that to really get rid of a large amount of these |
| 1024 | calls, this option must be used with disable_slat and disable_bw as well. |
| 1025 | .TP |
Steven Noonan | 836bad5 | 2011-09-14 09:21:33 +0200 | [diff] [blame] | 1026 | .BI disable_clat \fR=\fPbool |
Steven Noonan | c95f9da | 2011-06-22 09:47:09 +0200 | [diff] [blame] | 1027 | Disable measurements of completion latency numbers. See \fBdisable_lat\fR. |
Jens Axboe | 02af098 | 2010-06-24 09:59:34 +0200 | [diff] [blame] | 1028 | .TP |
Steven Noonan | 836bad5 | 2011-09-14 09:21:33 +0200 | [diff] [blame] | 1029 | .BI disable_slat \fR=\fPbool |
Jens Axboe | 02af098 | 2010-06-24 09:59:34 +0200 | [diff] [blame] | 1030 | Disable measurements of submission latency numbers. See \fBdisable_lat\fR. |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 1031 | .TP |
Steven Noonan | 836bad5 | 2011-09-14 09:21:33 +0200 | [diff] [blame] | 1032 | .BI disable_bw_measurement \fR=\fPbool |
Jens Axboe | 02af098 | 2010-06-24 09:59:34 +0200 | [diff] [blame] | 1033 | Disable measurements of throughput/bandwidth numbers. See \fBdisable_lat\fR. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1034 | .TP |
Jens Axboe | f7fa265 | 2009-03-09 14:20:20 +0100 | [diff] [blame] | 1035 | .BI lockmem \fR=\fPint |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1036 | Pin the specified amount of memory with \fBmlock\fR\|(2). Can be used to |
Jens Axboe | 81c6b6c | 2013-04-10 19:30:50 +0200 | [diff] [blame] | 1037 | simulate a smaller amount of memory. The amount specified is per worker. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1038 | .TP |
| 1039 | .BI exec_prerun \fR=\fPstr |
| 1040 | Before running the job, execute the specified command with \fBsystem\fR\|(3). |
| 1041 | .TP |
| 1042 | .BI exec_postrun \fR=\fPstr |
| 1043 | Same as \fBexec_prerun\fR, but the command is executed after the job completes. |
| 1044 | .TP |
| 1045 | .BI ioscheduler \fR=\fPstr |
| 1046 | Attempt to switch the device hosting the file to the specified I/O scheduler. |
| 1047 | .TP |
| 1048 | .BI cpuload \fR=\fPint |
| 1049 | If the job is a CPU cycle-eater, attempt to use the specified percentage of |
| 1050 | CPU cycles. |
| 1051 | .TP |
| 1052 | .BI cpuchunks \fR=\fPint |
| 1053 | If the job is a CPU cycle-eater, split the load into cycles of the |
| 1054 | given time in milliseconds. |
| 1055 | .TP |
| 1056 | .BI disk_util \fR=\fPbool |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1057 | Generate disk utilization statistics if the platform supports it. Default: true. |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 1058 | .TP |
Jens Axboe | 2389364 | 2012-12-17 14:44:08 +0100 | [diff] [blame] | 1059 | .BI clocksource \fR=\fPstr |
| 1060 | Use the given clocksource as the base of timing. The supported options are: |
| 1061 | .RS |
| 1062 | .TP |
| 1063 | .B gettimeofday |
| 1064 | gettimeofday(2) |
| 1065 | .TP |
| 1066 | .B clock_gettime |
| 1067 | clock_gettime(2) |
| 1068 | .TP |
| 1069 | .B cpu |
| 1070 | Internal CPU clock source |
| 1071 | .TP |
| 1072 | .RE |
| 1073 | .P |
| 1074 | \fBcpu\fR is the preferred clocksource if it is reliable, as it is very fast |
| 1075 | (and fio is heavy on time calls). Fio will automatically use this clocksource |
| 1076 | if it's supported and considered reliable on the system it is running on, |
| 1077 | unless another clocksource is specifically set. For x86/x86-64 CPUs, this |
| 1078 | means supporting TSC Invariant. |
| 1079 | .TP |
Jens Axboe | 901bb99 | 2009-03-14 20:17:36 +0100 | [diff] [blame] | 1080 | .BI gtod_reduce \fR=\fPbool |
| 1081 | Enable all of the gettimeofday() reducing options (disable_clat, disable_slat, |
| 1082 | disable_bw) plus reduce precision of the timeout somewhat to really shrink the |
| 1083 | gettimeofday() call count. With this option enabled, we only do about 0.4% of |
| 1084 | the gtod() calls we would have done if all time keeping was enabled. |
| 1085 | .TP |
| 1086 | .BI gtod_cpu \fR=\fPint |
| 1087 | Sometimes it's cheaper to dedicate a single thread of execution to just getting |
| 1088 | the current time. Fio (and databases, for instance) are very intensive on |
| 1089 | gettimeofday() calls. With this option, you can set one CPU aside for doing |
| 1090 | nothing but logging current time to a shared memory location. Then the other |
| 1091 | threads/processes that run IO workloads need only copy that segment, instead of |
| 1092 | entering the kernel with a gettimeofday() call. The CPU set aside for doing |
| 1093 | these time calls will be excluded from other uses. Fio will manually clear it |
| 1094 | from the CPU mask of other jobs. |
Radha Ramachandran | f2bba18 | 2009-06-15 08:40:16 +0200 | [diff] [blame] | 1095 | .TP |
Dmitry Monakhov | 8b28bd4 | 2012-09-23 15:46:09 +0400 | [diff] [blame] | 1096 | .BI ignore_error \fR=\fPstr |
| 1097 | Sometimes you want to ignore some errors during test in that case you can specify |
| 1098 | error list for each error type. |
| 1099 | .br |
| 1100 | ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST |
| 1101 | .br |
| 1102 | errors for given error type is separated with ':'. |
| 1103 | Error may be symbol ('ENOSPC', 'ENOMEM') or an integer. |
| 1104 | .br |
| 1105 | Example: ignore_error=EAGAIN,ENOSPC:122 . |
| 1106 | .br |
| 1107 | This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from WRITE. |
| 1108 | .TP |
| 1109 | .BI error_dump \fR=\fPbool |
| 1110 | If set dump every error even if it is non fatal, true by default. If disabled |
| 1111 | only fatal error will be dumped |
| 1112 | .TP |
Jens Axboe | a696fa2 | 2009-12-04 10:05:02 +0100 | [diff] [blame] | 1113 | .BI cgroup \fR=\fPstr |
| 1114 | Add job to this control group. If it doesn't exist, it will be created. |
Jens Axboe | 6adb38a | 2009-12-07 08:01:26 +0100 | [diff] [blame] | 1115 | The system must have a mounted cgroup blkio mount point for this to work. If |
| 1116 | your system doesn't have it mounted, you can do so with: |
| 1117 | |
Martin Steigerwald | 5982a92 | 2011-06-27 16:07:24 +0200 | [diff] [blame] | 1118 | # mount \-t cgroup \-o blkio none /cgroup |
Jens Axboe | a696fa2 | 2009-12-04 10:05:02 +0100 | [diff] [blame] | 1119 | .TP |
| 1120 | .BI cgroup_weight \fR=\fPint |
| 1121 | Set the weight of the cgroup to this value. See the documentation that comes |
| 1122 | with the kernel, allowed values are in the range of 100..1000. |
Jens Axboe | e0b0d89 | 2009-12-08 10:10:14 +0100 | [diff] [blame] | 1123 | .TP |
Vivek Goyal | 7de8709 | 2010-03-31 22:55:15 +0200 | [diff] [blame] | 1124 | .BI cgroup_nodelete \fR=\fPbool |
| 1125 | Normally fio will delete the cgroups it has created after the job completion. |
| 1126 | To override this behavior and to leave cgroups around after the job completion, |
| 1127 | set cgroup_nodelete=1. This can be useful if one wants to inspect various |
| 1128 | cgroup files after job completion. Default: false |
| 1129 | .TP |
Jens Axboe | e0b0d89 | 2009-12-08 10:10:14 +0100 | [diff] [blame] | 1130 | .BI uid \fR=\fPint |
| 1131 | Instead of running as the invoking user, set the user ID to this value before |
| 1132 | the thread/process does any work. |
| 1133 | .TP |
| 1134 | .BI gid \fR=\fPint |
| 1135 | Set group ID, see \fBuid\fR. |
Yu-ju Hong | 8334919 | 2011-08-13 00:53:44 +0200 | [diff] [blame] | 1136 | .TP |
Dan Ehrenberg | 9e684a4 | 2012-02-20 11:05:14 +0100 | [diff] [blame] | 1137 | .BI flow_id \fR=\fPint |
| 1138 | The ID of the flow. If not specified, it defaults to being a global flow. See |
| 1139 | \fBflow\fR. |
| 1140 | .TP |
| 1141 | .BI flow \fR=\fPint |
| 1142 | Weight in token-based flow control. If this value is used, then there is a |
| 1143 | \fBflow counter\fR which is used to regulate the proportion of activity between |
| 1144 | two or more jobs. fio attempts to keep this flow counter near zero. The |
| 1145 | \fBflow\fR parameter stands for how much should be added or subtracted to the |
| 1146 | flow counter on each iteration of the main I/O loop. That is, if one job has |
| 1147 | \fBflow=8\fR and another job has \fBflow=-1\fR, then there will be a roughly |
| 1148 | 1:8 ratio in how much one runs vs the other. |
| 1149 | .TP |
| 1150 | .BI flow_watermark \fR=\fPint |
| 1151 | The maximum value that the absolute value of the flow counter is allowed to |
| 1152 | reach before the job must wait for a lower value of the counter. |
| 1153 | .TP |
| 1154 | .BI flow_sleep \fR=\fPint |
| 1155 | The period of time, in microseconds, to wait after the flow watermark has been |
| 1156 | exceeded before retrying operations |
| 1157 | .TP |
Yu-ju Hong | 8334919 | 2011-08-13 00:53:44 +0200 | [diff] [blame] | 1158 | .BI clat_percentiles \fR=\fPbool |
| 1159 | Enable the reporting of percentiles of completion latencies. |
| 1160 | .TP |
| 1161 | .BI percentile_list \fR=\fPfloat_list |
| 1162 | Overwrite the default list of percentiles for completion |
| 1163 | latencies. Each number is a floating number in the range (0,100], and |
| 1164 | the maximum length of the list is 20. Use ':' to separate the |
Martin Steigerwald | 3eb0728 | 2011-10-05 11:41:54 +0200 | [diff] [blame] | 1165 | numbers. For example, \-\-percentile_list=99.5:99.9 will cause fio to |
Yu-ju Hong | 8334919 | 2011-08-13 00:53:44 +0200 | [diff] [blame] | 1166 | report the values of completion latency below which 99.5% and 99.9% of |
| 1167 | the observed latencies fell, respectively. |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 1168 | .SS "Ioengine Parameters List" |
| 1169 | Some parameters are only valid when a specific ioengine is in use. These are |
| 1170 | used identically to normal parameters, with the caveat that when used on the |
| 1171 | command line, the must come after the ioengine that defines them is selected. |
| 1172 | .TP |
Jens Axboe | e458593 | 2013-04-10 22:16:01 +0200 | [diff] [blame] | 1173 | .BI (cpu)cpuload \fR=\fPint |
| 1174 | Attempt to use the specified percentage of CPU cycles. |
| 1175 | .TP |
| 1176 | .BI (cpu)cpuchunks \fR=\fPint |
| 1177 | Split the load into cycles of the given time. In microseconds. |
| 1178 | .TP |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 1179 | .BI (libaio)userspace_reap |
| 1180 | Normally, with the libaio engine in use, fio will use |
| 1181 | the io_getevents system call to reap newly returned events. |
| 1182 | With this flag turned on, the AIO ring will be read directly |
| 1183 | from user-space to reap events. The reaping mode is only |
| 1184 | enabled when polling for a minimum of 0 events (eg when |
| 1185 | iodepth_batch_complete=0). |
| 1186 | .TP |
| 1187 | .BI (net,netsplice)hostname \fR=\fPstr |
| 1188 | The host name or IP address to use for TCP or UDP based IO. |
| 1189 | If the job is a TCP listener or UDP reader, the hostname is not |
| 1190 | used and must be omitted. |
| 1191 | .TP |
| 1192 | .BI (net,netsplice)port \fR=\fPint |
| 1193 | The TCP or UDP port to bind to or connect to. |
| 1194 | .TP |
Jens Axboe | 1d360ff | 2013-01-31 13:33:45 +0100 | [diff] [blame] | 1195 | .BI (net,netsplice)nodelay \fR=\fPbool |
| 1196 | Set TCP_NODELAY on TCP connections. |
| 1197 | .TP |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 1198 | .BI (net,netsplice)protocol \fR=\fPstr "\fR,\fP proto" \fR=\fPstr |
| 1199 | The network protocol to use. Accepted values are: |
| 1200 | .RS |
| 1201 | .RS |
| 1202 | .TP |
| 1203 | .B tcp |
| 1204 | Transmission control protocol |
| 1205 | .TP |
| 1206 | .B udp |
Bruce Cran | f5cc3d0 | 2012-10-10 08:17:44 -0600 | [diff] [blame] | 1207 | User datagram protocol |
Steven Lang | de890a1 | 2011-11-09 14:03:34 +0100 | [diff] [blame] | 1208 | .TP |
| 1209 | .B unix |
| 1210 | UNIX domain socket |
| 1211 | .RE |
| 1212 | .P |
| 1213 | When the protocol is TCP or UDP, the port must also be given, |
| 1214 | as well as the hostname if the job is a TCP listener or UDP |
| 1215 | reader. For unix sockets, the normal filename option should be |
| 1216 | used and the port is invalid. |
| 1217 | .RE |
| 1218 | .TP |
| 1219 | .BI (net,netsplice)listen |
| 1220 | For TCP network connections, tell fio to listen for incoming |
| 1221 | connections rather than initiating an outgoing connection. The |
| 1222 | hostname must be omitted if this option is used. |
Dmitry Monakhov | d54fce8 | 2012-09-20 15:37:17 +0400 | [diff] [blame] | 1223 | .TP |
Jens Axboe | 7aeb1e9 | 2012-12-06 20:53:57 +0100 | [diff] [blame] | 1224 | .BI (net, pingpong) \fR=\fPbool |
| 1225 | Normal a network writer will just continue writing data, and a network reader |
| 1226 | will just consume packages. If pingpong=1 is set, a writer will send its normal |
| 1227 | payload to the reader, then wait for the reader to send the same payload back. |
| 1228 | This allows fio to measure network latencies. The submission and completion |
| 1229 | latencies then measure local time spent sending or receiving, and the |
| 1230 | completion latency measures how long it took for the other end to receive and |
| 1231 | send back. |
| 1232 | .TP |
Dmitry Monakhov | d54fce8 | 2012-09-20 15:37:17 +0400 | [diff] [blame] | 1233 | .BI (e4defrag,donorname) \fR=\fPstr |
| 1234 | File will be used as a block donor (swap extents between files) |
| 1235 | .TP |
| 1236 | .BI (e4defrag,inplace) \fR=\fPint |
| 1237 | Configure donor file block allocation strategy |
| 1238 | .RS |
| 1239 | .BI 0(default) : |
| 1240 | Preallocate donor's file on init |
| 1241 | .TP |
| 1242 | .BI 1: |
| 1243 | allocate space immidietly inside defragment event, and free right after event |
| 1244 | .RE |
| 1245 | .TP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1246 | .SH OUTPUT |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1247 | While running, \fBfio\fR will display the status of the created jobs. For |
| 1248 | example: |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1249 | .RS |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1250 | .P |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1251 | Threads: 1: [_r] [24.8% done] [ 13509/ 8334 kb/s] [eta 00h:01m:31s] |
| 1252 | .RE |
| 1253 | .P |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1254 | The characters in the first set of brackets denote the current status of each |
| 1255 | threads. The possible values are: |
| 1256 | .P |
| 1257 | .PD 0 |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1258 | .RS |
| 1259 | .TP |
| 1260 | .B P |
| 1261 | Setup but not started. |
| 1262 | .TP |
| 1263 | .B C |
| 1264 | Thread created. |
| 1265 | .TP |
| 1266 | .B I |
| 1267 | Initialized, waiting. |
| 1268 | .TP |
| 1269 | .B R |
| 1270 | Running, doing sequential reads. |
| 1271 | .TP |
| 1272 | .B r |
| 1273 | Running, doing random reads. |
| 1274 | .TP |
| 1275 | .B W |
| 1276 | Running, doing sequential writes. |
| 1277 | .TP |
| 1278 | .B w |
| 1279 | Running, doing random writes. |
| 1280 | .TP |
| 1281 | .B M |
| 1282 | Running, doing mixed sequential reads/writes. |
| 1283 | .TP |
| 1284 | .B m |
| 1285 | Running, doing mixed random reads/writes. |
| 1286 | .TP |
| 1287 | .B F |
| 1288 | Running, currently waiting for \fBfsync\fR\|(2). |
| 1289 | .TP |
| 1290 | .B V |
| 1291 | Running, verifying written data. |
| 1292 | .TP |
| 1293 | .B E |
| 1294 | Exited, not reaped by main thread. |
| 1295 | .TP |
| 1296 | .B \- |
| 1297 | Exited, thread reaped. |
| 1298 | .RE |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1299 | .PD |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1300 | .P |
| 1301 | The second set of brackets shows the estimated completion percentage of |
| 1302 | the current group. The third set shows the read and write I/O rate, |
| 1303 | respectively. Finally, the estimated run time of the job is displayed. |
| 1304 | .P |
| 1305 | When \fBfio\fR completes (or is interrupted by Ctrl-C), it will show data |
| 1306 | for each thread, each group of threads, and each disk, in that order. |
| 1307 | .P |
| 1308 | Per-thread statistics first show the threads client number, group-id, and |
| 1309 | error code. The remaining figures are as follows: |
| 1310 | .RS |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1311 | .TP |
| 1312 | .B io |
| 1313 | Number of megabytes of I/O performed. |
| 1314 | .TP |
| 1315 | .B bw |
| 1316 | Average data rate (bandwidth). |
| 1317 | .TP |
| 1318 | .B runt |
| 1319 | Threads run time. |
| 1320 | .TP |
| 1321 | .B slat |
| 1322 | Submission latency minimum, maximum, average and standard deviation. This is |
| 1323 | the time it took to submit the I/O. |
| 1324 | .TP |
| 1325 | .B clat |
| 1326 | Completion latency minimum, maximum, average and standard deviation. This |
| 1327 | is the time between submission and completion. |
| 1328 | .TP |
| 1329 | .B bw |
| 1330 | Bandwidth minimum, maximum, percentage of aggregate bandwidth received, average |
| 1331 | and standard deviation. |
| 1332 | .TP |
| 1333 | .B cpu |
| 1334 | CPU usage statistics. Includes user and system time, number of context switches |
| 1335 | this thread went through and number of major and minor page faults. |
| 1336 | .TP |
| 1337 | .B IO depths |
| 1338 | Distribution of I/O depths. Each depth includes everything less than (or equal) |
| 1339 | to it, but greater than the previous depth. |
| 1340 | .TP |
| 1341 | .B IO issued |
| 1342 | Number of read/write requests issued, and number of short read/write requests. |
| 1343 | .TP |
| 1344 | .B IO latencies |
| 1345 | Distribution of I/O completion latencies. The numbers follow the same pattern |
| 1346 | as \fBIO depths\fR. |
| 1347 | .RE |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1348 | .P |
| 1349 | The group statistics show: |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1350 | .PD 0 |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1351 | .RS |
| 1352 | .TP |
| 1353 | .B io |
| 1354 | Number of megabytes I/O performed. |
| 1355 | .TP |
| 1356 | .B aggrb |
| 1357 | Aggregate bandwidth of threads in the group. |
| 1358 | .TP |
| 1359 | .B minb |
| 1360 | Minimum average bandwidth a thread saw. |
| 1361 | .TP |
| 1362 | .B maxb |
| 1363 | Maximum average bandwidth a thread saw. |
| 1364 | .TP |
| 1365 | .B mint |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1366 | Shortest runtime of threads in the group. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1367 | .TP |
| 1368 | .B maxt |
| 1369 | Longest runtime of threads in the group. |
| 1370 | .RE |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1371 | .PD |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1372 | .P |
| 1373 | Finally, disk statistics are printed with reads first: |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1374 | .PD 0 |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1375 | .RS |
| 1376 | .TP |
| 1377 | .B ios |
| 1378 | Number of I/Os performed by all groups. |
| 1379 | .TP |
| 1380 | .B merge |
| 1381 | Number of merges in the I/O scheduler. |
| 1382 | .TP |
| 1383 | .B ticks |
| 1384 | Number of ticks we kept the disk busy. |
| 1385 | .TP |
| 1386 | .B io_queue |
| 1387 | Total time spent in the disk queue. |
| 1388 | .TP |
| 1389 | .B util |
| 1390 | Disk utilization. |
| 1391 | .RE |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1392 | .PD |
Jens Axboe | 8423bd1 | 2012-04-12 09:18:38 +0200 | [diff] [blame] | 1393 | .P |
| 1394 | It is also possible to get fio to dump the current output while it is |
| 1395 | running, without terminating the job. To do that, send fio the \fBUSR1\fR |
| 1396 | signal. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1397 | .SH TERSE OUTPUT |
| 1398 | If the \fB\-\-minimal\fR option is given, the results will be printed in a |
David Nellans | 562c2d2 | 2010-09-23 08:38:17 +0200 | [diff] [blame] | 1399 | semicolon-delimited format suitable for scripted use - a job description |
| 1400 | (if provided) follows on a new line. Note that the first |
Jens Axboe | 525c2bf | 2010-06-30 15:22:21 +0200 | [diff] [blame] | 1401 | number in the line is the version number. If the output has to be changed |
| 1402 | for some reason, this number will be incremented by 1 to signify that |
| 1403 | change. The fields are: |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1404 | .P |
| 1405 | .RS |
Jens Axboe | 5e726d0 | 2011-10-14 08:08:10 +0200 | [diff] [blame] | 1406 | .B terse version, fio version, jobname, groupid, error |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1407 | .P |
| 1408 | Read status: |
| 1409 | .RS |
Jens Axboe | 312b4af | 2011-10-13 13:11:42 +0200 | [diff] [blame] | 1410 | .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1411 | .P |
| 1412 | Submission latency: |
| 1413 | .RS |
| 1414 | .B min, max, mean, standard deviation |
| 1415 | .RE |
| 1416 | Completion latency: |
| 1417 | .RS |
| 1418 | .B min, max, mean, standard deviation |
| 1419 | .RE |
Jens Axboe | 1db92cb | 2011-10-13 13:43:36 +0200 | [diff] [blame] | 1420 | Completion latency percentiles (20 fields): |
| 1421 | .RS |
| 1422 | .B Xth percentile=usec |
| 1423 | .RE |
Jens Axboe | 525c2bf | 2010-06-30 15:22:21 +0200 | [diff] [blame] | 1424 | Total latency: |
| 1425 | .RS |
| 1426 | .B min, max, mean, standard deviation |
| 1427 | .RE |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1428 | Bandwidth: |
| 1429 | .RS |
| 1430 | .B min, max, aggregate percentage of total, mean, standard deviation |
| 1431 | .RE |
| 1432 | .RE |
| 1433 | .P |
| 1434 | Write status: |
| 1435 | .RS |
Jens Axboe | 312b4af | 2011-10-13 13:11:42 +0200 | [diff] [blame] | 1436 | .B Total I/O \fR(KB)\fP, bandwidth \fR(KB/s)\fP, IOPS, runtime \fR(ms)\fP |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1437 | .P |
| 1438 | Submission latency: |
| 1439 | .RS |
| 1440 | .B min, max, mean, standard deviation |
| 1441 | .RE |
| 1442 | Completion latency: |
| 1443 | .RS |
| 1444 | .B min, max, mean, standard deviation |
| 1445 | .RE |
Jens Axboe | 1db92cb | 2011-10-13 13:43:36 +0200 | [diff] [blame] | 1446 | Completion latency percentiles (20 fields): |
| 1447 | .RS |
| 1448 | .B Xth percentile=usec |
| 1449 | .RE |
Jens Axboe | 525c2bf | 2010-06-30 15:22:21 +0200 | [diff] [blame] | 1450 | Total latency: |
| 1451 | .RS |
| 1452 | .B min, max, mean, standard deviation |
| 1453 | .RE |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1454 | Bandwidth: |
| 1455 | .RS |
| 1456 | .B min, max, aggregate percentage of total, mean, standard deviation |
| 1457 | .RE |
| 1458 | .RE |
| 1459 | .P |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1460 | CPU usage: |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1461 | .RS |
Carl Henrik Lunde | bd2626f | 2008-06-12 09:17:46 +0200 | [diff] [blame] | 1462 | .B user, system, context switches, major page faults, minor page faults |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1463 | .RE |
| 1464 | .P |
| 1465 | IO depth distribution: |
| 1466 | .RS |
| 1467 | .B <=1, 2, 4, 8, 16, 32, >=64 |
| 1468 | .RE |
| 1469 | .P |
David Nellans | 562c2d2 | 2010-09-23 08:38:17 +0200 | [diff] [blame] | 1470 | IO latency distribution: |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1471 | .RS |
David Nellans | 562c2d2 | 2010-09-23 08:38:17 +0200 | [diff] [blame] | 1472 | Microseconds: |
| 1473 | .RS |
| 1474 | .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000 |
| 1475 | .RE |
| 1476 | Milliseconds: |
| 1477 | .RS |
| 1478 | .B <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000 |
| 1479 | .RE |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1480 | .RE |
| 1481 | .P |
Jens Axboe | f2f788d | 2011-10-13 14:03:52 +0200 | [diff] [blame] | 1482 | Disk utilization (1 for each disk used): |
| 1483 | .RS |
| 1484 | .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 |
| 1485 | .RE |
| 1486 | .P |
Martin Steigerwald | 5982a92 | 2011-06-27 16:07:24 +0200 | [diff] [blame] | 1487 | Error Info (dependent on continue_on_error, default off): |
David Nellans | 562c2d2 | 2010-09-23 08:38:17 +0200 | [diff] [blame] | 1488 | .RS |
| 1489 | .B total # errors, first error code |
| 1490 | .RE |
| 1491 | .P |
| 1492 | .B text description (if provided in config - appears on newline) |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1493 | .RE |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1494 | .SH CLIENT / SERVER |
| 1495 | Normally you would run fio as a stand-alone application on the machine |
| 1496 | where the IO workload should be generated. However, it is also possible to |
| 1497 | run the frontend and backend of fio separately. This makes it possible to |
| 1498 | have a fio server running on the machine(s) where the IO workload should |
| 1499 | be running, while controlling it from another machine. |
| 1500 | |
| 1501 | To start the server, you would do: |
| 1502 | |
| 1503 | \fBfio \-\-server=args\fR |
| 1504 | |
| 1505 | on that machine, where args defines what fio listens to. The arguments |
Jens Axboe | 811826b | 2011-10-24 09:11:50 +0200 | [diff] [blame] | 1506 | are of the form 'type:hostname or IP:port'. 'type' is either 'ip' (or ip4) |
Martin Steigerwald | 20c67f1 | 2012-05-07 17:06:26 +0200 | [diff] [blame] | 1507 | for TCP/IP v4, 'ip6' for TCP/IP v6, or 'sock' for a local unix domain |
| 1508 | socket. 'hostname' is either a hostname or IP address, and 'port' is the port to |
Jens Axboe | 811826b | 2011-10-24 09:11:50 +0200 | [diff] [blame] | 1509 | listen to (only valid for TCP/IP, not a local socket). Some examples: |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1510 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1511 | 1) fio \-\-server |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1512 | |
| 1513 | Start a fio server, listening on all interfaces on the default port (8765). |
| 1514 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1515 | 2) fio \-\-server=ip:hostname,4444 |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1516 | |
| 1517 | Start a fio server, listening on IP belonging to hostname and on port 4444. |
| 1518 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1519 | 3) fio \-\-server=ip6:::1,4444 |
Jens Axboe | 811826b | 2011-10-24 09:11:50 +0200 | [diff] [blame] | 1520 | |
| 1521 | Start a fio server, listening on IPv6 localhost ::1 and on port 4444. |
| 1522 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1523 | 4) fio \-\-server=,4444 |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1524 | |
| 1525 | Start a fio server, listening on all interfaces on port 4444. |
| 1526 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1527 | 5) fio \-\-server=1.2.3.4 |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1528 | |
| 1529 | Start a fio server, listening on IP 1.2.3.4 on the default port. |
| 1530 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1531 | 6) fio \-\-server=sock:/tmp/fio.sock |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1532 | |
| 1533 | Start a fio server, listening on the local socket /tmp/fio.sock. |
| 1534 | |
| 1535 | When a server is running, you can connect to it from a client. The client |
| 1536 | is run with: |
| 1537 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1538 | fio \-\-local-args \-\-client=server \-\-remote-args <job file(s)> |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1539 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1540 | where \-\-local-args are arguments that are local to the client where it is |
| 1541 | running, 'server' is the connect string, and \-\-remote-args and <job file(s)> |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1542 | are sent to the server. The 'server' string follows the same format as it |
| 1543 | does on the server side, to allow IP/hostname/socket and port strings. |
| 1544 | You can connect to multiple clients as well, to do that you could run: |
| 1545 | |
Martin Steigerwald | e01e974 | 2012-05-07 17:06:54 +0200 | [diff] [blame] | 1546 | fio \-\-client=server2 \-\-client=server2 <job file(s)> |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1547 | .SH AUTHORS |
Jens Axboe | 49da124 | 2011-10-13 20:17:02 +0200 | [diff] [blame] | 1548 | |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1549 | .B fio |
Jens Axboe | aa58d25 | 2010-06-09 09:49:38 +0200 | [diff] [blame] | 1550 | was written by Jens Axboe <jens.axboe@oracle.com>, |
| 1551 | now Jens Axboe <jaxboe@fusionio.com>. |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1552 | .br |
| 1553 | This man page was written by Aaron Carroll <aaronc@cse.unsw.edu.au> based |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1554 | on documentation by Jens Axboe. |
| 1555 | .SH "REPORTING BUGS" |
Jens Axboe | 482900c | 2009-06-02 12:15:51 +0200 | [diff] [blame] | 1556 | Report bugs to the \fBfio\fR mailing list <fio@vger.kernel.org>. |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1557 | See \fBREADME\fR. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1558 | .SH "SEE ALSO" |
Aaron Carroll | d1429b5 | 2007-09-18 08:10:57 +0200 | [diff] [blame] | 1559 | For further documentation see \fBHOWTO\fR and \fBREADME\fR. |
| 1560 | .br |
| 1561 | Sample jobfiles are available in the \fBexamples\fR directory. |
Aaron Carroll | d60e92d | 2007-09-17 10:32:59 +0200 | [diff] [blame] | 1562 | |