Commit | Line | Data |
---|---|---|
ebac4655 JA |
1 | fio |
2 | --- | |
3 | ||
79809113 JA |
4 | fio is a tool that will spawn a number of threads or processes doing a |
5 | particular type of io action as specified by the user. fio takes a | |
6 | number of global parameters, each inherited by the thread unless | |
7 | otherwise parameters given to them overriding that setting is given. | |
8 | The typical use of fio is to write a job file matching the io load | |
9 | one wants to simulate. | |
ebac4655 | 10 | |
2b02b546 JA |
11 | |
12 | Source | |
13 | ------ | |
14 | ||
15 | fio resides in a git repo, the canonical place is: | |
16 | ||
6b3eccb1 | 17 | git://git.kernel.dk/fio.git |
97f049c9 JA |
18 | |
19 | The http protocol also works, path is the same. | |
2b02b546 | 20 | |
79809113 JA |
21 | Snapshots are frequently generated and they include the git meta data as |
22 | well. You can download them here: | |
2b02b546 JA |
23 | |
24 | http://brick.kernel.dk/snaps/ | |
25 | ||
a68594cb JA |
26 | Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository for |
27 | SUSE variants, you can find them here: | |
1053a106 JA |
28 | |
29 | http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio | |
30 | ||
a68594cb JA |
31 | Dag Wieƫrs has RPMs for Red Hat related distros, find them here: |
32 | ||
33 | http://dag.wieers.com/rpm/packages/fio/ | |
34 | ||
244e170e JA |
35 | Mandriva has integrated fio into their package repository, so installing |
36 | on that distro should be as easy as typing 'urpmi fio'. | |
37 | ||
2b02b546 | 38 | |
726f6ff0 JA |
39 | Mailing list |
40 | ------------ | |
41 | ||
42 | There's a mailing list associated with fio. It's meant for general | |
2e8552b0 JA |
43 | discussion, bug reporting, questions, and development - basically anything |
44 | that has to do with fio. An automated mail detailing recent commits is | |
45 | automatically sent to the list at most daily. The list address is | |
46 | fio@vger.kernel.org, subscribe by sending an email to | |
47 | majordomo@vger.kernel.org with | |
48 | ||
49 | subscribe fio | |
50 | ||
51 | in the body of the email. There is no archive for the new list yet, | |
52 | archives for the old list can be found here: | |
53 | ||
54 | http://maillist.kernel.dk/fio-devel/ | |
726f6ff0 JA |
55 | |
56 | ||
bbfd6b00 JA |
57 | Building |
58 | -------- | |
59 | ||
60 | Just type 'make' and 'make install'. If on FreeBSD, for now you have to | |
5f421004 | 61 | specify the FreeBSD Makefile with -f and use gmake (not make), eg: |
bbfd6b00 | 62 | |
5f421004 | 63 | $ gmake -f Makefile.Freebsd && gmake -f Makefile.FreeBSD install |
bbfd6b00 | 64 | |
edffcb96 | 65 | Likewise with OpenSolaris, use the Makefile.solaris to compile there. |
5f421004 JA |
66 | The OpenSolaris make should work fine. This might change in the |
67 | future if I opt for an autoconf type setup. | |
bbfd6b00 | 68 | |
6de43c1b JA |
69 | If your compile fails with an error like this: |
70 | ||
71 | CC gettime.o | |
72 | In file included from fio.h:23, | |
73 | from gettime.c:8: | |
74 | os/os.h:15:20: error: libaio.h: No such file or directory | |
75 | In file included from gettime.c:8: | |
76 | fio.h:119: error: field 'iocb' has incomplete type | |
77 | make: *** [gettime.o] Error 1 | |
78 | ||
79 | Check that you have the libaio development package installed. On RPM | |
80 | based distros, it's typically called libaio-devel. | |
81 | ||
bbfd6b00 | 82 | |
972cfd25 JA |
83 | Command line |
84 | ------------ | |
ebac4655 JA |
85 | |
86 | $ fio | |
ee56ad50 | 87 | --debug Enable some debugging options (see below) |
b4692828 | 88 | --output Write output to file |
062c6022 | 89 | --timeout Runtime in seconds |
b4692828 JA |
90 | --latency-log Generate per-job latency logs |
91 | --bandwidth-log Generate per-job bandwidth logs | |
92 | --minimal Minimal (terse) output | |
93 | --version Print version info and exit | |
fd28ca49 JA |
94 | --help Print this page |
95 | --cmdhelp=cmd Print command help, "all" for all of them | |
cca73aa7 | 96 | --showcmd Turn a job file into command line options |
062c6022 | 97 | --readonly Turn on safety read-only checks, preventing writes |
e592a06b AC |
98 | --eta=when When ETA estimate should be printed |
99 | May be "always", "never" or "auto" | |
01f06b63 | 100 | --section=name Only run specified section in job file |
2b386d25 | 101 | --alloc-size=kb Set smalloc pool to this size in kb (def 1024) |
e592a06b | 102 | |
b4692828 JA |
103 | |
104 | Any parameters following the options will be assumed to be job files, | |
105 | unless they match a job file parameter. You can add as many as you want, | |
106 | each job file will be regarded as a separate group and fio will stonewall | |
107 | its execution. | |
972cfd25 | 108 | |
724e4435 JA |
109 | The --readonly switch is an extra safety guard to prevent accidentically |
110 | turning on a write setting when that is not desired. Fio will only write | |
111 | if rw=write/randwrite/rw/randrw is given, but this extra safety net can | |
112 | be used as an extra precaution. It will also enable a write check in the | |
113 | io engine core to prevent an accidental write due to a fio bug. | |
114 | ||
ee56ad50 JA |
115 | The debug switch allows adding options that trigger certain logging |
116 | options in fio. Currently the options are: | |
117 | ||
118 | process Dump info related to processes | |
119 | file Dump info related to file actions | |
120 | io Dump info related to IO queuing | |
121 | mem Dump info related to memory allocations | |
bd6f78b2 JA |
122 | blktrace Dump info related to blktrace setup |
123 | verify Dump info related to IO verification | |
124 | all Enable all debug options | |
811a0d06 | 125 | random Dump info related to random offset generation |
a3d741fa | 126 | parse Dump info related to option matching and parsing |
cd991b9e | 127 | diskutil Dump info related to disk utilization updates |
5e1d306e | 128 | job:x Dump info only related to job number x |
29adda3c | 129 | mutex Dump info only related to mutex up/down ops |
bd6f78b2 | 130 | ? or help Show available debug options. |
ee56ad50 JA |
131 | |
132 | You can specify as many as you want, eg --debug=file,mem will enable | |
bd6f78b2 | 133 | file and memory debugging. |
ee56ad50 | 134 | |
01f06b63 JA |
135 | The section switch is meant to make it easier to ship a bigger job file |
136 | instead of several smaller ones. Say you define a job file with light, | |
137 | moderate, and heavy parts. Then you can ask fio to run the given part | |
138 | only by giving it a --section=heavy command line option. The section | |
139 | option only applies to job sections, the reserved 'global' section is | |
140 | always parsed and taken into account. | |
141 | ||
2b386d25 JA |
142 | Fio has an internal allocator for shared memory called smalloc. It |
143 | allocates shared structures from this pool. The pool defaults to 1024k | |
931823ca | 144 | in size, and can grow to 128 pools. If running large jobs with randommap |
2b386d25 | 145 | enabled it can run out of memory, in which case the --alloc-size switch |
931823ca JA |
146 | is handy for starting with a larger pool size. The backing store is |
147 | files in /tmp. Fio cleans up after itself, while it is running you | |
148 | may see .fio_smalloc.* files in /tmp. | |
2b386d25 | 149 | |
79809113 JA |
150 | |
151 | Job file | |
152 | -------- | |
153 | ||
71bfa161 | 154 | See the HOWTO file for a more detailed description of parameters and what |
4661f3d0 JA |
155 | they mean. This file contains the terse version. You can describe big and |
156 | complex setups with the command line, but generally it's a lot easier to | |
71bfa161 | 157 | just write a simple job file to describe the workload. The job file format |
4661f3d0 | 158 | is in the ini style format, as that is easy to read and write for the user. |
79809113 JA |
159 | |
160 | The job file parameters are: | |
ebac4655 | 161 | |
01452055 | 162 | name=x Use 'x' as the identifier for this job. |
61697c37 | 163 | description=x 'x' is a text description of the job. |
ebac4655 | 164 | directory=x Use 'x' as the top level directory for storing files |
b50b8755 JA |
165 | filename=x Force the use of 'x' as the filename for all files |
166 | in this thread. If not given, fio will make up | |
167 | a suitable filename based on the thread and file | |
168 | number. | |
3d60d1ed JA |
169 | rw=x 'x' may be: read, randread, write, randwrite, |
170 | rw (read-write mix), randrw (read-write random mix) | |
a6ccc7be JA |
171 | rwmixcycle=x Base cycle for switching between read and write |
172 | in msecs. | |
173 | rwmixread=x 'x' percentage of rw mix ios will be reads. If | |
174 | rwmixwrite is also given, the last of the two will | |
175 | be used if they don't add up to 100%. | |
176 | rwmixwrite=x 'x' percentage of rw mix ios will be writes. See | |
177 | rwmixread. | |
9ebc27e1 JA |
178 | rand_repeatable=x The sequence of random io blocks can be repeatable |
179 | across runs, if 'x' is 1. | |
ebac4655 JA |
180 | size=x Set file size to x bytes (x string can include k/m/g) |
181 | ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio, | |
78e7b3e7 JA |
182 | posixaio for POSIX aio, solarisaio for Solaris |
183 | native async IO, sync for regular read/write io, | |
1d2af02a JA |
184 | psync for regular pread/pwrite io, vsync for regular |
185 | readv/writev (with queuing emulation) mmap for mmap'ed | |
186 | io, syslet-rw for syslet driven read/write, splice for | |
d0c70934 GP |
187 | using splice/vmsplice, sg for direct SG_IO io, net |
188 | for network io, or cpuio for a cycler burner load. sg | |
1d2af02a JA |
189 | only works on Linux on SCSI (or SCSI-like devices, such |
190 | as usb-storage or sata/libata driven) devices. Fio also | |
191 | has a null io engine, which is mainly used for testing | |
192 | fio itself. | |
193 | ||
ebac4655 JA |
194 | iodepth=x For async io, allow 'x' ios in flight |
195 | overwrite=x If 'x', layout a write file first. | |
53cdc686 JA |
196 | nrfiles=x Spread io load over 'x' number of files per job, |
197 | if possible. | |
ebac4655 JA |
198 | prio=x Run io at prio X, 0-7 is the kernel allowed range |
199 | prioclass=x Run io at prio class X | |
200 | bs=x Use 'x' for thread blocksize. May include k/m postfix. | |
201 | bsrange=x-y Mix thread block sizes randomly between x and y. May | |
202 | also include k/m postfix. | |
203 | direct=x 1 for direct IO, 0 for buffered IO | |
204 | thinktime=x "Think" x usec after each io | |
205 | rate=x Throttle rate to x KiB/sec | |
206 | ratemin=x Quit if rate of x KiB/sec can't be met | |
207 | ratecycle=x ratemin averaged over x msecs | |
208 | cpumask=x Only allow job to run on CPUs defined by mask. | |
d2e268b0 | 209 | cpus_allowed=x Like 'cpumask', but allow text setting of CPU affinity. |
795407ca JA |
210 | fsync=x If writing with buffered IO, fsync after every |
211 | 'x' blocks have been written. | |
212 | end_fsync=x If 'x', run fsync() after end-of-job. | |
ebac4655 | 213 | startdelay=x Start this thread x seconds after startup |
03b74b3e | 214 | runtime=x Terminate x seconds after startup. Can include a |
906c8d75 JA |
215 | normal time suffix if not given in seconds, such as |
216 | 'm' for minutes, 'h' for hours, and 'd' for days. | |
ebac4655 JA |
217 | offset=x Start io at offset x (x string can include k/m/g) |
218 | invalidate=x Invalidate page cache for file prior to doing io | |
795407ca | 219 | sync=x Use sync writes if x and writing buffered IO. |
ebac4655 | 220 | mem=x If x == malloc, use malloc for buffers. If x == shm, |
795407ca JA |
221 | use shared memory for buffers. If x == mmap, use |
222 | anonymous mmap. | |
ebac4655 JA |
223 | exitall When one thread quits, terminate the others |
224 | bwavgtime=x Average bandwidth stats over an x msec window. | |
225 | create_serialize=x If 'x', serialize file creation. | |
226 | create_fsync=x If 'x', run fsync() after file creation. | |
f6cbb269 | 227 | unlink If set, unlink files when done. |
ebac4655 JA |
228 | loops=x Run the job 'x' number of times. |
229 | verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32, | |
230 | use crc32 for verifies. md5 is 'safer', but crc32 is | |
231 | a lot faster. Only makes sense for writing to a file. | |
bac39e0e | 232 | For other types of checksumming, see HOWTO. |
ebac4655 JA |
233 | stonewall Wait for preceeding jobs to end before running. |
234 | numjobs=x Create 'x' similar entries for this job | |
235 | thread Use pthreads instead of forked jobs | |
20dc95c4 JA |
236 | zonesize=x |
237 | zoneskip=y Zone options must be paired. If given, the job | |
238 | will skip y bytes for every x read/written. This | |
239 | can be used to gauge hard drive speed over the entire | |
240 | platter, without reading everything. Both x/y can | |
241 | include k/m/g suffix. | |
aea47d44 JA |
242 | iolog=x Open and read io pattern from file 'x'. The file must |
243 | contain one io action per line in the following format: | |
244 | rw, offset, length | |
245 | where with rw=0/1 for read/write, and the offset | |
246 | and length entries being in bytes. | |
843a7413 JA |
247 | write_iolog=x Write an iolog to file 'x' in the same format as iolog. |
248 | The iolog options are exclusive, if both given the | |
249 | read iolog will be performed. | |
ec94ec56 JA |
250 | write_bw_log Write a bandwidth log. |
251 | write_lat_log Write a latency log. | |
c04f7ec3 JA |
252 | lockmem=x Lock down x amount of memory on the machine, to |
253 | simulate a machine with less memory available. x can | |
254 | include k/m/g suffix. | |
b6f4d880 | 255 | nice=x Run job at given nice value. |
4e0ba8af JA |
256 | exec_prerun=x Run 'x' before job io is begun. |
257 | exec_postrun=x Run 'x' after job io has finished. | |
da86774e | 258 | ioscheduler=x Use ioscheduler 'x' for this job. |
b990b5c0 JA |
259 | cpuload=x For a CPU io thread, percentage of CPU time to attempt |
260 | to burn. | |
ba0fbe10 | 261 | cpuchunks=x Split burn cycles into pieces of x usecs. |
ebac4655 | 262 | |
79809113 | 263 | |
217bc04b JA |
264 | |
265 | Platforms | |
266 | --------- | |
267 | ||
268 | Fio works on (at least) Linux, Solaris, and FreeBSD. Some features and/or | |
269 | options may only be available on some of the platforms, typically because | |
270 | those features only apply to that platform (like the solarisaio engine, or | |
271 | the splice engine on Linux). | |
272 | ||
273 | Some features are not available on FreeBSD/Solaris even if they could be | |
274 | implemented, I'd be happy to take patches for that. An example of that is | |
275 | disk utility statistics and (I think) huge page support, support for that | |
276 | does exist in FreeBSD/Solaris. | |
277 | ||
278 | Fio uses pthread mutexes for signalling and locking and FreeBSD does not | |
279 | support process shared pthread mutexes. As a result, only threads are | |
280 | supported on FreeBSD. This could be fixed with sysv ipc locking or | |
281 | other locking alternatives. | |
282 | ||
283 | Other *BSD platforms are untested, but fio should work there almost out | |
284 | of the box. Since I don't do test runs or even compiles on those platforms, | |
285 | your mileage may vary. Sending me patches for other platforms is greatly | |
286 | appreciated. There's a lot of value in having the same test/benchmark tool | |
287 | available on all platforms. | |
288 | ||
289 | ||
290 | ||
79809113 JA |
291 | Author |
292 | ------ | |
293 | ||
aae22ca7 | 294 | Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing |
79809113 JA |
295 | of the Linux IO subsystem and schedulers. He got tired of writing |
296 | specific test applications to simulate a given workload, and found that | |
297 | the existing io benchmark/test tools out there weren't flexible enough | |
298 | to do what he wanted. | |
299 | ||
aae22ca7 | 300 | Jens Axboe <axboe@kernel.dk> 20060905 |
79809113 | 301 |