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
15 fio resides in a git repo, the canonical place is:
17 git://git.kernel.dk/fio.git
19 The http protocol also works, path is the same.
21 Snapshots are frequently generated and they include the git meta data as
22 well. You can download them here:
24 http://brick.kernel.dk/snaps/
31 Starting with Debian "Squeeze", fio packages are part of the official
32 Debian repository. http://packages.debian.org/search?keywords=fio
35 Starting with Ubuntu 10.04 LTS (aka "Lucid Lynx"), fio packages are part
36 of the Ubuntu "universe" repository.
37 http://packages.ubuntu.com/search?keywords=fio
40 Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository for SUSE
41 variants, you can find them here:
42 http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio
45 Dag Wieƫrs has RPMs for Red Hat related distros, find them here:
46 http://dag.wieers.com/rpm/packages/fio/
49 Mandriva has integrated fio into their package repository, so installing
50 on that distro should be as easy as typing 'urpmi fio'.
53 Packages for Solaris are available from OpenCSW. Install their pkgutil
54 tool (http://www.opencsw.org/get-it/pkgutil/) and then install fio via
58 Bruce Cran <bruce@cran.org.uk> has fio packages for Windows at
59 http://www.bluestop.org/fio .
65 There's a mailing list associated with fio. It's meant for general
66 discussion, bug reporting, questions, and development - basically anything
67 that has to do with fio. An automated mail detailing recent commits is
68 automatically sent to the list at most daily. The list address is
69 fio@vger.kernel.org, subscribe by sending an email to
70 majordomo@vger.kernel.org with
74 in the body of the email. Archives can be found here:
76 http://www.spinics.net/lists/fio/
78 and archives for the old list can be found here:
80 http://maillist.kernel.dk/fio-devel/
86 Just type 'make' and 'make install'.
88 Note that GNU make is required. On BSD it's available from devel/gmake;
89 on Solaris it's in the SUNWgmake package. On platforms where GNU make
90 isn't the default, type 'gmake' instead of 'make'.
92 If your compile fails with an error like this:
95 In file included from fio.h:23,
97 os/os.h:15:20: error: libaio.h: No such file or directory
98 In file included from gettime.c:8:
99 fio.h:119: error: field 'iocb' has incomplete type
100 make: *** [gettime.o] Error 1
102 Check that you have the libaio development package installed. On RPM
103 based distros, it's typically called libaio-devel.
109 On Windows Cygwin (http://www.cygwin.com) is required with at least
110 devel/gcc4 and devel/make installed in order to build fio, and
111 admin/cygrunsrv to run it. You can also install devel/git to fetch/update
112 the source files. To create an MSI installer package put a copy of Cygwin
113 in os\windows\fio, install WiX 3.6 from http://wix.sourceforge.net/releases/
114 and run dobuild.cmd from the os/windows directory.
116 Before running fio you'll need to have a copy of cygserver running. Run
117 "/usr/bin/cygserver-config" from an elevated Cygwin shell (i.e. launch the
118 Cygwin shell under the Administrator account) to configure it. Once
119 configured, run "net start cygserver" to start it, or type
120 "/usr/sbin/cygserver &" in the Cygwin shell to start a local copy.
122 If fio exits with the message "Bad system call" it normally means that
123 Cygserver isn't running.
130 --debug Enable some debugging options (see below)
131 --output Write output to file
132 --timeout Runtime in seconds
133 --latency-log Generate per-job latency logs
134 --bandwidth-log Generate per-job bandwidth logs
135 --minimal Minimal (terse) output
136 --version Print version info and exit
137 --terse-version=type Terse version output format
138 --help Print this page
139 --cmdhelp=cmd Print command help, "all" for all of them
140 --showcmd Turn a job file into command line options
141 --readonly Turn on safety read-only checks, preventing
143 --eta=when When ETA estimate should be printed
144 May be "always", "never" or "auto"
145 --section=name Only run specified section in job file. Multiple
146 sections can be specified.
147 --alloc-size=kb Set smalloc pool to this size in kb (def 1024)
148 --warnings-fatal Fio parser warnings are fatal
149 --max-jobs Maximum number of threads/processes to support
152 Any parameters following the options will be assumed to be job files,
153 unless they match a job file parameter. You can add as many as you want,
154 each job file will be regarded as a separate group and fio will stonewall
157 The --readonly switch is an extra safety guard to prevent accidentally
158 turning on a write setting when that is not desired. Fio will only write
159 if rw=write/randwrite/rw/randrw is given, but this extra safety net can
160 be used as an extra precaution. It will also enable a write check in the
161 io engine core to prevent an accidental write due to a fio bug.
163 The debug switch allows adding options that trigger certain logging
164 options in fio. Currently the options are:
166 process Dump info related to processes
167 file Dump info related to file actions
168 io Dump info related to IO queuing
169 mem Dump info related to memory allocations
170 blktrace Dump info related to blktrace setup
171 verify Dump info related to IO verification
172 all Enable all debug options
173 random Dump info related to random offset generation
174 parse Dump info related to option matching and parsing
175 diskutil Dump info related to disk utilization updates
176 job:x Dump info only related to job number x
177 mutex Dump info only related to mutex up/down ops
178 profile Dump info related to profile extensions
179 time Dump info related to internal time keeping
180 ? or help Show available debug options.
182 You can specify as many as you want, eg --debug=file,mem will enable
183 file and memory debugging.
185 The section switch is meant to make it easier to ship a bigger job file
186 instead of several smaller ones. Say you define a job file with light,
187 moderate, and heavy parts. Then you can ask fio to run the given part
188 only by giving it a --section=heavy command line option. The section
189 option only applies to job sections, the reserved 'global' section is
190 always parsed and taken into account.
192 Fio has an internal allocator for shared memory called smalloc. It
193 allocates shared structures from this pool. The pool defaults to 1024k
194 in size, and can grow to 128 pools. If running large jobs with randommap
195 enabled it can run out of memory, in which case the --alloc-size switch
196 is handy for starting with a larger pool size. The backing store is
197 files in /tmp. Fio cleans up after itself, while it is running you
198 may see .fio_smalloc.* files in /tmp.
204 See the HOWTO file for a more detailed description of parameters and what
205 they mean. This file contains the terse version. You can describe big and
206 complex setups with the command line, but generally it's a lot easier to
207 just write a simple job file to describe the workload. The job file format
208 is in the ini style format, as that is easy to read and write for the user.
210 The job file parameters are:
212 name=x Use 'x' as the identifier for this job.
213 description=x 'x' is a text description of the job.
214 directory=x Use 'x' as the top level directory for storing files
215 filename=x Force the use of 'x' as the filename for all files
216 in this thread. If not given, fio will make up
217 a suitable filename based on the thread and file
219 rw=x 'x' may be: read, randread, write, randwrite,
220 rw (read-write mix), randrw (read-write random mix)
221 rwmixcycle=x Base cycle for switching between read and write
223 rwmixread=x 'x' percentage of rw mix ios will be reads. If
224 rwmixwrite is also given, the last of the two will
225 be used if they don't add up to 100%.
226 rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
228 rand_repeatable=x The sequence of random io blocks can be repeatable
229 across runs, if 'x' is 1.
230 size=x Set file size to x bytes (x string can include k/m/g)
231 ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
232 posixaio for POSIX aio, solarisaio for Solaris
233 native async IO, windowsaio for Windows native async IO,
234 sync for regular read/write io,
235 psync for regular pread/pwrite io, vsync for regular
236 readv/writev (with queuing emulation) mmap for mmap'ed
237 io, syslet-rw for syslet driven read/write, splice for
238 using splice/vmsplice, sg for direct SG_IO io, net
239 for network io, or cpuio for a cycler burner load. sg
240 only works on Linux on SCSI (or SCSI-like devices, such
241 as usb-storage or sata/libata driven) devices. Fio also
242 has a null io engine, which is mainly used for testing
245 iodepth=x For async io, allow 'x' ios in flight
246 overwrite=x If 'x', layout a write file first.
247 nrfiles=x Spread io load over 'x' number of files per job,
249 prio=x Run io at prio X, 0-7 is the kernel allowed range
250 prioclass=x Run io at prio class X
251 bs=x Use 'x' for thread blocksize. May include k/m postfix.
252 bsrange=x-y Mix thread block sizes randomly between x and y. May
253 also include k/m postfix.
254 direct=x 1 for direct IO, 0 for buffered IO
255 thinktime=x "Think" x usec after each io
256 rate=x Throttle rate to x KB/sec
257 ratemin=x Quit if rate of x KB/sec can't be met
258 ratecycle=x ratemin averaged over x msecs
259 cpumask=x Only allow job to run on CPUs defined by mask.
260 cpus_allowed=x Like 'cpumask', but allow text setting of CPU affinity.
261 fsync=x If writing with buffered IO, fsync after every
262 'x' blocks have been written.
263 end_fsync=x If 'x', run fsync() after end-of-job.
264 startdelay=x Start this thread x seconds after startup
265 runtime=x Terminate x seconds after startup. Can include a
266 normal time suffix if not given in seconds, such as
267 'm' for minutes, 'h' for hours, and 'd' for days.
268 offset=x Start io at offset x (x string can include k/m/g)
269 invalidate=x Invalidate page cache for file prior to doing io
270 sync=x Use sync writes if x and writing buffered IO.
271 mem=x If x == malloc, use malloc for buffers. If x == shm,
272 use shared memory for buffers. If x == mmap, use
274 exitall When one thread quits, terminate the others
275 bwavgtime=x Average bandwidth stats over an x msec window.
276 create_serialize=x If 'x', serialize file creation.
277 create_fsync=x If 'x', run fsync() after file creation.
278 unlink If set, unlink files when done.
279 loops=x Run the job 'x' number of times.
280 verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
281 use crc32 for verifies. md5 is 'safer', but crc32 is
282 a lot faster. Only makes sense for writing to a file.
283 For other types of checksumming, see HOWTO.
284 stonewall Wait for preceeding jobs to end before running.
285 numjobs=x Create 'x' similar entries for this job
286 thread Use pthreads instead of forked jobs
288 zoneskip=y Zone options must be paired. If given, the job
289 will skip y bytes for every x read/written. This
290 can be used to gauge hard drive speed over the entire
291 platter, without reading everything. Both x/y can
292 include k/m/g suffix.
293 read_iolog=x Open and read io pattern from file 'x'. The file format
294 is described in the HOWTO.
295 write_iolog=x Write an iolog to file 'x' in the same format as iolog.
296 The iolog options are exclusive, if both given the
297 read iolog will be performed. Specify a separate file
298 for each job, otherwise the iologs will be interspersed
299 and the file may be corrupt.
300 write_bw_log Write a bandwidth log.
301 write_lat_log Write a latency log.
302 lockmem=x Lock down x amount of memory on the machine, to
303 simulate a machine with less memory available. x can
304 include k/m/g suffix.
305 nice=x Run job at given nice value.
306 exec_prerun=x Run 'x' before job io is begun.
307 exec_postrun=x Run 'x' after job io has finished.
308 ioscheduler=x Use ioscheduler 'x' for this job.
309 cpuload=x For a CPU io thread, percentage of CPU time to attempt
311 cpuchunks=x Split burn cycles into pieces of x usecs.
318 Fio works on (at least) Linux, Solaris, AIX, HP-UX, OSX, NetBSD, Windows
319 and FreeBSD. Some features and/or options may only be available on some of
320 the platforms, typically because those features only apply to that platform
321 (like the solarisaio engine, or the splice engine on Linux).
323 Some features are not available on FreeBSD/Solaris even if they could be
324 implemented, I'd be happy to take patches for that. An example of that is
325 disk utility statistics and (I think) huge page support, support for that
326 does exist in FreeBSD/Solaris.
328 Fio uses pthread mutexes for signalling and locking and FreeBSD does not
329 support process shared pthread mutexes. As a result, only threads are
330 supported on FreeBSD. This could be fixed with sysv ipc locking or
331 other locking alternatives.
333 Other *BSD platforms are untested, but fio should work there almost out
334 of the box. Since I don't do test runs or even compiles on those platforms,
335 your mileage may vary. Sending me patches for other platforms is greatly
336 appreciated. There's a lot of value in having the same test/benchmark tool
337 available on all platforms.
339 Note that POSIX aio is not enabled by default on AIX. If you get messages like:
341 Symbol resolution failed for /usr/lib/libc.a(posix_aio.o) because:
342 Symbol _posix_kaio_rdwr (number 2) is not exported from dependent module /unix.
344 you need to enable POSIX aio. Run the following commands as root:
346 # lsdev -C -l posix_aio0
347 posix_aio0 Defined Posix Asynchronous I/O
348 # cfgmgr -l posix_aio0
349 # lsdev -C -l posix_aio0
350 posix_aio0 Available Posix Asynchronous I/O
352 POSIX aio should work now. To make the change permanent:
354 # chdev -l posix_aio0 -P -a autoconfig='available'
361 Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
362 of the Linux IO subsystem and schedulers. He got tired of writing
363 specific test applications to simulate a given workload, and found that
364 the existing io benchmark/test tools out there weren't flexible enough
365 to do what he wanted.
367 Jens Axboe <axboe@kernel.dk> 20060905