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 --help Print this page
138 --cmdhelp=cmd Print command help, "all" for all of them
139 --showcmd Turn a job file into command line options
140 --readonly Turn on safety read-only checks, preventing
142 --eta=when When ETA estimate should be printed
143 May be "always", "never" or "auto"
144 --section=name Only run specified section in job file. Multiple
145 sections can be specified.
146 --alloc-size=kb Set smalloc pool to this size in kb (def 1024)
147 --warnings-fatal Fio parser warnings are fatal
150 Any parameters following the options will be assumed to be job files,
151 unless they match a job file parameter. You can add as many as you want,
152 each job file will be regarded as a separate group and fio will stonewall
155 The --readonly switch is an extra safety guard to prevent accidentally
156 turning on a write setting when that is not desired. Fio will only write
157 if rw=write/randwrite/rw/randrw is given, but this extra safety net can
158 be used as an extra precaution. It will also enable a write check in the
159 io engine core to prevent an accidental write due to a fio bug.
161 The debug switch allows adding options that trigger certain logging
162 options in fio. Currently the options are:
164 process Dump info related to processes
165 file Dump info related to file actions
166 io Dump info related to IO queuing
167 mem Dump info related to memory allocations
168 blktrace Dump info related to blktrace setup
169 verify Dump info related to IO verification
170 all Enable all debug options
171 random Dump info related to random offset generation
172 parse Dump info related to option matching and parsing
173 diskutil Dump info related to disk utilization updates
174 job:x Dump info only related to job number x
175 mutex Dump info only related to mutex up/down ops
176 profile Dump info related to profile extensions
177 time Dump info related to internal time keeping
178 ? or help Show available debug options.
180 You can specify as many as you want, eg --debug=file,mem will enable
181 file and memory debugging.
183 The section switch is meant to make it easier to ship a bigger job file
184 instead of several smaller ones. Say you define a job file with light,
185 moderate, and heavy parts. Then you can ask fio to run the given part
186 only by giving it a --section=heavy command line option. The section
187 option only applies to job sections, the reserved 'global' section is
188 always parsed and taken into account.
190 Fio has an internal allocator for shared memory called smalloc. It
191 allocates shared structures from this pool. The pool defaults to 1024k
192 in size, and can grow to 128 pools. If running large jobs with randommap
193 enabled it can run out of memory, in which case the --alloc-size switch
194 is handy for starting with a larger pool size. The backing store is
195 files in /tmp. Fio cleans up after itself, while it is running you
196 may see .fio_smalloc.* files in /tmp.
202 See the HOWTO file for a more detailed description of parameters and what
203 they mean. This file contains the terse version. You can describe big and
204 complex setups with the command line, but generally it's a lot easier to
205 just write a simple job file to describe the workload. The job file format
206 is in the ini style format, as that is easy to read and write for the user.
208 The job file parameters are:
210 name=x Use 'x' as the identifier for this job.
211 description=x 'x' is a text description of the job.
212 directory=x Use 'x' as the top level directory for storing files
213 filename=x Force the use of 'x' as the filename for all files
214 in this thread. If not given, fio will make up
215 a suitable filename based on the thread and file
217 rw=x 'x' may be: read, randread, write, randwrite,
218 rw (read-write mix), randrw (read-write random mix)
219 rwmixcycle=x Base cycle for switching between read and write
221 rwmixread=x 'x' percentage of rw mix ios will be reads. If
222 rwmixwrite is also given, the last of the two will
223 be used if they don't add up to 100%.
224 rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
226 rand_repeatable=x The sequence of random io blocks can be repeatable
227 across runs, if 'x' is 1.
228 size=x Set file size to x bytes (x string can include k/m/g)
229 ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
230 posixaio for POSIX aio, solarisaio for Solaris
231 native async IO, windowsaio for Windows native async IO,
232 sync for regular read/write io,
233 psync for regular pread/pwrite io, vsync for regular
234 readv/writev (with queuing emulation) mmap for mmap'ed
235 io, syslet-rw for syslet driven read/write, splice for
236 using splice/vmsplice, sg for direct SG_IO io, net
237 for network io, or cpuio for a cycler burner load. sg
238 only works on Linux on SCSI (or SCSI-like devices, such
239 as usb-storage or sata/libata driven) devices. Fio also
240 has a null io engine, which is mainly used for testing
243 iodepth=x For async io, allow 'x' ios in flight
244 overwrite=x If 'x', layout a write file first.
245 nrfiles=x Spread io load over 'x' number of files per job,
247 prio=x Run io at prio X, 0-7 is the kernel allowed range
248 prioclass=x Run io at prio class X
249 bs=x Use 'x' for thread blocksize. May include k/m postfix.
250 bsrange=x-y Mix thread block sizes randomly between x and y. May
251 also include k/m postfix.
252 direct=x 1 for direct IO, 0 for buffered IO
253 thinktime=x "Think" x usec after each io
254 rate=x Throttle rate to x KB/sec
255 ratemin=x Quit if rate of x KB/sec can't be met
256 ratecycle=x ratemin averaged over x msecs
257 cpumask=x Only allow job to run on CPUs defined by mask.
258 cpus_allowed=x Like 'cpumask', but allow text setting of CPU affinity.
259 fsync=x If writing with buffered IO, fsync after every
260 'x' blocks have been written.
261 end_fsync=x If 'x', run fsync() after end-of-job.
262 startdelay=x Start this thread x seconds after startup
263 runtime=x Terminate x seconds after startup. Can include a
264 normal time suffix if not given in seconds, such as
265 'm' for minutes, 'h' for hours, and 'd' for days.
266 offset=x Start io at offset x (x string can include k/m/g)
267 invalidate=x Invalidate page cache for file prior to doing io
268 sync=x Use sync writes if x and writing buffered IO.
269 mem=x If x == malloc, use malloc for buffers. If x == shm,
270 use shared memory for buffers. If x == mmap, use
272 exitall When one thread quits, terminate the others
273 bwavgtime=x Average bandwidth stats over an x msec window.
274 create_serialize=x If 'x', serialize file creation.
275 create_fsync=x If 'x', run fsync() after file creation.
276 unlink If set, unlink files when done.
277 loops=x Run the job 'x' number of times.
278 verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
279 use crc32 for verifies. md5 is 'safer', but crc32 is
280 a lot faster. Only makes sense for writing to a file.
281 For other types of checksumming, see HOWTO.
282 stonewall Wait for preceeding jobs to end before running.
283 numjobs=x Create 'x' similar entries for this job
284 thread Use pthreads instead of forked jobs
286 zoneskip=y Zone options must be paired. If given, the job
287 will skip y bytes for every x read/written. This
288 can be used to gauge hard drive speed over the entire
289 platter, without reading everything. Both x/y can
290 include k/m/g suffix.
291 iolog=x Open and read io pattern from file 'x'. The file must
292 contain one io action per line in the following format:
294 where with rw=0/1 for read/write, and the offset
295 and length entries being in bytes.
296 write_iolog=x Write an iolog to file 'x' in the same format as iolog.
297 The iolog options are exclusive, if both given the
298 read iolog will be performed. Specify a separate file
299 for each job, otherwise the iologs will be interspersed
300 and the file may be corrupt.
301 write_bw_log Write a bandwidth log.
302 write_lat_log Write a latency log.
303 lockmem=x Lock down x amount of memory on the machine, to
304 simulate a machine with less memory available. x can
305 include k/m/g suffix.
306 nice=x Run job at given nice value.
307 exec_prerun=x Run 'x' before job io is begun.
308 exec_postrun=x Run 'x' after job io has finished.
309 ioscheduler=x Use ioscheduler 'x' for this job.
310 cpuload=x For a CPU io thread, percentage of CPU time to attempt
312 cpuchunks=x Split burn cycles into pieces of x usecs.
319 Fio works on (at least) Linux, Solaris, AIX, OSX, NetBSD, Windows and FreeBSD.
320 Some features and/or options may only be available on some of the platforms,
321 typically because those features only apply to that platform (like the
322 solarisaio engine, or the splice engine on Linux).
324 Some features are not available on FreeBSD/Solaris even if they could be
325 implemented, I'd be happy to take patches for that. An example of that is
326 disk utility statistics and (I think) huge page support, support for that
327 does exist in FreeBSD/Solaris.
329 Fio uses pthread mutexes for signalling and locking and FreeBSD does not
330 support process shared pthread mutexes. As a result, only threads are
331 supported on FreeBSD. This could be fixed with sysv ipc locking or
332 other locking alternatives.
334 Other *BSD platforms are untested, but fio should work there almost out
335 of the box. Since I don't do test runs or even compiles on those platforms,
336 your mileage may vary. Sending me patches for other platforms is greatly
337 appreciated. There's a lot of value in having the same test/benchmark tool
338 available on all platforms.
340 Note that POSIX aio is not enabled by default on AIX. If you get messages like:
342 Symbol resolution failed for /usr/lib/libc.a(posix_aio.o) because:
343 Symbol _posix_kaio_rdwr (number 2) is not exported from dependent module /unix.
345 you need to enable POSIX aio. Run the following commands as root:
347 # lsdev -C -l posix_aio0
348 posix_aio0 Defined Posix Asynchronous I/O
349 # cfgmgr -l posix_aio0
350 # lsdev -C -l posix_aio0
351 posix_aio0 Available Posix Asynchronous I/O
353 POSIX aio should work now. To make the change permanent:
355 # chdev -l posix_aio0 -P -a autoconfig='available'
362 Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
363 of the Linux IO subsystem and schedulers. He got tired of writing
364 specific test applications to simulate a given workload, and found that
365 the existing io benchmark/test tools out there weren't flexible enough
366 to do what he wanted.
368 Jens Axboe <axboe@kernel.dk> 20060905