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/
26 Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository for
27 SUSE variants, you can find them here:
29 http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio
31 Dag Wieƫrs has RPMs for Red Hat related distros, find them here:
33 http://dag.wieers.com/rpm/packages/fio/
35 Mandriva has integrated fio into their package repository, so installing
36 on that distro should be as easy as typing 'urpmi fio'.
42 There's a mailing list associated with fio. It's meant for general
43 discussion, bug reporting, questions - basically anything that has to
44 do with fio. An automated mail detailing recent commits is automatically
45 sent to the list at most daily. The list address is fio-devel@kernel.dk,
46 subscribe by sending an empty email to fio-devel+subscribe@kernel.dk.
52 Just type 'make' and 'make install'. If on FreeBSD, for now you have to
53 specify the FreeBSD Makefile with -f, eg:
55 $ make -f Makefile.Freebsd && make -f Makefile.FreeBSD install
57 Likewise with OpenSolaris, use the Makefile.solaris to compile there.
58 This might change in the future if I opt for an autoconf type setup.
65 --output Write output to file
66 --runtime Runtime in seconds
67 --latency-log Generate per-job latency logs
68 --bandwidth-log Generate per-job bandwidth logs
69 --minimal Minimal (terse) output
70 --version Print version info and exit
71 --help Print this page
72 --cmdhelp=cmd Print command help, "all" for all of them
73 --showcmd Turn a job file into command line options
74 --readonly Turn on safety read-only checks
76 Any parameters following the options will be assumed to be job files,
77 unless they match a job file parameter. You can add as many as you want,
78 each job file will be regarded as a separate group and fio will stonewall
81 The --readonly switch is an extra safety guard to prevent accidentically
82 turning on a write setting when that is not desired. Fio will only write
83 if rw=write/randwrite/rw/randrw is given, but this extra safety net can
84 be used as an extra precaution. It will also enable a write check in the
85 io engine core to prevent an accidental write due to a fio bug.
91 See the HOWTO file for a more detailed description of parameters and what
92 they mean. This file contains the terse version. You can describe big and
93 complex setups with the command line, but generally it's a lot easier to
94 just write a simple job file to describe the workload. The job file format
95 is in the ini style format, as that is easy to read and write for the user.
97 The job file parameters are:
99 name=x Use 'x' as the identifier for this job.
100 description=x 'x' is a text description of the job.
101 directory=x Use 'x' as the top level directory for storing files
102 filename=x Force the use of 'x' as the filename for all files
103 in this thread. If not given, fio will make up
104 a suitable filename based on the thread and file
106 rw=x 'x' may be: read, randread, write, randwrite,
107 rw (read-write mix), randrw (read-write random mix)
108 rwmixcycle=x Base cycle for switching between read and write
110 rwmixread=x 'x' percentage of rw mix ios will be reads. If
111 rwmixwrite is also given, the last of the two will
112 be used if they don't add up to 100%.
113 rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
115 rand_repeatable=x The sequence of random io blocks can be repeatable
116 across runs, if 'x' is 1.
117 size=x Set file size to x bytes (x string can include k/m/g)
118 ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
119 posixaio for POSIX aio, sync for regular read/write io,
120 mmap for mmap'ed io, syslet-rw for syslet driven
121 read/write, splice for using splice/vmsplice,
122 sgio for direct SG_IO io, net for network io, or cpuio
123 for a cycler burner load. sgio only works on Linux on
124 SCSI (or SCSI-like devices, such as usb-storage or
125 sata/libata driven) devices. Fio also has a null io
126 engine, which is mainly used for testing fio itself.
127 iodepth=x For async io, allow 'x' ios in flight
128 overwrite=x If 'x', layout a write file first.
129 nrfiles=x Spread io load over 'x' number of files per job,
131 prio=x Run io at prio X, 0-7 is the kernel allowed range
132 prioclass=x Run io at prio class X
133 bs=x Use 'x' for thread blocksize. May include k/m postfix.
134 bsrange=x-y Mix thread block sizes randomly between x and y. May
135 also include k/m postfix.
136 direct=x 1 for direct IO, 0 for buffered IO
137 thinktime=x "Think" x usec after each io
138 rate=x Throttle rate to x KiB/sec
139 ratemin=x Quit if rate of x KiB/sec can't be met
140 ratecycle=x ratemin averaged over x msecs
141 cpumask=x Only allow job to run on CPUs defined by mask.
142 cpus_allowed=x Like 'cpumask', but allow text setting of CPU affinity.
143 fsync=x If writing with buffered IO, fsync after every
144 'x' blocks have been written.
145 end_fsync=x If 'x', run fsync() after end-of-job.
146 startdelay=x Start this thread x seconds after startup
147 runtime=x Terminate x seconds after startup. Can include a
148 normal time suffix if not given in seconds, such as
149 'm' for minutes, 'h' for hours, and 'd' for days.
150 offset=x Start io at offset x (x string can include k/m/g)
151 invalidate=x Invalidate page cache for file prior to doing io
152 sync=x Use sync writes if x and writing buffered IO.
153 mem=x If x == malloc, use malloc for buffers. If x == shm,
154 use shared memory for buffers. If x == mmap, use
156 exitall When one thread quits, terminate the others
157 bwavgtime=x Average bandwidth stats over an x msec window.
158 create_serialize=x If 'x', serialize file creation.
159 create_fsync=x If 'x', run fsync() after file creation.
160 unlink If set, unlink files when done.
161 loops=x Run the job 'x' number of times.
162 verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
163 use crc32 for verifies. md5 is 'safer', but crc32 is
164 a lot faster. Only makes sense for writing to a file.
165 stonewall Wait for preceeding jobs to end before running.
166 numjobs=x Create 'x' similar entries for this job
167 thread Use pthreads instead of forked jobs
169 zoneskip=y Zone options must be paired. If given, the job
170 will skip y bytes for every x read/written. This
171 can be used to gauge hard drive speed over the entire
172 platter, without reading everything. Both x/y can
173 include k/m/g suffix.
174 iolog=x Open and read io pattern from file 'x'. The file must
175 contain one io action per line in the following format:
177 where with rw=0/1 for read/write, and the offset
178 and length entries being in bytes.
179 write_iolog=x Write an iolog to file 'x' in the same format as iolog.
180 The iolog options are exclusive, if both given the
181 read iolog will be performed.
182 write_bw_log Write a bandwidth log.
183 write_lat_log Write a latency log.
184 lockmem=x Lock down x amount of memory on the machine, to
185 simulate a machine with less memory available. x can
186 include k/m/g suffix.
187 nice=x Run job at given nice value.
188 exec_prerun=x Run 'x' before job io is begun.
189 exec_postrun=x Run 'x' after job io has finished.
190 ioscheduler=x Use ioscheduler 'x' for this job.
191 cpuload=x For a CPU io thread, percentage of CPU time to attempt
193 cpuchunks=x Split burn cycles into pieces of x usecs.
199 Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
200 of the Linux IO subsystem and schedulers. He got tired of writing
201 specific test applications to simulate a given workload, and found that
202 the existing io benchmark/test tools out there weren't flexible enough
203 to do what he wanted.
205 Jens Axboe <axboe@kernel.dk> 20060905