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://brick.kernel.dk/data/git/fio.git
19 Snapshots are frequently generated and they include the git meta data as
20 well. You can download them here:
22 http://brick.kernel.dk/snaps/
24 Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository, you
27 http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio
33 Just type 'make' and 'make install'. If on FreeBSD, for now you have to
34 specify the FreeBSD Makefile with -f, eg:
36 $ make -f Makefile.Freebsd && make -f Makefile.FreeBSD install
38 Likewise with OpenSolaris, use the Makefile.solaris to compile there.
39 This might change in the future if I opt for an autoconf type setup.
46 -t <sec> Runtime in seconds
47 -l Generate per-job latency logs
48 -w Generate per-job bandwidth logs
49 -o <file> Log output to file
50 -m Minimal (terse) output
52 -v Print version information and exit
54 Any parameters following the options will be assumed to be job files.
55 You can add as many as you want, each job file will be regarded as a
56 separate group and fio will stonewall it's execution.
62 See the HOWTO file for a more detailed description of parameters and what
63 they mean. This file contains the terse version. Only a few options can
64 be controlled with command line parameters, generally it's a lot easier to
65 just write a simple job file to describe the workload. The job file format
66 is in the ini style format, as it's easy to read and write for the user.
68 The job file parameters are:
70 name=x Use 'x' as the identifier for this job.
71 directory=x Use 'x' as the top level directory for storing files
72 filename=x Force the use of 'x' as the filename for all files
73 in this thread. If not given, fio will make up
74 a suitable filename based on the thread and file
76 rw=x 'x' may be: read, randread, write, randwrite,
77 rw (read-write mix), randrw (read-write random mix)
78 rwmixcycle=x Base cycle for switching between read and write
80 rwmixread=x 'x' percentage of rw mix ios will be reads. If
81 rwmixwrite is also given, the last of the two will
82 be used if they don't add up to 100%.
83 rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
85 rand_repeatable=x The sequence of random io blocks can be repeatable
86 across runs, if 'x' is 1.
87 size=x Set file size to x bytes (x string can include k/m/g)
88 ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
89 posixaio for POSIX aio, sync for regular read/write io,
90 mmap for mmap'ed io, splice for using splice/vmsplice,
91 or sgio for direct SG_IO io. The latter only works on
92 Linux on SCSI (or SCSI-like devices, such as
93 usb-storage or sata/libata driven) devices.
94 iodepth=x For async io, allow 'x' ios in flight
95 overwrite=x If 'x', layout a write file first.
96 nrfiles=x Spread io load over 'x' number of files per job,
98 prio=x Run io at prio X, 0-7 is the kernel allowed range
99 prioclass=x Run io at prio class X
100 bs=x Use 'x' for thread blocksize. May include k/m postfix.
101 bsrange=x-y Mix thread block sizes randomly between x and y. May
102 also include k/m postfix.
103 direct=x 1 for direct IO, 0 for buffered IO
104 thinktime=x "Think" x usec after each io
105 rate=x Throttle rate to x KiB/sec
106 ratemin=x Quit if rate of x KiB/sec can't be met
107 ratecycle=x ratemin averaged over x msecs
108 cpumask=x Only allow job to run on CPUs defined by mask.
109 fsync=x If writing with buffered IO, fsync after every
110 'x' blocks have been written.
111 end_fsync=x If 'x', run fsync() after end-of-job.
112 startdelay=x Start this thread x seconds after startup
113 timeout=x Terminate x seconds after startup. Can include a
114 normal time suffix if not given in seconds, such as
115 'm' for minutes, 'h' for hours, and 'd' for days.
116 offset=x Start io at offset x (x string can include k/m/g)
117 invalidate=x Invalidate page cache for file prior to doing io
118 sync=x Use sync writes if x and writing buffered IO.
119 mem=x If x == malloc, use malloc for buffers. If x == shm,
120 use shared memory for buffers. If x == mmap, use
122 exitall When one thread quits, terminate the others
123 bwavgtime=x Average bandwidth stats over an x msec window.
124 create_serialize=x If 'x', serialize file creation.
125 create_fsync=x If 'x', run fsync() after file creation.
126 unlink If set, unlink files when done.
127 loops=x Run the job 'x' number of times.
128 verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
129 use crc32 for verifies. md5 is 'safer', but crc32 is
130 a lot faster. Only makes sense for writing to a file.
131 stonewall Wait for preceeding jobs to end before running.
132 numjobs=x Create 'x' similar entries for this job
133 thread Use pthreads instead of forked jobs
135 zoneskip=y Zone options must be paired. If given, the job
136 will skip y bytes for every x read/written. This
137 can be used to gauge hard drive speed over the entire
138 platter, without reading everything. Both x/y can
139 include k/m/g suffix.
140 iolog=x Open and read io pattern from file 'x'. The file must
141 contain one io action per line in the following format:
143 where with rw=0/1 for read/write, and the offset
144 and length entries being in bytes.
145 write_iolog=x Write an iolog to file 'x' in the same format as iolog.
146 The iolog options are exclusive, if both given the
147 read iolog will be performed.
148 write_bw_log Write a bandwidth log.
149 write_lat_log Write a latency log.
150 lockmem=x Lock down x amount of memory on the machine, to
151 simulate a machine with less memory available. x can
152 include k/m/g suffix.
153 nice=x Run job at given nice value.
154 exec_prerun=x Run 'x' before job io is begun.
155 exec_postrun=x Run 'x' after job io has finished.
156 ioscheduler=x Use ioscheduler 'x' for this job.
157 cpuload=x For a CPU io thread, percentage of CPU time to attempt
159 cpuchunks=x Split burn cycles into pieces of x.
165 Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
166 of the Linux IO subsystem and schedulers. He got tired of writing
167 specific test applications to simulate a given workload, and found that
168 the existing io benchmark/test tools out there weren't flexible enough
169 to do what he wanted.
171 Jens Axboe <axboe@kernel.dk> 20060905