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 Only a few options can be controlled with command line parameters,
63 generally it's a lot easier to just write a simple job file to describe
64 the workload. The job file format is in the ini style format, as it's
65 easy to read and write for the user.
67 The job file parameters are:
69 name=x Use 'x' as the identifier for this job.
70 directory=x Use 'x' as the top level directory for storing files
71 filename=x Force the use of 'x' as the filename for all files
72 in this thread. If not given, fio will make up
73 a suitable filename based on the thread and file
75 rw=x 'x' may be: read, randread, write, randwrite,
76 rw (read-write mix), randrw (read-write random mix)
77 rwmixcycle=x Base cycle for switching between read and write
79 rwmixread=x 'x' percentage of rw mix ios will be reads. If
80 rwmixwrite is also given, the last of the two will
81 be used if they don't add up to 100%.
82 rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
84 rand_repeatable=x The sequence of random io blocks can be repeatable
85 across runs, if 'x' is 1.
86 size=x Set file size to x bytes (x string can include k/m/g)
87 ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
88 posixaio for POSIX aio, sync for regular read/write io,
89 mmap for mmap'ed io, splice for using splice/vmsplice,
90 or sgio for direct SG_IO io. The latter only works on
91 Linux on SCSI (or SCSI-like devices, such as
92 usb-storage or sata/libata driven) devices.
93 iodepth=x For async io, allow 'x' ios in flight
94 overwrite=x If 'x', layout a write file first.
95 nrfiles=x Spread io load over 'x' number of files per job,
97 prio=x Run io at prio X, 0-7 is the kernel allowed range
98 prioclass=x Run io at prio class X
99 bs=x Use 'x' for thread blocksize. May include k/m postfix.
100 bsrange=x-y Mix thread block sizes randomly between x and y. May
101 also include k/m postfix.
102 direct=x 1 for direct IO, 0 for buffered IO
103 thinktime=x "Think" x usec after each io
104 rate=x Throttle rate to x KiB/sec
105 ratemin=x Quit if rate of x KiB/sec can't be met
106 ratecycle=x ratemin averaged over x msecs
107 cpumask=x Only allow job to run on CPUs defined by mask.
108 fsync=x If writing, fsync after every x blocks have been written
109 startdelay=x Start this thread x seconds after startup
110 timeout=x Terminate x seconds after startup. Can include a
111 normal time suffix if not given in seconds, such as
112 'm' for minutes, 'h' for hours, and 'd' for days.
113 offset=x Start io at offset x (x string can include k/m/g)
114 invalidate=x Invalidate page cache for file prior to doing io
115 sync=x Use sync writes if x and writing
116 mem=x If x == malloc, use malloc for buffers. If x == shm,
117 use shm for buffers. If x == mmap, use anon mmap.
118 exitall When one thread quits, terminate the others
119 bwavgtime=x Average bandwidth stats over an x msec window.
120 create_serialize=x If 'x', serialize file creation.
121 create_fsync=x If 'x', run fsync() after file creation.
122 unlink If set, unlink files when done.
123 end_fsync=x If 'x', run fsync() after end-of-job.
124 loops=x Run the job 'x' number of times.
125 verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
126 use crc32 for verifies. md5 is 'safer', but crc32 is
127 a lot faster. Only makes sense for writing to a file.
128 stonewall Wait for preceeding jobs to end before running.
129 numjobs=x Create 'x' similar entries for this job
130 thread Use pthreads instead of forked jobs
132 zoneskip=y Zone options must be paired. If given, the job
133 will skip y bytes for every x read/written. This
134 can be used to gauge hard drive speed over the entire
135 platter, without reading everything. Both x/y can
136 include k/m/g suffix.
137 iolog=x Open and read io pattern from file 'x'. The file must
138 contain one io action per line in the following format:
140 where with rw=0/1 for read/write, and the offset
141 and length entries being in bytes.
142 write_iolog=x Write an iolog to file 'x' in the same format as iolog.
143 The iolog options are exclusive, if both given the
144 read iolog will be performed.
145 write_bw_log Write a bandwidth log.
146 write_lat_log Write a latency log.
147 lockmem=x Lock down x amount of memory on the machine, to
148 simulate a machine with less memory available. x can
149 include k/m/g suffix.
150 nice=x Run job at given nice value.
151 exec_prerun=x Run 'x' before job io is begun.
152 exec_postrun=x Run 'x' after job io has finished.
153 ioscheduler=x Use ioscheduler 'x' for this job.
154 cpuload=x For a CPU io thread, percentage of CPU time to attempt
156 cpuchunks=x Split burn cycles into pieces of x.
159 Examples using a job file
160 -------------------------
162 Example 1) Two random readers
164 Lets say we want to simulate two threads reading randomly from a file
165 each. They will be doing IO in 4KiB chunks, using raw (O_DIRECT) IO.
166 Since they share most parameters, we'll put those in the [global]
167 section. Job 1 will use a 128MiB file, job 2 will use a 256MiB file.
172 ioengine=sync ; regular read/write(2), the default
185 Generally the [] bracketed name specifies a file name, but the "global"
186 keyword is reserved for setting options that are inherited by each
187 subsequent job description. It's possible to have several [global]
188 sections in the job file, each one adds options that are inherited by
189 jobs defined below it. The name can also point to a block device, such
190 as /dev/sda. To run the above job file, simply do:
194 Example 2) Many random writers
196 Say we want to exercise the IO subsystem some more. We'll define 64
197 threads doing random buffered writes. We'll let each thread use async io
198 with a depth of 4 ios in flight. A job file would then look like this:
215 This will create files.[0-63] and perform the random writes to them.
217 There are endless ways to define jobs, the examples/ directory contains
221 Interpreting the output
222 -----------------------
224 fio spits out a lot of output. While running, fio will display the
225 status of the jobs created. An example of that would be:
227 Threads running: 1: [_r] [24.79% done] [eta 00h:01m:31s]
229 The characters inside the square brackets denote the current status of
230 each thread. The possible values (in typical life cycle order) are:
234 P Thread setup, but not started.
236 I Thread initialized, waiting.
237 R Running, doing sequential reads.
238 r Running, doing random reads.
239 W Running, doing sequential writes.
240 w Running, doing random writes.
241 M Running, doing mixed sequential reads/writes.
242 m Running, doing mixed random reads/writes.
243 F Running, currently waiting for fsync()
244 V Running, doing verification of written data.
245 E Thread exited, not reaped by main thread yet.
248 The other values are fairly self explanatory - number of threads
249 currently running and doing io, and the estimated completion percentage
250 and time for the running group. It's impossible to estimate runtime
251 of the following groups (if any).
253 When fio is done (or interrupted by ctrl-c), it will show the data for
254 each thread, group of threads, and disks in that order. For each data
255 direction, the output looks like:
257 Client1 (g=0): err= 0:
258 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
259 slat (msec): min= 0, max= 136, avg= 0.03, dev= 1.92
260 clat (msec): min= 0, max= 631, avg=48.50, dev=86.82
261 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, dev=681.68
262 cpu : usr=1.49%, sys=0.25%, ctx=7969
264 The client number is printed, along with the group id and error of that
265 thread. Below is the io statistics, here for writes. In the order listed,
268 io= Number of megabytes io performed
269 bw= Average bandwidth rate
270 runt= The runtime of that thread
271 slat= Submission latency (avg being the average, dev being the
272 standard deviation). This is the time it took to submit
273 the io. For sync io, the slat is really the completion
274 latency, since queue/complete is one operation there.
275 clat= Completion latency. Same names as slat, this denotes the
276 time from submission to completion of the io pieces. For
277 sync io, clat will usually be equal (or very close) to 0,
278 as the time from submit to complete is basically just
279 CPU time (io has already been done, see slat explanation).
280 bw= Bandwidth. Same names as the xlat stats, but also includes
281 an approximate percentage of total aggregate bandwidth
282 this thread received in this group. This last value is
283 only really useful if the threads in this group are on the
284 same disk, since they are then competing for disk access.
285 cpu= CPU usage. User and system time, along with the number
286 of context switches this thread went through.
288 After each client has been listed, the group statistics are printed. They
291 Run status group 0 (all jobs):
292 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
293 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
295 For each data direction, it prints:
297 io= Number of megabytes io performed.
298 aggrb= Aggregate bandwidth of threads in this group.
299 minb= The minimum average bandwidth a thread saw.
300 maxb= The maximum average bandwidth a thread saw.
301 mint= The smallest runtime of the threads in that group.
302 maxt= The longest runtime of the threads in that group.
304 And finally, the disk statistics are printed. They will look like this:
306 Disk stats (read/write):
307 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
309 Each value is printed for both reads and writes, with reads first. The
312 ios= Number of ios performed by all groups.
313 merge= Number of merges io the io scheduler.
314 ticks= Number of ticks we kept the disk busy.
315 io_queue= Total time spent in the disk queue.
316 util= The disk utilization. A value of 100% means we kept the disk
317 busy constantly, 50% would be a disk idling half of the time.
323 For scripted usage where you typically want to generate tables or graphs
324 of the results, fio can output the results in a comma seperated format.
325 The format is one long line of values, such as:
327 client1,0,0,936,331,2894,0,0,0.000000,0.000000,1,170,22.115385,34.290410,16,714,84.252874%,366.500000,566.417819,3496,1237,2894,0,0,0.000000,0.000000,0,246,6.671625,21.436952,0,2534,55.465300%,1406.600000,2008.044216,0.000000%,0.431928%,1109
329 Split up, the format is as follows:
331 jobname, groupid, error
333 KiB IO, bandwidth (KiB/sec), runtime (msec)
334 Submission latency: min, max, mean, deviation
335 Completion latency: min, max, mean, deviation
336 Bw: min, max, aggreate percentage of total, mean, deviation
338 KiB IO, bandwidth (KiB/sec), runtime (msec)
339 Submission latency: min, max, mean, deviation
340 Completion latency: min, max, mean, deviation
341 Bw: min, max, aggreate percentage of total, mean, deviation
342 CPU usage: user, system, context switches
348 Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
349 of the Linux IO subsystem and schedulers. He got tired of writing
350 specific test applications to simulate a given workload, and found that
351 the existing io benchmark/test tools out there weren't flexible enough
352 to do what he wanted.
354 Jens Axboe <axboe@kernel.dk> 20060905