[PATCH] Add ioscheduler=x option to allow setting io scheduler
[fio.git] / README
CommitLineData
ebac4655
JA
1fio
2---
3
4fio is a tool that will spawn a number of thread doing a particular
5type of io action as specified by the user. fio takes a number of
6global parameters, each inherited by the thread unless otherwise
7parameters given to them overriding that setting is given.
8
2b02b546
JA
9
10Source
11------
12
13fio resides in a git repo, the canonical place is:
14
15git://brick.kernel.dk/data/git/fio.git
16
17Snapshots are frequently generated as well and they include the git
18meta data as well. You can download them here:
19
20http://brick.kernel.dk/snaps/
21
1053a106
JA
22Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository, you
23can find them here:
24
25http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio
26
2b02b546 27
ebac4655
JA
28Options
29-------
30
31$ fio
32 -s IO is sequential
33 -b block size in KiB for each io
34 -t <sec> Runtime in seconds
35 -r For random io, sequence must be repeatable
36 -R <on> If one thread fails to meet rate, quit all
37 -o <on> Use direct IO is 1, buffered if 0
38 -l Generate per-job latency logs
39 -w Generate per-job bandwidth logs
40 -f <file> Read <file> for job descriptions
4785f995 41 -h Print help info
ebac4655
JA
42 -v Print version information and exit
43
44The <jobs> format is as follows:
45
46 directory=x Use 'x' as the top level directory for storing files
3d60d1ed
JA
47 rw=x 'x' may be: read, randread, write, randwrite,
48 rw (read-write mix), randrw (read-write random mix)
a6ccc7be
JA
49 rwmixcycle=x Base cycle for switching between read and write
50 in msecs.
51 rwmixread=x 'x' percentage of rw mix ios will be reads. If
52 rwmixwrite is also given, the last of the two will
53 be used if they don't add up to 100%.
54 rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
55 rwmixread.
ebac4655
JA
56 size=x Set file size to x bytes (x string can include k/m/g)
57 ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
58 posixaio for POSIX aio, sync for regular read/write io,
8756e4d4
JA
59 mmap for mmap'ed io, splice for using splice/vmsplice,
60 or sgio for direct SG_IO io. The latter only works on
61 Linux on SCSI (or SCSI-like devices, such as
62 usb-storage or sata/libata driven) devices.
ebac4655
JA
63 iodepth=x For async io, allow 'x' ios in flight
64 overwrite=x If 'x', layout a write file first.
65 prio=x Run io at prio X, 0-7 is the kernel allowed range
66 prioclass=x Run io at prio class X
67 bs=x Use 'x' for thread blocksize. May include k/m postfix.
68 bsrange=x-y Mix thread block sizes randomly between x and y. May
69 also include k/m postfix.
70 direct=x 1 for direct IO, 0 for buffered IO
71 thinktime=x "Think" x usec after each io
72 rate=x Throttle rate to x KiB/sec
73 ratemin=x Quit if rate of x KiB/sec can't be met
74 ratecycle=x ratemin averaged over x msecs
75 cpumask=x Only allow job to run on CPUs defined by mask.
76 fsync=x If writing, fsync after every x blocks have been written
77 startdelay=x Start this thread x seconds after startup
78 timeout=x Terminate x seconds after startup
79 offset=x Start io at offset x (x string can include k/m/g)
80 invalidate=x Invalidate page cache for file prior to doing io
81 sync=x Use sync writes if x and writing
82 mem=x If x == malloc, use malloc for buffers. If x == shm,
83 use shm for buffers. If x == mmap, use anon mmap.
84 exitall When one thread quits, terminate the others
85 bwavgtime=x Average bandwidth stats over an x msec window.
86 create_serialize=x If 'x', serialize file creation.
87 create_fsync=x If 'x', run fsync() after file creation.
fc1a4713 88 end_fsync=x If 'x', run fsync() after end-of-job.
ebac4655
JA
89 loops=x Run the job 'x' number of times.
90 verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
91 use crc32 for verifies. md5 is 'safer', but crc32 is
92 a lot faster. Only makes sense for writing to a file.
93 stonewall Wait for preceeding jobs to end before running.
94 numjobs=x Create 'x' similar entries for this job
95 thread Use pthreads instead of forked jobs
20dc95c4
JA
96 zonesize=x
97 zoneskip=y Zone options must be paired. If given, the job
98 will skip y bytes for every x read/written. This
99 can be used to gauge hard drive speed over the entire
100 platter, without reading everything. Both x/y can
101 include k/m/g suffix.
aea47d44
JA
102 iolog=x Open and read io pattern from file 'x'. The file must
103 contain one io action per line in the following format:
104 rw, offset, length
105 where with rw=0/1 for read/write, and the offset
106 and length entries being in bytes.
843a7413
JA
107 write_iolog=x Write an iolog to file 'x' in the same format as iolog.
108 The iolog options are exclusive, if both given the
109 read iolog will be performed.
c04f7ec3
JA
110 lockmem=x Lock down x amount of memory on the machine, to
111 simulate a machine with less memory available. x can
112 include k/m/g suffix.
b6f4d880 113 nice=x Run job at given nice value.
4e0ba8af
JA
114 exec_prerun=x Run 'x' before job io is begun.
115 exec_postrun=x Run 'x' after job io has finished.
da86774e 116 ioscheduler=x Use ioscheduler 'x' for this job.
ebac4655
JA
117
118Examples using a job file
119-------------------------
120
121A sample job file doing the same as above would look like this:
122
123[read_file]
124rw=0
125bs=4096
126
127[write_file]
128rw=1
129bs=16384
130
131And fio would be invoked as:
132
133$ fio -o1 -s -f file_with_above
134
135The second example would look like this:
136
137[rf1]
138rw=0
139prio=6
140
141[rf2]
142rw=0
143prio=3
144
145[rf3]
146rw=0
147prio=0
148direct=1
149
150And fio would be invoked as:
151
152$ fio -o0 -s -b4096 -f file_with_above
153
154'global' is a reserved keyword. When used as the filename, it sets the
155default options for the threads following that section. It is possible
156to have more than one global section in the file, as it only affects
157subsequent jobs.
158
159Also see the examples/ dir for sample job files.
160
161
162Interpreting the output
163-----------------------
164
165fio spits out a lot of output. While running, fio will display the
166status of the jobs created. An example of that would be:
167
168Threads now running: 2 : [ww] [5.73% done]
169
170The characters inside the square brackets denote the current status of
171each thread. The possible values (in typical life cycle order) are:
172
173Idle Run
174---- ---
175P Thread setup, but not started.
176C Thread created and running, but not doing anything yet
177 R Running, doing sequential reads.
178 r Running, doing random reads.
179 W Running, doing sequential writes.
180 w Running, doing random writes.
181V Running, doing verification of written data.
182E Thread exited, not reaped by main thread yet.
183_ Thread reaped.
184
185The other values are fairly self explanatory - number of thread currently
186running and doing io, and the estimated completion percentage.
187
188When fio is done (or interrupted by ctrl-c), it will show the data for
189each thread, group of threads, and disks in that order. For each data
190direction, the output looks like:
191
192Client1 (g=0): err= 0:
193 write: io= 32MiB, bw= 666KiB/s, runt= 50320msec
194 slat (msec): min= 0, max= 136, avg= 0.03, dev= 1.92
195 clat (msec): min= 0, max= 631, avg=48.50, dev=86.82
196 bw (KiB/s) : min= 0, max= 1196, per=51.00%, avg=664.02, dev=681.68
197 cpu : usr=1.49%, sys=0.25%, ctx=7969
198
199The client number is printed, along with the group id and error of that
200thread. Below is the io statistics, here for writes. In the order listed,
201they denote:
202
203io= Number of megabytes io performed
204bw= Average bandwidth rate
205runt= The runtime of that thread
206 slat= Submission latency (avg being the average, dev being the
207 standard deviation). This is the time it took to submit
208 the io. For sync io, the slat is really the completion
209 latency, since queue/complete is one operation there.
210 clat= Completion latency. Same names as slat, this denotes the
211 time from submission to completion of the io pieces. For
212 sync io, clat will usually be equal (or very close) to 0,
213 as the time from submit to complete is basically just
214 CPU time (io has already been done, see slat explanation).
215 bw= Bandwidth. Same names as the xlat stats, but also includes
216 an approximate percentage of total aggregate bandwidth
217 this thread received in this group. This last value is
218 only really useful if the threads in this group are on the
219 same disk, since they are then competing for disk access.
220cpu= CPU usage. User and system time, along with the number
221 of context switches this thread went through.
222
223After each client has been listed, the group statistics are printed. They
224will look like this:
225
226Run status group 0 (all jobs):
227 READ: io=64MiB, aggrb=22178, minb=11355, maxb=11814, mint=2840msec, maxt=2955msec
228 WRITE: io=64MiB, aggrb=1302, minb=666, maxb=669, mint=50093msec, maxt=50320msec
229
230For each data direction, it prints:
231
232io= Number of megabytes io performed.
233aggrb= Aggregate bandwidth of threads in this group.
234minb= The minimum average bandwidth a thread saw.
235maxb= The maximum average bandwidth a thread saw.
236mint= The minimum runtime of a thread.
237maxt= The maximum runtime of a thread.
238
239And finally, the disk statistics are printed. They will look like this:
240
241Disk stats (read/write):
242 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
243
244Each value is printed for both reads and writes, with reads first. The
245numbers denote:
246
247ios= Number of ios performed by all groups.
248merge= Number of merges io the io scheduler.
249ticks= Number of ticks we kept the disk busy.
250io_queue= Total time spent in the disk queue.
251util= The disk utilization. A value of 100% means we kept the disk
252 busy constantly, 50% would be a disk idling half of the time.