[disktools.git] / README.fio

fio
---

fio is a tool that will spawn a number of thread doing a particular
type of io action as specified by the user. fio takes a number of
global parameters, each inherited by the thread unless otherwise
parameters given to them overriding that setting is given.

Options
-------

$ fio
	-s IO is sequential
	-b block size in KiB for each io
	-t <sec> Runtime in seconds
	-r For random io, sequence must be repeatable
	-R <on> If one thread fails to meet rate, quit all
	-o <on> Use direct IO is 1, buffered if 0
	-l Generate per-job latency logs
	-w Generate per-job bandwidth logs
	-f <file> Read <file> for job descriptions

The <jobs> format is as follows:

	directory=x	Use 'x' as the top level directory for storing files
	rw=x		'x' may be: read, randread, write, or randwrite
	overwrite=x	If 'x', layout a write file first.
	prio=x		Run io at prio X, 0-7 is the kernel allowed range
	prioclass=x	Run io at prio class X
	file=foo	Do the io to file foo
	bs=x		Thread blocksize is x bytes
	bsrange=x-y	Mix thread block sizes randomly between x and y
	direct=x	1 for direct IO, 0 for buffered IO
	thinktime=x	"Think" x usec after each io
	rate=x		Throttle rate to x KiB/sec
	ratemin=x	Quit if rate of x KiB/sec can't be met
	ratecycle=x	ratemin averaged over x msecs
	cpumask=x	Allow job to run on CPUs defined by mask
	fsync=x		If writing, fsync after every x blocks have been written
	startdelay=x	Start this thread x seconds after startup
	timeout=x	Terminate x seconds after startup
	aio		Use Linux async io
	aio_depth=x	Allow x iocbs in flight
	size=x		Set file size to x bytes (x string can include k/m/g)
	offset=x	Start io at offset x (x string can include k/m/g)
	invalidate=x	Invalidate page cache for file prior to doing io
	sync=x		Use sync writes if x and writing
	mem=x		If x == malloc, use malloc for buffers. If x == shm,
			use shm for buffers. If x == mmap, use anon mmap.
	exitall		When one thread quits, terminate the others
	bwavgtime=x	Average bandwidth stats over x msec
	create_serialize=x	If 'x', serialize file creation.
	create_fsync=x	If 'x', run fsync() after file creation.
	loops=x		Run the job 'x' number of times.
	verify=x	If 'x' == md5, use md5 for verifies. If 'x' == crc32,
			use crc32 for verifies.
	stonewall	Wait for preceeding jobs to end before running.
	numjobs=x	Create 'x' similar entries for this job
	mmap=x		If 'x', use mmap for data transfers


Examples using a job file
-------------------------

A sample job file doing the same as above would look like this:

[read_file]
rw=0
bs=4096

[write_file]
rw=1
bs=16384

And fio would be invoked as:

$ fio -o1 -s -f file_with_above

The second example would look like this:

[rf1]
rw=0
prio=6

[rf2]
rw=0
prio=3

[rf3]
rw=0
prio=0
direct=1

And fio would be invoked as:

$ fio -o0 -s -b4096 -f file_with_above

'global' is a reserved keyword. When used as the filename, it sets the
default options for the threads following that section. It is possible
to have more than one global section in the file, as it only affects
subsequent jobs.

Also see the examples/ dir for sample job files.
Commit	Line	Data
e709787c JA	1	fio
	2	---
	3
	4	fio is a tool that will spawn a number of thread doing a particular
	5	type of io action as specified by the user. fio takes a number of
	6	global parameters, each inherited by the thread unless otherwise
	7	parameters given to them overriding that setting is given.
	8
	9	Options
	10	-------
	11
	12	$ fio
	13	-s IO is sequential
86184d14	14	-b block size in KiB for each io
e709787c	15	-t <sec> Runtime in seconds
e709787c	16	-r For random io, sequence must be repeatable
02bdd9ba	17	-R <on> If one thread fails to meet rate, quit all
e709787c	18	-o <on> Use direct IO is 1, buffered if 0
a0a9b35b JA	19	-l Generate per-job latency logs
a0a9b35b JA	20	-w Generate per-job bandwidth logs
7dd1389e	21	-f <file> Read <file> for job descriptions
e709787c	22
e709787c JA	23	The <jobs> format is as follows:
e709787c JA	24
fe74f555	25	directory=x Use 'x' as the top level directory for storing files
d2d598b2	26	rw=x 'x' may be: read, randread, write, or randwrite
e8403a9c	27	overwrite=x If 'x', layout a write file first.
e709787c JA	28	prio=x Run io at prio X, 0-7 is the kernel allowed range
	29	prioclass=x Run io at prio class X
	30	file=foo Do the io to file foo
8c033f93	31	bs=x Thread blocksize is x bytes
7889f07b	32	bsrange=x-y Mix thread block sizes randomly between x and y
e709787c	33	direct=x 1 for direct IO, 0 for buffered IO
e97712ed	34	thinktime=x "Think" x usec after each io
86184d14	35	rate=x Throttle rate to x KiB/sec
4240cfa1 JA	36	ratemin=x Quit if rate of x KiB/sec can't be met
4240cfa1 JA	37	ratecycle=x ratemin averaged over x msecs
18e0b78c	38	cpumask=x Allow job to run on CPUs defined by mask
4240cfa1	39	fsync=x If writing, fsync after every x blocks have been written
fc24389f	40	startdelay=x Start this thread x seconds after startup
67903a2e	41	timeout=x Terminate x seconds after startup
43000118 JA	42	aio Use Linux async io
43000118 JA	43	aio_depth=x Allow x iocbs in flight
02983297 JA	44	size=x Set file size to x bytes (x string can include k/m/g)
02983297 JA	45	offset=x Start io at offset x (x string can include k/m/g)
b95799ca	46	invalidate=x Invalidate page cache for file prior to doing io
74b4b5fb	47	sync=x Use sync writes if x and writing
99c6704f	48	mem=x If x == malloc, use malloc for buffers. If x == shm,
891e70f8	49	use shm for buffers. If x == mmap, use anon mmap.
98dd52d6	50	exitall When one thread quits, terminate the others
1d035750	51	bwavgtime=x Average bandwidth stats over x msec
b6794fbf JA	52	create_serialize=x If 'x', serialize file creation.
	53	create_fsync=x If 'x', run fsync() after file creation.
	54	loops=x Run the job 'x' number of times.
7f46ef08 JA	55	verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
7f46ef08 JA	56	use crc32 for verifies.
2a81240d	57	stonewall Wait for preceeding jobs to end before running.
fe74f555	58	numjobs=x Create 'x' similar entries for this job
6e2c38cc	59	mmap=x If 'x', use mmap for data transfers
e709787c JA	60
e709787c JA	61
7dd1389e JA	62	Examples using a job file
	63	-------------------------
	64
	65	A sample job file doing the same as above would look like this:
	66
	67	[read_file]
	68	rw=0
	69	bs=4096
	70
	71	[write_file]
	72	rw=1
	73	bs=16384
	74
	75	And fio would be invoked as:
	76
	77	$ fio -o1 -s -f file_with_above
	78
	79	The second example would look like this:
	80
	81	[rf1]
	82	rw=0
	83	prio=6
	84
	85	[rf2]
	86	rw=0
	87	prio=3
	88
	89	[rf3]
	90	rw=0
	91	prio=0
	92	direct=1
	93
	94	And fio would be invoked as:
	95
	96	$ fio -o0 -s -b4096 -f file_with_above
	97
47d45203 JA	98	'global' is a reserved keyword. When used as the filename, it sets the
	99	default options for the threads following that section. It is possible
	100	to have more than one global section in the file, as it only affects
	101	subsequent jobs.
ed864e24 JA	102
ed864e24 JA	103	Also see the examples/ dir for sample job files.
47d45203	104