[fio.git] / README

fio
---

fio is a tool that will spawn a number of threads or processes 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.
The typical use of fio is to write a job file matching the io load
one wants to simulate.


Source
------

fio resides in a git repo, the canonical place is:

git://brick.kernel.dk/data/git/fio.git

Snapshots are frequently generated and they include the git meta data as
well. You can download them here:

http://brick.kernel.dk/snaps/

Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository, you
can find them here:

http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio


Mailing list
------------

There's a mailing list associated with fio. It's meant for general
discussion, bug reporting, questions - basically anything that has to
do with fio. An automated mail detailing recent commits is automatically
sent to the list at most daily. The list address is fio-devel@kernel.dk,
subscribe by sending an empty email to fio-devel+subscribe@kernel.dk.


Building
--------

Just type 'make' and 'make install'. If on FreeBSD, for now you have to
specify the FreeBSD Makefile with -f, eg:

$ make -f Makefile.Freebsd && make -f Makefile.FreeBSD install

Likewise with OpenSolaris, use the Makefile.solaris to compile there.
This might change in the future if I opt for an autoconf type setup.


Command line
------------

$ fio
        --output        Write output to file
        --timeout       Runtime in seconds
        --latency-log   Generate per-job latency logs
        --bandwidth-log Generate per-job bandwidth logs
        --minimal       Minimal (terse) output
        --version       Print version info and exit

Any parameters following the options will be assumed to be job files,
unless they match a job file parameter. You can add as many as you want,
each job file will be regarded as a separate group and fio will stonewall
its execution.


Job file
--------

See the HOWTO file for a more detailed description of parameters and what
they mean. This file contains the terse version.  Only a few options can
be controlled with command line parameters, generally it's a lot easier to
just write a simple job file to describe the workload. The job file format
is in the ini style format, as it's easy to read and write for the user.

The job file parameters are:

	name=x		Use 'x' as the identifier for this job.
	directory=x	Use 'x' as the top level directory for storing files
	filename=x	Force the use of 'x' as the filename for all files
			in this thread. If not given, fio will make up
			a suitable filename based on the thread and file
			number.
	rw=x		'x' may be: read, randread, write, randwrite,
			rw (read-write mix), randrw (read-write random mix)
	rwmixcycle=x	Base cycle for switching between read and write
			in msecs.
	rwmixread=x	'x' percentage of rw mix ios will be reads. If
			rwmixwrite is also given, the last of the two will
			 be used if they don't add up to 100%.
	rwmixwrite=x	'x' percentage of rw mix ios will be writes. See
			rwmixread.
	rand_repeatable=x  The sequence of random io blocks can be repeatable
			across runs, if 'x' is 1.
	size=x		Set file size to x bytes (x string can include k/m/g)
	ioengine=x	'x' may be: aio/libaio/linuxaio for Linux aio,
			posixaio for POSIX aio, sync for regular read/write io,
			mmap for mmap'ed io, splice for using splice/vmsplice,
			or sgio for direct SG_IO io. The latter only works on
			Linux on SCSI (or SCSI-like devices, such as
			usb-storage or sata/libata driven) devices. Fio also
			has a null io engine, which is mainly used for testing
			fio itself.
	iodepth=x	For async io, allow 'x' ios in flight
	overwrite=x	If 'x', layout a write file first.
	nrfiles=x	Spread io load over 'x' number of files per job,
			if possible.
	prio=x		Run io at prio X, 0-7 is the kernel allowed range
	prioclass=x	Run io at prio class X
	bs=x		Use 'x' for thread blocksize. May include k/m postfix.
	bsrange=x-y	Mix thread block sizes randomly between x and y. May
			also include k/m postfix.
	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	Only allow job to run on CPUs defined by mask.
	fsync=x		If writing with buffered IO, fsync after every
			'x' blocks have been written.
	end_fsync=x	If 'x', run fsync() after end-of-job.
	startdelay=x	Start this thread x seconds after startup
	timeout=x	Terminate x seconds after startup. Can include a
			normal time suffix if not given in seconds, such as
			'm' for minutes, 'h' for hours, and 'd' for days.
	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 buffered IO.
	mem=x		If x == malloc, use malloc for buffers. If x == shm,
			use shared memory for buffers. If x == mmap, use
			anonymous mmap.
	exitall		When one thread quits, terminate the others
	bwavgtime=x	Average bandwidth stats over an x msec window.
	create_serialize=x	If 'x', serialize file creation.
	create_fsync=x	If 'x', run fsync() after file creation.
	unlink		If set, unlink files when done.
	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. md5 is 'safer', but crc32 is
			a lot faster. Only makes sense for writing to a file.
	stonewall	Wait for preceeding jobs to end before running.
	numjobs=x	Create 'x' similar entries for this job
	thread		Use pthreads instead of forked jobs
	zonesize=x
	zoneskip=y	Zone options must be paired. If given, the job
			will skip y bytes for every x read/written. This
			can be used to gauge hard drive speed over the entire
			platter, without reading everything. Both x/y can
			include k/m/g suffix.
	iolog=x		Open and read io pattern from file 'x'. The file must
			contain one io action per line in the following format:
			rw, offset, length
			where with rw=0/1 for read/write, and the offset
			and length entries being in bytes.
	write_iolog=x	Write an iolog to file 'x' in the same format as iolog.
			The iolog options are exclusive, if both given the
			read iolog will be performed.
	write_bw_log	Write a bandwidth log.
	write_lat_log	Write a latency log.
	lockmem=x	Lock down x amount of memory on the machine, to
			simulate a machine with less memory available. x can
			include k/m/g suffix.
	nice=x		Run job at given nice value.
	exec_prerun=x	Run 'x' before job io is begun.
	exec_postrun=x	Run 'x' after job io has finished.
	ioscheduler=x	Use ioscheduler 'x' for this job.
	cpuload=x	For a CPU io thread, percentage of CPU time to attempt
			to burn.
	cpuchunks=x	Split burn cycles into pieces of x.


Author
------

Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
of the Linux IO subsystem and schedulers. He got tired of writing
specific test applications to simulate a given workload, and found that
the existing io benchmark/test tools out there weren't flexible enough
to do what he wanted.

Jens Axboe <axboe@kernel.dk> 20060905
Commit	Line	Data
	1	fio
	2	---
	3
	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
	9	one wants to simulate.
	10
	11
	12	Source
	13	------
	14
	15	fio resides in a git repo, the canonical place is:
	16
	17	git://brick.kernel.dk/data/git/fio.git
	18
	19	Snapshots are frequently generated and they include the git meta data as
	20	well. You can download them here:
	21
	22	http://brick.kernel.dk/snaps/
	23
	24	Pascal Bleser <guru@unixtech.be> has fio RPMs in his repository, you
	25	can find them here:
	26
	27	http://linux01.gwdg.de/~pbleser/rpm-navigation.php?cat=System/fio
	28
	29
	30	Mailing list
	31	------------
	32
	33	There's a mailing list associated with fio. It's meant for general
	34	discussion, bug reporting, questions - basically anything that has to
	35	do with fio. An automated mail detailing recent commits is automatically
	36	sent to the list at most daily. The list address is fio-devel@kernel.dk,
	37	subscribe by sending an empty email to fio-devel+subscribe@kernel.dk.
	38
	39
	40	Building
	41	--------
	42
	43	Just type 'make' and 'make install'. If on FreeBSD, for now you have to
	44	specify the FreeBSD Makefile with -f, eg:
	45
	46	$ make -f Makefile.Freebsd && make -f Makefile.FreeBSD install
	47
	48	Likewise with OpenSolaris, use the Makefile.solaris to compile there.
	49	This might change in the future if I opt for an autoconf type setup.
	50
	51
	52	Command line
	53	------------
	54
	55	$ fio
	56	--output Write output to file
	57	--timeout Runtime in seconds
	58	--latency-log Generate per-job latency logs
	59	--bandwidth-log Generate per-job bandwidth logs
	60	--minimal Minimal (terse) output
	61	--version Print version info and exit
	62
	63	Any parameters following the options will be assumed to be job files,
	64	unless they match a job file parameter. You can add as many as you want,
	65	each job file will be regarded as a separate group and fio will stonewall
	66	its execution.
	67
	68
	69	Job file
	70	--------
	71
	72	See the HOWTO file for a more detailed description of parameters and what
	73	they mean. This file contains the terse version. Only a few options can
	74	be controlled with command line parameters, generally it's a lot easier to
	75	just write a simple job file to describe the workload. The job file format
	76	is in the ini style format, as it's easy to read and write for the user.
	77
	78	The job file parameters are:
	79
	80	name=x Use 'x' as the identifier for this job.
	81	directory=x Use 'x' as the top level directory for storing files
	82	filename=x Force the use of 'x' as the filename for all files
	83	in this thread. If not given, fio will make up
	84	a suitable filename based on the thread and file
	85	number.
	86	rw=x 'x' may be: read, randread, write, randwrite,
	87	rw (read-write mix), randrw (read-write random mix)
	88	rwmixcycle=x Base cycle for switching between read and write
	89	in msecs.
	90	rwmixread=x 'x' percentage of rw mix ios will be reads. If
	91	rwmixwrite is also given, the last of the two will
	92	be used if they don't add up to 100%.
	93	rwmixwrite=x 'x' percentage of rw mix ios will be writes. See
	94	rwmixread.
	95	rand_repeatable=x The sequence of random io blocks can be repeatable
	96	across runs, if 'x' is 1.
	97	size=x Set file size to x bytes (x string can include k/m/g)
	98	ioengine=x 'x' may be: aio/libaio/linuxaio for Linux aio,
	99	posixaio for POSIX aio, sync for regular read/write io,
	100	mmap for mmap'ed io, splice for using splice/vmsplice,
	101	or sgio for direct SG_IO io. The latter only works on
	102	Linux on SCSI (or SCSI-like devices, such as
	103	usb-storage or sata/libata driven) devices. Fio also
	104	has a null io engine, which is mainly used for testing
	105	fio itself.
	106	iodepth=x For async io, allow 'x' ios in flight
	107	overwrite=x If 'x', layout a write file first.
	108	nrfiles=x Spread io load over 'x' number of files per job,
	109	if possible.
	110	prio=x Run io at prio X, 0-7 is the kernel allowed range
	111	prioclass=x Run io at prio class X
	112	bs=x Use 'x' for thread blocksize. May include k/m postfix.
	113	bsrange=x-y Mix thread block sizes randomly between x and y. May
	114	also include k/m postfix.
	115	direct=x 1 for direct IO, 0 for buffered IO
	116	thinktime=x "Think" x usec after each io
	117	rate=x Throttle rate to x KiB/sec
	118	ratemin=x Quit if rate of x KiB/sec can't be met
	119	ratecycle=x ratemin averaged over x msecs
	120	cpumask=x Only allow job to run on CPUs defined by mask.
	121	fsync=x If writing with buffered IO, fsync after every
	122	'x' blocks have been written.
	123	end_fsync=x If 'x', run fsync() after end-of-job.
	124	startdelay=x Start this thread x seconds after startup
	125	timeout=x Terminate x seconds after startup. Can include a
	126	normal time suffix if not given in seconds, such as
	127	'm' for minutes, 'h' for hours, and 'd' for days.
	128	offset=x Start io at offset x (x string can include k/m/g)
	129	invalidate=x Invalidate page cache for file prior to doing io
	130	sync=x Use sync writes if x and writing buffered IO.
	131	mem=x If x == malloc, use malloc for buffers. If x == shm,
	132	use shared memory for buffers. If x == mmap, use
	133	anonymous mmap.
	134	exitall When one thread quits, terminate the others
	135	bwavgtime=x Average bandwidth stats over an x msec window.
	136	create_serialize=x If 'x', serialize file creation.
	137	create_fsync=x If 'x', run fsync() after file creation.
	138	unlink If set, unlink files when done.
	139	loops=x Run the job 'x' number of times.
	140	verify=x If 'x' == md5, use md5 for verifies. If 'x' == crc32,
	141	use crc32 for verifies. md5 is 'safer', but crc32 is
	142	a lot faster. Only makes sense for writing to a file.
	143	stonewall Wait for preceeding jobs to end before running.
	144	numjobs=x Create 'x' similar entries for this job
	145	thread Use pthreads instead of forked jobs
	146	zonesize=x
	147	zoneskip=y Zone options must be paired. If given, the job
	148	will skip y bytes for every x read/written. This
	149	can be used to gauge hard drive speed over the entire
	150	platter, without reading everything. Both x/y can
	151	include k/m/g suffix.
	152	iolog=x Open and read io pattern from file 'x'. The file must
	153	contain one io action per line in the following format:
	154	rw, offset, length
	155	where with rw=0/1 for read/write, and the offset
	156	and length entries being in bytes.
	157	write_iolog=x Write an iolog to file 'x' in the same format as iolog.
	158	The iolog options are exclusive, if both given the
	159	read iolog will be performed.
	160	write_bw_log Write a bandwidth log.
	161	write_lat_log Write a latency log.
	162	lockmem=x Lock down x amount of memory on the machine, to
	163	simulate a machine with less memory available. x can
	164	include k/m/g suffix.
	165	nice=x Run job at given nice value.
	166	exec_prerun=x Run 'x' before job io is begun.
	167	exec_postrun=x Run 'x' after job io has finished.
	168	ioscheduler=x Use ioscheduler 'x' for this job.
	169	cpuload=x For a CPU io thread, percentage of CPU time to attempt
	170	to burn.
	171	cpuchunks=x Split burn cycles into pieces of x.
	172
	173
	174	Author
	175	------
	176
	177	Fio was written by Jens Axboe <axboe@kernel.dk> to enable flexible testing
	178	of the Linux IO subsystem and schedulers. He got tired of writing
	179	specific test applications to simulate a given workload, and found that
	180	the existing io benchmark/test tools out there weren't flexible enough
	181	to do what he wanted.
	182
	183	Jens Axboe <axboe@kernel.dk> 20060905
	184