mirror_zfs/man/man1/zpios.1
Don Brady 56b3986316 Add large block support to zpios(1) benchmark
As part of the large block support effort, it makes sense to add
support for large blocks to **zpios(1)**. The specifying of a zfs
block size for zpios is optional and will default to 128K if the
block size is not specified.

  `zpios ... -S size | --blocksize size ...`

This will use *size* ZFS blocks for each test, specified as a comma
delimited list with an optional unit suffix. The supported range is
powers of two from 128K through 16M. A range of block sizes can be
tested as follows: `-S 128K,256K,512K,1M`

Example run below
(non realistic results from a VM and output abbreviated for space)

```
 --regioncount=750 --regionsize=8M --chunksize=1M --offset=4K
 --threaddelay=0 --cleanup --human-readable --verbose --cleanup
 --blocksize=128K,256K,512K,1M

 th-cnt  rg-cnt  rg-sz  ch-sz  blksz  wr-data wr-bw   rd-data rd-bw
---------------------------------------------------------------------
 4       750     8m     1m     128k   5g      90.06m  5g      93.37m
 4       750     8m     1m     256k   5g      79.71m  5g      99.81m
 4       750     8m     1m     512k   5g      42.20m  5g      93.14m
 4       750     8m     1m     1m     5g      35.51m  5g      89.36m
 8       750     8m     1m     128k   5g      85.49m  5g      90.81m
 8       750     8m     1m     256k   5g      61.42m  5g      99.24m
 8       750     8m     1m     512k   5g      49.09m  5g     108.78m
 16      750     8m     1m     128k   5g      86.28m  5g      88.73m
 16      750     8m     1m     256k   5g      64.34m  5g      93.47m
 16      750     8m     1m     512k   5g      68.84m  5g     124.47m
 16      750     8m     1m     1m     5g      53.97m  5g      97.20m
---------------------------------------------------------------------
```

Signed-off-by: Don Brady <don.brady@intel.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #3795
Closes #2071
2015-09-22 09:13:20 -07:00

242 lines
7.6 KiB
Groff

'\" t
.\"
.\" CDDL HEADER START
.\"
.\" The contents of this file are subject to the terms of the
.\" Common Development and Distribution License (the "License").
.\" You may not use this file except in compliance with the License.
.\"
.\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
.\" or http://www.opensolaris.org/os/licensing.
.\" See the License for the specific language governing permissions
.\" and limitations under the License.
.\"
.\" When distributing Covered Code, include this CDDL HEADER in each
.\" file and include the License file at usr/src/OPENSOLARIS.LICENSE.
.\" If applicable, add the following below this CDDL HEADER, with the
.\" fields enclosed by brackets "[]" replaced with your own identifying
.\" information: Portions Copyright [yyyy] [name of copyright owner]
.\"
.\" CDDL HEADER END
.\"
.\"
.\" Copyright 2013 Darik Horn <dajhorn@vanadac.com>. All rights reserved.
.\"
.\" Copyright (c) 2015, Intel Corporation.
.\"
.TH zpios 1 "2013 FEB 28" "ZFS on Linux" "User Commands"
.SH NAME
zpios \- Directly test the DMU.
.SH SYNOPSIS
.LP
.BI "zpios [options] <\-p " pool ">"
.SH DESCRIPTION
This utility runs in-kernel DMU performance and stress tests that do
not depend on the ZFS Posix Layer ("ZPL").
.SH OPTIONS
.HP
.BI "\-t" " regex" ", \-\-threadcount" " regex"
.IP
Start this many threads for each test series, specified as a comma
delimited regular expression. (eg: "-t 1,2,3")
.IP
This option is mutually exclusive with the \fBthreadcount_*\fR
options.
.HP
.BI "\-l" " regex_low" ", \-\-threadcount_low" " regex_low"
.HP
.BI "\-h" " regex_high" ", \-\-threadcount_high" " regex_high"
.HP
.BI "\-e" " regex_incr" ", \-\-threadcount_incr" " regex_incr"
.IP
Start \fIregex_low\fR threads for the first test,
add \fIregex_incr\fR threads for each subsequent test,
and start \fIregex_high\fR threads for the last test.
.IP
These three options must be specified together and are mutually
exclusive with the \fBthreadcount\fR option.
.HP
.BI "\-n" " regex" ", \-\-regioncount" " regex"
.IP
Create this many regions for each test series, specified as a comma
delimited regular expression. (eg: "-n 512,4096,65536")
.IP
This option is mutually exclusive with the \fBregioncount_*\fR
options.
.HP
.BI "\-i" " regex_low" ", \-\-regioncount_low" " regex_low"
.HP
.BI "\-j" " regex_high" ", \-\-regioncount_high" " regex_high"
.HP
.BI "\-k" " regex_incr" ", \-\-regioncount_incr" " regex_incr"
.IP
Create \fIregex_low\fR regions for the first test,
add \fIregex_incr\fR regions for each subsequent test, and
create \fIregex_high\fR regions for the last test.
.IP
These three options must be specified together and are mutually
exclusive with the \fBregioncount\fR option.
.HP
.BI "\-o" " size" ", \-\-offset" " size"
.IP
Create regions at \fIsize\fR offset for each test series, specified as
a comma delimited regular expression with an optional unit suffix.
(eg: "-o 4M" means four megabytes.)
.IP
This option is mutually exclusive with the \fBoffset_*\fR options.
.HP
.BI "\-m" " size_low" ", \-\-offset_low" " size_low"
.HP
.BI "\-q" " size_high" ", \-\-offset_high" " size_high"
.HP
.BI "\-r" " size_incr" ", \-\-offset_incr" " size_incr"
.IP
Create a region at \fIsize_low\fR offset for the first test, add
\fIsize_incr\fR to the offset for each subsequent test, and create
a region at \fIsize_high\fR offset for the last test.
.IP
These three options must be specified together and are mutually
exclusive with the \fBoffset\fR option.
.HP
.BI "\-c" " size" ", \-\-chunksize" " size"
.IP
Use \fIsize\fR chunks for each test, specified as a comma delimited
regular expression with an optional unit suffix. (eg: "-c 1M" means
one megabyte.) The chunk size must be at least the region size.
.IP
This option is mutually exclusive with the \fBchunksize_*\fB options.
.HP
.BI "\-a" " size_low" ", \-\-chunksize_low" " size_low"
.HP
.BI "\-b" " size_high" ", \-\-chunksize_high" " size_high"
.HP
.BI "\-g" " size_incr" ", \-\-chunksize_incr" " size_incr"
.IP
Use a \fIsize_low\fR chunk size for the first test, add \fIsize_incr\fR
to the chunk size for each subsequent test, and use a \fIsize_high\fR
chunk size for the last test.
.IP
These three options must be specified together and are mutually
exclusive with the \fBchunksize\fR option.
.HP
.BI "\-s" " size" ", \-\-regionsize" " size"
.IP
Use \fIsize\fR regions for each test, specified as a comma delimited
regular expression with an optional unit suffix. (eg: "-s 1M" means
one megabyte.)
.IP
This option is mutually exclusive with the \fBregionsize_*\fB options.
.HP
.BI "\-A" " size_low" ", \-\-regionsize_low" " size_low"
.HP
.BI "\-B" " size_high" ", \-\-regionsize_high" " size_high"
.HP
.BI "\-C" " size_incr" ", \-\-regionsize_incr" " size_incr"
.IP
Use a \fIsize_low\fR region size for the first test, add \fIsize_incr\fR
to the region size for each subsequent test, and use a \fIsize_high\fR
region size for the last test.
.IP
These three options must be specified together and are mutually
exclusive with the \fBregionsize\fR option.
.HP
.BI "\-S" " size | sizes" ", \-\-blocksize" " size | sizes"
.IP
Use \fIsize\fR ZFS blocks for each test, specified as a comma delimited
regular expression with an optional unit suffix. (eg: "-S 1M" means
one megabyte.) The supported range is powers of two from 128K through 16M.
A range of blocks can be tested as follows: "-S 128K,256K,512K,1M".
.IP
.HP
.BI "\-L" " dmu_flags" ", \-\-load" " dmu_flags"
.IP
Specify \fIdmuio\fR for regular DMU_IO, \fIssf\fR for single shared
file access, or \fIfpp\fR for per thread access. Use commas to delimit
multiple flags. (eg: "-L dmuio,ssf")
.HP
.BI "\-p" " name" ", \-\-pool" " name"
.IP
The pool name, which is mandatory.
.HP
.BI "\-M" " test" ", \-\-name" " test"
.IP
An arbitrary string that appears in the program output.
.HP
.BI "-x, \-\-cleanup"
.IP
Enable the DMU_REMOVE flag.
.HP
.BI "\-P" " command" ", \-\-prerun" " command"
.IP
Invoke \fIcommand\fR from the kernel before running the test. Shell
expansion is not performed and the environment is set to
HOME=/; TERM=linux; PATH=/sbin:/usr/sbin:/bin:/usr/bin.
.HP
.BI "\-R" " command" ", \-\-postrun" " command"
.IP
Invoke \fIcommand\fR from the kernel after running the test. Shell
expansion is not performed and the environment is set to
HOME=/; TERM=linux; PATH=/sbin:/usr/sbin:/bin:/usr/bin.
.HP
.BI "\-G" " directory" ", \-\-log" " directory"
.IP
Put logging output in this directory.
.HP
.BI "\-I" " size" ", \-\-regionnoise" " size"
.IP
Randomly vary the \fBregionsize\fR parameter for each test
modulo \fIsize\fR bytes.
.HP
.BI "\-N" " size" ", \-\-chunknoise" " size"
.IP
Randomly vary the \fBchunksize\fR parameter for each test
modulo \fIsize\fR bytes.
.HP
.BI "\-T" " time" ", \-\-threaddelay" " time"
.IP
Randomly vary the execution time for each test
modulo \fItime\fR kernel jiffies.
.HP
.BI "\-V" "" ", \-\-verify" ""
.IP
Enable the DMU_VERIFY flag for trivial data verification.
.HP
.BI "\-z" "" ", \-\-zerocopy" ""
.IP
Enable the DMU_READ_ZC and DMU_WRITE_ZC flags, which are
currently unimplemented for Linux.
.IP
.HP
.BI "\-O" "" ", \-\-nowait" ""
.IP
Enable the DMU_WRITE_NOWAIT flag.
.HP
.BI "\-f" "" ", \-\-noprefetch" ""
.IP
Enable the DMU_READ_NOPF flag.
.HP
.BI "\-H" "" ", \-\-human\-readable" ""
.IP
Print PASS and FAIL results explicitly and put unit suffixes on large
numbers.
.HP
.BI "\-v" "" ", \-\-verbose" ""
.IP
Increase output verbosity.
.HP
.BI "\-?" " " ", \-\-help" " "
.IP
Print the usage message.
.SH "AUTHORS"
The original zpios implementation was created by Cluster File Systems
Inc and adapted to ZFS on Linux by Brian Behlendorf
<behlendorf1@llnl.gov>.
This man page was written by Darik Horn <dajhorn@vanadac.com>.
.SH "SEE ALSO"
.BR zpool (8),
.BR zfs (8)