Date: Fri, 25 May 2001 15:25:33 -0700 (PDT) From: Matt Dillon <dillon@earth.backplane.com> To: Dave Hayes <dave@jetcafe.org>, hackers@FreeBSD.ORG Subject: Preliminary Tuning man page (was Re: Benchmarking FreeBSD (was ...)) Message-ID: <200105252225.f4PMPXI44229@earth.backplane.com> References: <200105250638.XAA06408@hokkshideh.jetcafe.org> <200105251951.f4PJp1b42293@earth.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Ok, here is my first shot at a 'tuning' manual page. If anyone wants
to review it, I am open to all suggestions, grammatical and formatting,
fixes etc... just email me with the changes (do not email the entire
document back to me, just a diff). It's not 100% complete (well duh!),
but it's a good start.
While this topic is broken open, if anyone has adjustments or updates
for my 'security' man page (which is already in the system), email
diffs for that to me as well. Now is the time.
Sometime this weekend I'll commit it.
I think I'll do a 'firewall' manual page as well. I'm tired of people
not configuring firewalls properly and then whining about it.
-Matt
.\" Copyright (c) 2001, Matthew Dillon. Terms and conditions are those of
.\" the BSD Copyright as specified in the file "/usr/src/COPYRIGHT" in
.\" the source tree.
.\"
.\" $FreeBSD$
.\"
.Dd May 25, 2001
.Dt TUNING 7
.Os FreeBSD
.Sh NAME
.Nm tuning
.Nd performance tuning under FreeBSD
.Sh SYSTEM SETUP - DISKLABEL, NEWFS, TUNEFS, SWAP
.Pp
When using
.Xr disklabel 8
to lay out your filesystems on a hard disk it is important to remember
that hard drives can transfer data much more quickly from outer tracks
then they can from inner tracks. To take advantage of this you should
try to pack your smaller filesystems and swap closer to the outer tracks,
follow with the larger filesystems, and end with the largest filesystems.
It is also important to size system standard filesystems such that you
will not be forced to resize them later as you scale the machine up.
I usually create, in order, a 128M root, 1G swap, 128M /var, 128M /var/tmp,
3G /usr, and use any remaining space for /home. You should typically size
your swap space to approximately 2x main memory. If you do not have a lot
of ram, though, you will generally want a lot more swap. It is not
recommended that you configure any less then 256M of swap on a system.
The kernel's VM paging algorithms are tuned to perform best when there is
at least 2x swap verses main memory. Configuring too little swap can lead
to inefficiencies in the VM page scanning code as well as create issues
later on if you add more memory to your machine.
.Pp
How you size your
.Em /var
partition depends heavily on what you intend to use the machine for. This
partition is primarily used to hold mailboxes and the print spool. If your
machine is intended to act as a mail or print server you should consider
creating a much larger partition - perhaps a gig or more.
Sizing
.Em /var/tmp
depends on the kind of temp file usage you think you will need. 128M is
the minimum we recommend.
.Pp
The
.Em /usr
partition holds the bulk of the files required to support the system and
a subdirectory within it called
.Em /usr/local
holds the bulk of the files installed from the
.Xr ports 7
hierarchy. If you do not use ports all that much and do not intend to keep
system source (/usr/src) on the machine, you can get away with
a 1 gigabyte /usr partition. However, if you install a lot of ports
(especially window managers and linux-emulated binaries), we recommend
at least a 2 gigabyte /usr and if you also intend to keep system source
on the machine, we recommend a 3 gigabyte /usr. Do not underestimate the
amount of space you will need in this partition, it can creep up and
surprise you!
.Pp
The
.Em /home
partition is typically used to hold user-specific data. I usually size it
to the remainder of the disk.
.Pp
Why partition at all? Why not create one big
.Em /dev/kmem
partition and be done with it? Then I don't have to worry about undersizing
things! Well, there are several reasons this isn't a good idea. First,
each partition has different operational characteristics and separating them
allows the filesystem to tune itself to those characteristics. For example,
the root and /usr partitions are read-mostly, with very little writing, while
a lot of reading and writing could occur in /var and /var/tmp. By properly
partitioning your system, fragmentation introduced in the smaller more
heavily write-loaded partitions will not bleed over into the mostly-read
partitions. Additionally, keeping the write-loaded partitions closer to
the edge of the disk (i.e. before the really big partitions instead of after
in the partition table) will increase I/O performance in the partitions
where you need it the most. Now it is true that you might also need I/O
performance in the larger partitions, but they are so large that shifting
them more towards the edge of the disk will not lead to a significnat
performance improvement whereas moving /var to the edge can have a huge impact.
.Pp
Properly partitioning your system also allows you to tune
.Xr newfs 8 ,
and
.Xr tunefs 8
parameters. Tuning
.Fn newfs
requires more experience but can lead to significant improvements in
performance. There are three parameters that are relatively safe to
tune:
.Em blocksize ,
.Em bytes/inode ,
and
.Em cylinders/group .
.Pp
.Fx
performs best when using 8K or 16K filesystem block sizes. The default
filesystem block size is 8K. For larger partitions it is usually a good
idea to use a 16K block size. This also requires you to specify a larger
fragment size. We recommend always using a fragment size that is 1/8
the block size (less testing has been done on other fragment size factors).
The
.Fn newfs
options for this would be
.Li Em newfs -f 2048 -b 16384 ...
Using a larger block size can cause fragmentation of the buffer cache and
lead to lower performance.
.Pp
If a large partition is intended to be used to hold fewer, larger files, such
as a database files, you can increase the
.Em bytes/inode
ratio which reduces the number if inodes (maximum number of files and
directories that can be created) for that partition. Decreasing the number
of inodes in a filesystem can greatly reduce
.Xr fsck 8
recovery times after a crash. Do not use this option
unless you are actually storing large files on the partition, because if you
overcompensate you can wind up with a filesystem that has lots of free
space remaining but cannot accomodate any more files. Using
32768, 65536, or 262144 bytes/inode is recommended. You can go higher but
it will have only incremental effects on fsck recovery times. For
example,
.Li Em newfs -i 32768 ...
.Pp
Finally, increasing the
.Em cyliners/group
ratio has the effect of packing the inodes closer together. This can increase
directory performance and also decrease fsck times. If you use this option
at all, we recommend maxing it out. Use
.Li Em newfs -c 999
and newfs will error out and tell you what the maximum is, then use that.
.Pp
.Xr tunefs 8
may be used to further tune a filesystem. This command can be run in
single-user mode without having to reformat the filesystem. However, this
is possibly the most abused program in the system. Many people attempt to
increase available filesystem space by setting the min-free percentage to 0.
This can lead to severe filesystem fragmentation and we do not recommend
that you do this. Really the only tunefs option worthwhile here is turning on
.Em softupdates
with
.Li Em tunefs -n enable /filesystem.
Softupdates drastically improves meta-data performance, mainly file
creation and deletion. We recommend turning softupdates on on all of your
filesystems. There are two downsides to softupdates that you should be
aware of: First, softupdates guarentees filesystem consistency in the
case of a crash but could very easily be several seconds (even a minute!)
behind updating the physical disk. If you crash you may loose more work
then otherwise. Secondly, softupdates delays the freeing of filesystem
blocks. If you have a filesystem (such as the root filesystem) which is
close to full, doing a major update of it, e.g.
.Em make installworld,
can run it out of space and cause the update to fail.
.Sh SYSCTL TUNING
.Pp
There are several hundred
.Xr sysctl 8
variables in the system, including many that appear to be candidates for
tuning but actually aren't. In this document we will only cover the ones
that have the greatest effect on the system.
.Pp
The
.Em kern.ipc.shm_use_phys
sysctl defaults to 0 (off) and may be set to 0 (off) or 1 (on). Setting
this parameter to 1 will cause all SysV shared memory segments to be
mapped to unpageable physical ram. This feature only has an effect if you
are either (A) mapping small amounts of shared memory across many (hundreds)
of processes, or (B) mapping large amounts of shared memory across any
number of processes. This feature allows the kernel to remove a great deal
of internal memory management page-tracking overhead at the cost of wiring
the shared memory into core, making it unswappable.
.Pp
The
.Em vfs.vmiodirenable
sysctl defaults to 0 (off) (though soon it will default to 1) and may be
set to 0 (off) or 1 (on). This parameter controls how directories are cached
by the system. Most directories are small and use but a single fragment
(typically 1K) in the filesystem and even less (typically 512 bytes) in
the buffer cache. However, when operating in the default mode the buffer
cache will only cache a fixed number of directories even if you have a huge
amount of memory. Turning on this sysctl allows the buffer cache to use
the VM Page Cache to cache the directories. The advantage is that all of
memory is now available for caching directories. The disadvantage is that
the minimum in-core memory used to cache a directory is the physical page
size (typically 4K) rather then 512 bytes. We recommend turning this option
on if you are running any services which manipulate large numbers of files.
Such services can include web caches, large mail systems, and news systems.
Turning on this option will generally not reduce performance even with the
wasted memory but you should experiment to find out.
.Pp
There are various buffer-cache and VM page cache related sysctls. We do
not recommend messing around with these at all. As of
.Fx 4.3 ,
the VM system does an extremely good job tuning itself.
.Pp
The
.Em net.inet.tcp.sendspace
and
.Em net.inet.tcp.recvspace
sysctls are of particular interest if you are running network intensive
applications. This controls the amount of send and receive buffer space
allowed for any given TCP connection. The default is 16K. You can often
improve bandwidth utilization by increasing the default at the cost of
eating up more kernel memory for each connection. We do not recommend
increasing the defaults if you are serving hundreds or thousands of
simultanious connections because it is possible to quickly run the system
out of memory due to stalled connections building up. But if you need
high bandwidth over a fewer number of connections, especially if you have
gigabit ethernet, increasing these defaults can make a huge difference.
You can adjust the buffer size for incoming and outgoing data separately.
For example, if your machine is primarily doing web serving you may want
to decrease the recvspace in order to be able to increase the sendspace
without eating too much kernel memory. Note that the route table, see
.Xr route 8 ,
can be used to introduce route-specific send and receive buffer size
defaults. As an additional mangagement tool you can use pipes in your
firewall rules, see
.Xr ipfw 8 ,
to limit the bandwidth going to or from particular IP blocks or ports.
For example, if you have a T1 you might want to limit your web traffic
to 70% of the T1's bandwidth in order to leave the remainder available
for mail and interactive use. Normally a heavily loaded web server
will not introduce significant latencies into other services even if
the network link is maxed out, but enforcing a limit can smooth things
out and lead to longer term stability. Many people also enforce artificial
bandwidth limitations in order to ensure that they are not charged for
using too much bandwidth.
.Pp
We recommend that you turn on (set to 1) and leave on the
.Em net.inet.tcp.always_keepalive
control. The default is usually off. This introduces a small amount of
additional network bandwidth but guarentees that dead tcp connections
will eventually be recognized and cleared. Dead tcp connections are a
particular problem on systems accesed by users operating over dialups,
because users often disconnect their modems without properly closing active
connections.
.Sh KERNEL CONFIG TUNING
.Pp
There are a number of kernel options that you may have to fiddle with in
a large scale system. In order to change these options you need to be
able to compile a new kernel from source. The
.Xr config 8
manual page and the handbook are good starting points for learning how to
do this. Generally the first thing you do when creating your own custom
kernel is to strip out all the drivers and services you don't use. Removing
things like
.Em INET6
and drivers you don't have will reduce the size of your kernel, sometimes
by a megabyte or more, leaving more memory available for applications.
.Pp
The
.Em maxusers
kernel option defaults to an incredibly low value. For most modern machines,
you probably want to increase this value to 64, 128, or 256. We do not
recommend going above 256 unless you need a huge number of file descriptors.
Network buffers are also effected but can be controlled with a separate
kernel option. Do not increase maxusers just to get more network mbufs.
.Pp
.Em NMBCLUSTERS
may be adjusted to increase the number of network mbufs the system is
willing to allocate. Each cluster represents approximately 16K of memory,
so a value of 1024 represents 16M of kernel memory reserved for network
buffers. You can do a simple calculation to figure out how many you need.
If you have a web server which maxes out at 1000 simultanious connections,
and each connection eats a 16K receive and 16K send buffer, you need
approximate 32MB worth of network buffers to deal with it. A good rule of
thumb is to multiply by 2, so 32MBx2 = 64MB/16K = 4096. So for this case
you would want to se NMBCLUSTERS to 4096. We recommend values between
1024 and 4096 for machines with moderates amount of memory, and 4096, 8192,
or 16384 for machines with greater amounts of memory. Under no circumstances
should you specify an arbitrarily high value for this parameter, it could
lead to a machine crash.
.Pp
More and more programs are using the
.Fn sendfile
system call to transmit files over the network. The
.Em NFSBUFS
kernel parameter controls the number of filesystem buffers
.Fn sendfile
is allowed to use to perform its work. This parameter nominally scales
with
.Em maxusers
so you should not need to mess with this parameter except under extreme
circumstances.
.Pp
.Em SCSI_DELAY
and
.Em IDE_DELAY
may be used to reduce system boot times. The defaults are fairly high and
can be responsible for 15+ seconds of delay in the boot process. Reducing
SCSI_DELAY to 5 seconds usually works (especially with modern drives).
Reducing IDE_DELAY also works but you have to be a little more careful.
.Pp
There are a number of
.Em XXX_CPU
options that can be commented out. If you only want the kernel to run
on a Pentium class cpu, you can easily remove
.Em I386_CPU
and
.Em I486_CPU,
but only remove
.Em I586_CPU
if you are sure your cpu is being recognized as a Pentium II or better.
Some clones may be recognized as a pentium or even a 486 and not be able
to boot without those options. If it works, great! The operating system
will be able to better-use higher-end cpu features for mmu, task switching,
timebase, and even device operations. Additionally, higher-end cpus support
4MB MMU pages which the kernel uses to map the kernel itself into memory,
which increases its efficiency under heavy syscall loads.
.Sh IDE WRITE CACHING
As of
.Fx 4.3 ,
IDE write caching is turned off by default. This will reduce write bandwidth
to IDE disks but is considered necessary due to serious data consistency
issues introduced by hard drive vendors. Basically the problem is that
IDE drives lie about when a write completes. With IDE write caching turned
on, IDE hard drives will not only write data to disk out of order, they
will sometimes delay some of the blocks indefinitely when under heavy disk
loads. A crash or power failure can result in serious filesystem
corruption. So our default is to be safe. If you are willing to risk
filesystem corruption, you can return to the old behavior by setting the
hw.ata.wc
kernel variable back to 1. This must be done from the boot loader at boot
time. Please see
.Xr ata 4 ,
and
.Xr loader 8 .
.Pp
There is a new experimental feature for IDE hard drives called hw.ata.tags
(you also set this in the bootloader) which allows write caching to be safely
turned on. This brings SCSI tagging features to IDE drives. As of this
writing only IBMDPTA and DTLA drives support the feature.
.Sh CPU, MEMORY, DISK, NETWORK
The type of tuning you do depends heavily on where your system begins to
bottleneck as load increases. If your system runs out of cpu (idle times
are pepetually 0%) then you need to consider upgrading the cpu or moving to
an SMP motherboard (multiple cpu's), or perhaps you need to revisit the
programs that are causing the load and try to optimize them. If your system
is paing to swap a lot you need to consider adding more memory. If your
system is saturating the disk you typically see high cpu idle times and
total disk saturation.
.Xr systat 1
can be used to monitor this. There are many solutions to saturated disks:
increasing memory for caching, mirroring disks, distributing operations across
several machines, and so forth. If disk performance is an issue and you
are using IDE drives, switching to SCSI can help a great deal. While modern
IDE drives compare with SCSI in raw sequential bandwidth, the moment you
start seeking around the disk SCSI drives usually win.
.Pp
Finally, you might run out of network suds. The first line of defense for
improving network performance is to make sure you are using switches instead
of hubs, especially these days where switches are almost as cheap. Hubs
are severely limited due to collision backoff and can cause other
inefficiencies to build in the system (such as causing more tcp packet
retries to occur due to nondeterministic delays). If you are hitting a
bottleneck in your WAN link (e.g. modem, T1, DSL, whatever) and do not
have the luxury of getting a bigger pipe, you may be able to use
.XR ipfw 8
to partition and traffic-shape the bandwidth going over the pipe. Obviously
you cannot push more bandwidth then the pipe will hold, but you can tune
it so an overload with one service does not effect all services running over
the pipe.
.Sh SEE ALSO
.Pp
.Xr ata 4 ,
.Xr boot 8 ,
.Xr config 8 ,
.Xr disklabel 8 ,
.Xr fsck 8 ,
.Xr ifconfig 8 ,
.Xr ipfw 8 ,
.Xr loader 8 ,
.Xr login.conf 5 ,
.Xr newfs 8 ,
.Xr ports 7 ,
.Xr route 8 ,
.Xr sysctl 8 ,
.Xr systat 1 ,
.Xr tunefs 8
.Sh HISTORY
The
.Nm
manual page was originally written by
.An Matthew Dillon
and first appeared
in
.Fx 4.3 ,
May 2001.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200105252225.f4PMPXI44229>