Maximizing Performance

This article is a retrospective analysis and basic rundown about gaining performance in Arch Linux.

Contents [hide] 1 The basics 1.1 Know your system 1.2 The first thing to do 1.3 Compromise 1.4 Benchmarking 2 Storage devices 2.1 Choosing and tuning your filesystem 2.1.1 Summary 2.1.2 Mount options 2.1.3 Ext3 2.1.4 Ext4 2.1.5 JFS 2.1.6 XFS 2.1.7 Reiserfs 2.1.8 BTRFS 2.2 Compressing /usr 2.3 Tuning for an SSD 3 CPU 3.1 Verynice 3.2 Ulatencyd 4 Network 5 Graphics 5.1 Xorg.conf configuration 5.2 Driconf 5.3 GPU Overclocking 6 RAM and swap 6.1 Swappiness 6.2 Compcache 6.3 Mounting /tmp to RAM 6.4 Using the graphic card's RAM 6.5 Preloading 6.5.1 Go-preload 6.5.2 Preload 6.5.3 Readahead 7 Boot time 7.1 Suspend to ram 7.2 Kernel boot options 7.3 Custom kernel 8 Application-specific tips 8.1 Firefox 8.2 Gcc/Makepkg 8.3 Mkinitcpio 8.4 LibreOffice 8.5 Pacman 8.6 SSH 9 Laptops

The basics

Know your system

The best way to tune a system is to target the bottlenecks, that is the subsystems that limit the overall speed. They usually can be identified by knowing the specifications of the system, but there are some basic indications:

• If the computer becomes slow when big applications, like OpenOffice.org and Firefox, are running at the same time, then there is a good chance the amount of RAM is insufficient. To verify available RAM, use this command, and check for the line beginning with -/+buffers:

$ free -m

• If boot time is really slow, and if applications take a lot of time to load the first time they are launched, but run fine afterwards, then the hard drive is probably too slow. The speed of a hard drive can be measured using the hdparm command:

$ hdparm -t /dev/harddrive

This is only the pure read speed of the hard drive, and is not a valid benchmark, but a value superior to 40MB/s (assuming drive tested while idle) can be considered decent on an average system.

• If the CPU load is consistently high even when RAM is available, then lowering CPU usage should be a priority. CPU load can be monitored in many ways, like using the top command:

$ top

• If the only applications lagging are the ones using direct rendering, meaning they use the graphic card, like video players and games, then improving the graphic performance should help. First step would be to verify if direct rendering simply is not enabled. This is indicated by the glxinfo command:

$ glxinfo | grep direct

The first thing to do

The simplest and most efficient way of improving overall performance is to run lightweight environments and applications.

• Use a window manager instead of a Desktop Environment. Choices include dwm, wmii, Awesome, Openbox, Fluxbox and JWM.
• Choose a minimal Desktop Environment over GNOME and KDE. Choices include LXDE and Xfce.
• Using lightweight applications. See Lightweight Software and the Light and Fast Applications Awards threads in the forum: 2007, 2008, 2009,2010, and 2011.
• Remove unnecessary daemons and background what daemons you can in /etc/rc.conf.

Compromise

Almost all tuning brings drawbacks. Lighter applications usually come with less features and some tweaks may make a system unstable, or simply require time to implement and maintain. This page tries to highlight those drawbacks, but the final judgment rests on the user.

Benchmarking

The effects of optimization are often difficult to judge. They can however be measured by benchmarking tools

Storage devices

Choosing and tuning your filesystem

Choosing the best filesystem for a specific system is very important because each has its own strengths. The beginner's guide provides a short summary of the most popular ones. You can also find relevant articles here.

Summary

• XFS: Excellent performance with large files. Low speed with small files. A good choice for /home.
• Reiserfs: Excellent performance with small files. A good choice for /var.
• Ext3: Average performance, reliable.
• Ext4: Great overall performance, reliable, has performance issues with sqlite and some other databases.
• JFS: Good overall performance, very low CPU usage, extremely fast resume after power failure.
• Btrfs: Great overall performance (better than ext4), reliable (once it becomes stable). Lots of features. Still in heavy development and considered as unstable. Do not use this filesystem yet unless you know what you are doing and are prepared for potential data loss.

Mount options

Mount options offer an easy way to improve speed without reformatting. They can be set using the mount command:

$ mount -o option1,option2 /dev/partition /mnt/partition

To set them permanently, you can modify /etc/fstab to make the relevant line look like this:

/dev/partition /mnt/partition partitiontype option1,option2 0 0

A couple of mount options improving performance on almost all file-systems is noatime,nodiratime. The former is a superset of the latter (which applies to directories only -- noatime applies to both files and directories). In rare cases, for example if you use mutt, it can cause minor problems. You can instead use the relatime option (NB relatime is the default in >2.6.30)

Ext3

See Ext3 Filesystem Tips.

Ext4

See the Ext4 wiki page.

JFS

See JFS Filesystem.

XFS

For optimal speed, create an XFS file-system with:

$ mkfs.xfs -l internal,lazy-count=1,size=128m -d agcount=2 /dev/thetargetpartition

An XFS specific mount option that may increase performance is logbufs=8.

#/etc/fstab
LABEL=XFSHOME /home xfs noatime,logbufs=8 0 1

Reiserfs

The data=writeback mount option improves speed, but may corrupt data during power loss. The notail mount option increases the space used by the filesystem by about 5%, but also improves overall speed. You can also reduce disk load by putting the journal and data on separate drives. This is be done when creating the filesystem:

$ mkreiserfs –j /dev/hda1 /dev/hdb1

Replace /dev/hda1 with the partition reserved for the journal, and /dev/hdb1 with the partition for data. You can learn more about reiserfs with this article.

BTRFS

See compression.

Compressing /usr

A way to speed up reading from the hard drive is to compress the data, because there is less data to be read. It must however be decompressed, which means a greater CPU load. Some filesystems support transparent compression, most notably btrfs and reiserfs4, but their compression ratio is limited by the 4k block size. A good alternative is to compress /usr in a squashfs file, with a 64k(128k) block size, as instructed in this Gentoo forums thread. What this tutorial does is basically to compress the /usr folder into a compressed squashfs file-system, then mounts it with aufs. A lot of space is saved, usually two thirds of the original size of /usr, and applications load faster. However, each time an application is installed or reinstalled, it is written uncompressed, so /usr must be re-compressed periodically. Squashfs is already in the kernel, and aufs2 is in the extra repository, so no kernel compilation is needed if using the stock kernel. Since the linked guide is for Gentoo the next commands outline the steps especially for Arch. Basically we have got install two packages to get it working:

# pacman -S aufs2 squashfs-tools

This command installs the aufs-modules and some userspace-tools for the squash-filesystem. Now we need some extra directories where we can store the archive of /usr as read-only and another folder where we can store the data changed after the last compression as writeable:

# mkdir -p /squashed/usr/{ro,rw}

Now that we got a rough setup you should perform a complete system-upgrade since every change of content in /usr after the compression will be excluded from this speedup. If you use prelink you should also perform a complete prelink before creating the archive. Now it is time to invoke the command to compress /usr:

# mksquashfs /usr /squashed/usr/usr.sfs -b 65536

These parameters/options are the ones suggested by the Gentoo link but there might be some room for improvement using some of the options described here. Now to get the archive mounted together with the writeable folder it is necessary to edit /etc/fstab:

# nano /etc/fstab

Add the following lines:

/squashed/usr/usr.sfs   /squashed/usr/ro   squashfs   loop,ro   0 0 
usr    /usr    aufs    udba=reval,br:/squashed/usr/rw:/squashed/usr/ro  0 0

Now you should be done and able to reboot. The original Author suggests to delete all the old content of /usr, but this might cause some problems if anything goes wrong during some later re-compression. It is more safe to leave the old files in place just to be on the safe side.
A bash script has been created that will automate the process of re-compressing (read updating) the archive since the tutorial is meant for Gentoo and some options do not correlate to what they should be in Arch.

Tuning for an SSD

SSD#Tips_for_Maximizing_SSD_Performance

CPU

The only way to directly improve CPU speed is overclocking. As it is a complicated and risky task, it is not recommended for anyone except experts. The best way to overclock is through the BIOS. When purchasing your system, keep in mind that most Intel motherboards are notorious for disabling the capacity to overclock.
A way to modify performance (ref) is to use Con Kolivas' desktop-centric kernel patchset, which, among other things, replaces the Completely Fair Scheduler (CFS) with the Brain Fuck Scheduler (BFS).
Kernel PKGBUILDs that include the BFS patch can be installed from the AUR or Unofficial_User_Repositories. See the respective pages for linux-ck and linux-ck wiki page, linus-bfs or linux-pf for more information on their additional patches.

Note: BFS/CK are designed for desktop/laptop use and not servers. They provide low latency and work well for 16 CPUs or less. Also, Con Kolivas suggests setting HZ to 1000. For more information, see the BFS FAQ and ck patches.

Verynice

Verynice is a daemon, available on AUR, for dynamically adjusting the nice levels of executables. The nice level represent the priority of the executable when allocating CPU resources. Simply define executables for which responsiveness is important, like X or multimedia applications, as goodexe in /etc/verynice.conf. Similarly, CPU-hungry executables running in the background, like make, can be defined as badexe. This prioritization greatly improves system responsiveness under heavy load.

Ulatencyd

Ulatencyd is a daemon that controls how the Linux kernel will spend its resources on the running processes. It uses dynamic cgroups to give the kernel hints and limitations on processes.

Network

See relevant section in General Recommendations.

Graphics

Xorg.conf configuration

Graphic performance heavily depends on the settings in /etc/X11/xorg.conf. There are tutorials for Nvidia, ATI and Intel cards. Improper settings may stop Xorg from working, so caution is advised.

Driconf

Driconf is a small utility that allows you to change the direct rendering settings for open source drivers. Enabling HyperZ can drastically improve performance.

GPU Overclocking

Overclocking a graphics card is typically more expedient than with a CPU, since there are readily accessible software packages which allow for on-the-fly GPU clock adjustments. For ATI users, get rovclock, and Nvidia users should get nvclock in the extra repository. Intel chipsets users can install GMABooster from AUR
The changes can be made permanent by running the appropriate command after X boots, for example by adding it to ~/.xinitrc. A safer approach would be to only apply the overclocked settings when needed.

RAM and swap

Swappiness

The swappiness represent how much the kernel prefers swap to RAM. Setting it to a very low value, meaning the kernel will almost always use RAM, is known to improve responsiveness on many systems. To do that, simply add those line to /etc/sysctl.conf:

vm.swappiness=20
vm.vfs_cache_pressure=50

To test and more on why this may work, take a look at this article.

Compcache

Compcache, also known as the zram kernel module, creates a swap device in RAM and compresses it. That means that part of the RAM can hold much more information but uses more CPU. Still, it is much quicker than swapping to a hard drive. If a system often falls back to swap, this could improve responsiveness. Zram is in mainline staging (therefore its not stable yet, use with caution).

 modprobe zram

It is also possible (and recommended) to tell compcache to fall back on the hard drive swap when full. To do this, define a backing swap device in the configuration file. This swap device must not be in use when zram is started, so remove it from your /etc/fstab!
This is also a good way to reduce disk read/write cycles due to swap on SSDs.

Mounting /tmp to RAM

This will make your system a tiny bit faster, but will take up some of your RAM. It also reduces disk read/write cycles, and is therefore a good choice if using an SSD or if you have RAM to spare. Simply add this line to /etc/fstab and reboot:

tmpfs /tmp tmpfs defaults,noatime,nodev,nosuid,mode=1777 0 0

Using the graphic card's RAM

In the unlikely case that you have very little RAM and a surplus of video RAM, you can use the latter as swap. See Swap on video ram.

Preloading

Preloading is the action of putting and keeping target files into the RAM. The practical use is that preloaded applications always start very quickly, because reading from the RAM is always quicker than from the hard drive. However, part of your RAM will be dedicated to this task, but no more than if you kept the application open. Therefore, preloading is best used with heavy, often-used applications, like firefox and openoffice.

Go-preload

Go-preload is a small daemon created in the gentoo forum. To use it, first run this command in a terminal for each program you want to preload at boot:

# gopreload-prepare program

Then, as instructed, press enter when the program is fully loaded. This will add a list of files needed by the program in /usr/share/gopreload/enabled. To load all lists at boot, simply add gopreload to your DAEMONS array in /etc/rc.conf. To disable the loading of a program, remove the appropriate list in /usr/share/gopreload/enabled, or move it to /usr/share/gopreload/disabled.

Preload

A more automated, albeit less KISS, approach is used by Preload. All you have to do is add it to your DAEMONS array in /etc/rc.conf. It will monitor the most used files on your system, and with time build its own list of files to preload at boot.

Readahead

Readahead is a tool that can cache files before needed and help you accelerating program loading.

Boot time

You can find tutorials with good tips in the article Improve Boot Performance.

Suspend to ram

The best way to reduce boot time is not booting at all. Consider suspending your system to ram instead.

Kernel boot options

Some boot options can decrease kernel boot time. The fastboot option usually can take off one second or so. Also, if you see a message saying "Waiting 8s for device XXX" at boot, adding rootdelay=1 can reduce the waiting time, but be careful, as it may break the booting process. Those options are set in /boot/grub/menu.lst or /etc/lilo.conf, depending on which bootloader you use.

Custom kernel

Compiling a custom kernel will reduce boot time and memory usage, but can be long, complicated and even painful. It usually is not worth the effort, but can be very interesting and a great learning experience. If you really know what you are doing, start here.

Application-specific tips

Firefox

The Firefox article offers good tips; most notably disabling pango, cleaning the sqlite database, and using firefox-pgo. See also: Speed-up Firefox using tmpfs, and Turning off anti-phishing.

Gcc/Makepkg

See Ccache.

Mkinitcpio

User josh_ from the forum has made impressive changes to the mkinitcpio script, making it two or three times faster. While waiting for these changes to be implemented, you can get them here.

LibreOffice

See Speed up LibreOffice.

Pacman

See Improve Pacman Performance.

SSH

See Speed up SSH.

https://wiki.archlinux.org/index.php/Maximizing_Performance