Embedded Linux Optimizations

Embedded Linux optimizations
Embedded Linux
optimizations
Size, RAM, speed,
power, cost
Michael Opdenacker
Thomas Petazzoni
Gilles Chanteperdrix
Free Electrons
© Copyright 20042009, Free Electrons.
Creative Commons BYSA 3.0 license
Latest update: May 26, 2010,
Document sources, updates and translations:
http://freeelectrons.com/docs/optimizations
Corrections, suggestions, contributions and translations are welcome!
1
Free Electrons. Kernel, drivers and embedded Linux development, consulting, training and support. http//freeelectrons.com
Penguin weight watchers
Make your penguin slimmer, faster, and reduce its consumption of fish!
Before 2 weeks after
2
CE Linux Forum
http://celinuxforum.org/
Non profit organization, whose members are embedded Linux
companies and Consumer Electronics (CE) devices makers.
Mission: develop the use of Linux in CE devices
Hosts many projects to improve the suitability of Linux for CE
devices and embedded systems. All patches are meant to be
included in the mainline Linux kernel.
Most of the ideas introduced in this presentation have been
gathered or even implemented by CE Linux Forum projects!
3
Contents
Ideas for optimizing the Linux kernel and executables
Increasing speed
Reducing size: disk footprint and RAM
Reducing power consumption
Global perspective: cost and combined optimization effects
The ultimate optimization tool!
4
Embedded Linux Optimizations
Increasing speed
Reducing kernel boot time
5
Measuring kernel boot time
CONFIG_PRINTK_TIME
Configure it in the Kernel Hacking section.
Adds timing information to kernel messages. Simple and robust.
Not accurate enough on some platforms (1 jiffy = 10 ms on arm!)
See http://elinux.org/Printk_Times
...
[42949372.970000] Memory: 64MB = 64MB total
[42949372.970000] Memory: 54784KB available (1404K code, 296K data, 72K init)
[42949373.180000] Mountcache hash table entries: 512
[42949373.180000] CPU: Testing write buffer coherency: ok
[42949373.180000] checking if image is initramfs...it isn't (bad gzip magic numb
ers); looks like an initrd
[42949373.200000] Freeing initrd memory: 8192K
[42949373.210000] NET: Registered protocol family 16
...
6
Boot tracer
CONFIG_BOOT_TRACER in kernel configuration
Introduced in Linux 2.6.28
Based on the ftrace tracing infrastructure
Allows to record the timings of initcalls
Boot with the initcall_debug and printk.time=1 parameters,
and run dmesg | perl scripts/bootgraph.pl > output.svg
to generate a graphical representation
7
Disable console output
The output of kernel bootup messages to the console takes
time! Even worse: scrolling up in framebuffer consoles!
Console output not needed in production systems.
Console output can be disabled with the quiet
argument in the Linux kernel command line (bootloader settings)
Example:
root=/dev/ram0 rw init=/startup.sh quiet
Benchmarks: can reduce boot time by 30 or even 50%!
See http://elinux.org/Disable_Console
8
Preset loops_per_jiffy
At each boot, the Linux kernel calibrates a delay loop (for the
udelay function). This measures a loops_per_jiffy (lpj)
value. This takes about 25 jiffies (1 jiffy = time between 2 timer
interrupts).
In embedded systems, it can be about 250 ms!
You just need to measure this once! Find the lpj value in kernel
boot messages (if you don't get it in the console, boot Linux with
the loglevel=8 parameter). Example:
Calibrating using timer specific routine... 187.59
BogoMIPS (lpj=937984)
At the next boots, start Linux with the below option:
lpj=<value>
9
Faster rebooting (1)
kexec system call: executes a new kernel from a running one.
Must faster rebooting: doesn't go through bootstrap / bootloader
code.
Great solution for rebooting after system (“firmware”) upgrades.
Useful for automatic rebooting after kernel panics.
See http://developer.osdl.org/andyp/kexec/whitepaper/kexec.pdf
and Documentation/kdump/kdump.txt in kernel sources.
10
Faster rebooting (2)
Another option: use reboot=soft in the kernel command line
When you reboot, the firmware will be skipped.
Drawback: unlike kexec, cannot be chosen from userspace.
Supported platforms: i386, x86_64, arm, arm26 (Aug. 2006)
See Documentation/kernelparameters.txt
in the kernel sources for details. Not supported on all platforms.
11
Skip memory allocation
Idea: spare memory at boot time and manage it by yourself!
Assume you have 32 MB of RAM
Boot your kernel with mem=30
The kernel will just manage the first 30 MB of RAM.
Driver code can now reclaim the 2 MB left:
buf = ioremap (
0x1e00000, /* Start: 30 MB */
0x200000 /* Size: 2 MB */
);
This saves time allocating memory.
Critical drivers are also sure to always have the RAM they need.
12
Other ideas
Copy kernel from flash to RAM using DMA.
Reducing probe time on some IDE operations.
Compile drivers as modules for devices not used at boot
time. This reduces time spent initializing drivers.
A smaller kernel is also faster to copy to RAM.
Multithreaded device probing.
See http://elinux.org/Boot_Time for more resources
See also Tim Bird's presentation on boot time reduction:
http://freeelectrons.com/redir/tbirdelce2008.html
13
Increasing speed
System startup time and application speed
14
Starting system services
SysV init:
Starts services sequentially. Waits for the current startup script to
be complete to start the next one! While dependencies exist,
some tasks can be run in parallel!
Initng: http://initng.org
New alternative to SysV init, which can start services in parallel,
as soon as their preconditions are met.
15
Initng vs. SysV init: bootcharts
Initng wins!
System utilization is much
better.
Note: you can hunt system startup trouble by using the Bootchart program
(http://www.bootchart.org/). Used to investigate Fedora booting time.
16
Reading ahead
Linux keeps the contents of all the files it reads in RAM (in the
page cache), as long as it doesn't need the RAM pages for
something else.
Idea: load files (programs and libraries in particular) in RAM
cache before using them. Best done when the system is not
doing any I/O.
Thanks to this, programs are not stuck waiting for I/O.
Used the Knoppix distribution to achieve very nice boot speed
ups.
Also planned to be used by Initng.
Not very useful for systems with very little RAM:
cached pages are recycled before the files are accessed.
17
Implementing readahead
You can use the sys_readahead() system call
in your C programs. See man readahead for details.
You can also use the readaheadlist utility, which reads a file
containing the list of files to load in cache.
Available on: http://freshmeat.net/projects/readaheadlist/.
In embedded systems using Busybox, you can use the
readahead command (implemented by Free Electrons).
18
Compiler speed optimizations
By default, most tools are compiled with compiler optimizations.
Make sure you use them for your own programs!
O2 is the most common optimization switch of gcc.
Lots of optimization techniques are available.
See http://en.wikipedia.org/wiki/Compiler_optimization
O3 can be also be used for speed critical executables.
However, there is done at the expense of code size (for example
“inlining”: replacing function calls by the function code itself).
19
Using processor acceleration instructions
liboil http://liboil.freedesktop.org/
Library of functions optimized for special instructions
from several processors (Altivec, MMX, SSE, etc.)
Mainly functions implementing loops on data arrays:
type conversion, copying, simple arithmetics, direct cosine
transform, random number generation...
Transparent: keeps your application portable!
So far mainly supports desktop processors
License: BSD type
20
Prelinking (1)
Applies to executables using shared libraries
To load and start an executable, the dynamic linker has a
significant amount of work to do (mainly address relocation)
It can take a lot of time for executables using many shared
libraries!
In many systems in which executables and shared libraries never
change, the same job is done every time the executable is
started.
21
Prelinking (2)
prelink
http://people.redhat.com/jakub/prelink/
prelink modifies executables and shared libraries to simplify
the dynamic linker relocation work.
This can greatly reduce startup time for big applications (50%
less for KDE!). This also saves memory consumed by
relocations.
Can be used to reduce the startup time of a Linux system.
Just needs to be run again when libraries or executables are
updated.
Details on http://elinux.org/Pre_Linking
22
Use simpler Unix executables
Big, feature rich executables take time to load.
Particularly true for shell scripts calling the bash shell!
Idea: replace standard Unix / GNU executables by lightweight, simplified
implementations by busybox (http://busybox.net).
Implemented by Ubuntu 6.10 to reduce boot time, replacing bash (649 K)
by dash (79 K, see http://en.wikipedia.org/wiki/Debian_Almquist_shell).
This broke various shell scripts which used bash specific features
(“bashisms”).
In nonembedded Linux systems
where featurerich executables are still needed,
should at least use busybox ash for system scripts.
23
Shells: reducing forking
fork / exec system calls are very heavy.
Because of this, calls to executables from shells are slow.
Even executing echo in busybox shells results in a fork syscall!
Select Shells > Standalone shell in busybox
configuration to make the busybox shell call applets whenever
possible.
Pipes and backquotes are also implemented by fork / exec.
You can reduce their usage in scripts. Example:
cat /proc/cpuinfo | grep model
Replace it with: grep model /proc/cpuinfo
See http://elinux.org/Optimize_RC_Scripts
24
Use faster filesystems
Run faster by using the most appropriate filesystems!
Compressed readonly filesystem (block device):
use SquashFS (http://squashfs.sourceforge.net)
instead of CramFS (much slower, getting obsolete).
NAND flash storage: you should try UBIFS
(http://www.linuxmtd.infradead.org/doc/ubifs.html), the
successor of JFFS2. It is much faster. You could also use
SquashFS. See our Choosing filesystems presentation
(http://freeelectrons.com/docs/filesystems).
25
Use faster filesystems (2)
Use RAM filesystems for temporary, speed critical files
with no need for permanent storage. Details in the kernel
sources: Documentation/filesystems/tmpfs.txt
Benchmark your system and application on competing
filesystems! Reiser4 is more innovative and benchmarks
found it faster than ext3.
Good to benchmark your system with JFS or XFS too.
XFS is reported to be the fastest to mount (good for
startup time), and JFS to have the lowest CPU utilization.
See http://www.debianadministration.org/articles/388
ext4 is also ready to be used now.
26
Speed up applications with tmpfs
When enough RAM is available, the OS keeps recently accessed files and
applications in RAM (page cache). This significantly speeds up any new usage.
However, depending on system activity, this may not last long.
For programs that need fast startup even if they haven't been run for a long time:
copy them to a tmpfs filesystem at system startup! This makes sure they are
always accessed from the file cache in RAM (provided you do not have a swap
partition).
See Documentation/filesystems/tmpfs.txt in kernel sources for details about
tmpfs.
Caution: don't use ramdisks instead!
Ramdisks duplicate files in RAM and unused space cannot be reclaimed.
Caution: use with care. May impact overall performance.
Not needed if there's enough RAM to cache all files and programs.
27
Boot from a hibernate image
The ultimate technique for instant boot!
In development: start the system, required applications and the
user interface. Hibernate the system to disk / flash in this state.
In production: boot the kernel and restore the system state from
with this predefined hibernation image.
This way, you don't have to initialize the programs one by one.
You just get the back to a valid state.
Used in Sony cameras to achieve instant power on time.
Unlike Suspend to RAM, still allows to remove batteries!
28
Use a profiler
Using a profiler can help to identify unexpected behavior degrading
application performance.
For example, a profiler can tell you in which functions most of the
time is spent.
Possible to start with strace and ltrace
Advanced profiling with Valgrind: http://valgrind.org/
Compile your application for x86 architecture
You can then profile it with the whole Valgrind toolsuite:
Cachegrind: sources of cache misses and function statistics.
Massif: sources of memory allocation.
See our Software Development presentation for details:
http://freeelectrons.com/docs/swdev/
29
Reducing size
Kernel size and RAM usage
30
LinuxTiny
Goal: reduce the disk footprint and RAM size of the Linux kernel
http://elinux.org/Linux_Tiny
Set of patches against the mainstream Linux kernel.
Mergeability in mainstream is a priority.
Many changes have already been merged in recent kernels.
All features can be selected in kernel configuration
(CONFIG_EMBEDDED).
Also ships utilities or patches for tracking sources
of memory usage or code size.
31
LinuxTiny ideas (1)
Remove kernel messages (printk, BUG, panic...)
Hunt excess inlining (speed vs. size tradeoff)
2.6.26: can allow gcc to uninline functions marked as inline:
(CONFIG_OPTIMIZE_INLINING=y). Only used by x86 so far.
Hunt excess memory allocations
Memory (slob instead of slab) allocator more space efficient for
small systems.
Reduce the size of kernel data structures (may impact
performance)
Simpler alternative implementations of kernel functionalities
with less features, or not supporting special cases.
32
LinuxTiny ideas (2)
Remove some features which may not be needed
in some systems.
Compiling optimizations for size.
A smaller kernel executable also saves RAM
(unless executed in place from storage).
33
LinuxTiny: kernel configuration screenshot
Many features
configured out
34
With and without CONFIG_EMBEDDED
Tests on Linux 2.6.29, on a minimalistic but working x86 kernel
1800
1600
1400
1200
1000 Raw
Compressed
800
600
400
200
0
CONFIG_EMBEDDED=n CONFIG_EMBEDDED=y
Raw: 272 KB (17%), Compressed: 136 KB (20%)
35
Replace initrd by initramfs
Replace init ramdisks (initrd) with initramfs:
much less overhead and ram waste!
Access
to file Regular Ramdisk ramfs
Access Access
block device to file block device to file
Virtual File
System
Virtual File Virtual File
System File System File
Filesystem File
driver cache cache cache
Filesystem
driver Copy
Block
driver
Block
Copy
driver Block storage
Block storage
RAM RAM RAM
36
ramfs advantages over ramdisks
No block and filesystem overhead.
No duplication in RAM.
Files can be removed (reclaiming RAM) after use.
Initramfs: ramfs archive embedded in the Linux kernel file.
37
Reducing size
Application size and RAM usage
38
Static or dynamic linking? (1)
Static linking
All shared library code duplicated in the executables
Allows not to copy the C library in the filesystem.
Simpler and smaller when very few executables (busybox)
Library code duplication: bad for systems with more executables
(code size and RAM)
Best for small systems (< 12 MB) with few executables!
39
Static or dynamic linking? (2)
Dynamic linking
Shared library code not duplicated in the executables
Makes much smaller executables
Saves space in RAM (bigger executables take more RAM)
Requires the library to the copied to the filesystem
Best for medium to big systems (> 500 KB 1 MB)
40
Using a lighter C library
glibc (GNU C library): http://www.gnu.org/software/libc/
Found on most computer type GNU/Linux machines
Size on arm: approx 1.7 MB
uClibc: http://www.uclibc.org/
Found in more and more embedded Linux systems!
Size on arm: approx 400 KB (you save 1.2 MB!)
Executables are slightly smaller too:
C program Compiled with shared libraries Compiled statically

glibc uClibc glibc uClibc
Plain “hello world” 4.6 K 4.4 K 475 K 25 K
Busybox 245 K 231 K 843 K 311 K
41
How to use uClibc?
Need to compile all your executables with a uClibc toolchain.
Readytouse toolchains can be found on
http://freeelectrons.com/community/tools/uclibc
You can very easily build your own with buildroot:
http://buildroot.uclibc.org/
You also have to copy the uClibc files from the toolchain to the
/lib directory in the target root filesystem.
Readytouse filesystems can also be generated by buildroot.
You just need to add your specific stuff then.
42
Need for stripping
Compiled executables and libraries contain extra information
which can be used to investigate problems in a debugger.
This was useful for the tool developer, but not for the final user.
To remove debugging information, use the strip command.
This can save a very significant amount of space!
gcc o hello hello.c (output size: 4635 bytes)
strip hello (output size: 2852 bytes, 38.5%)
Don't forget to strip libraries too!
43
Are my executables stripped?
You can use the file command to get the answer
gcc o hello hello.c
file hello
hello: ELF 32bit LSB executable, Intel 80386, version 1
(SYSV), for GNU/Linux 2.2.5, dynamically linked (uses
shared libs), not stripped
strip hello
hello: ELF 32bit LSB executable, Intel 80386, version 1
(SYSV), for GNU/Linux 2.2.5, dynamically linked (uses
shared libs), stripped
You can use findstrip (http://packages.debian.org/stable/source/perforate)
to find all executables and libraries that need stripping in your system.
44
How to strip
Some lightweight tools, like busybox, are automatically stripped
when you build them.
Makefiles for many standard tools offer a special command:
make installstrip
Caution: stripping is architecture dependent.
Use the strip command from your crosscompiling toolchain:
armlinuxstrip potato
45
sstrip: “super strip”
http://muppetlabs.com/~breadbox/software/elfkickers.html
Goes beyond strip and can strip out a few more bits that are
not used by Linux to start an executable.
Can be used on libraries too. Minor limitation: processed
libraries can no longer be used to compile new executables.
Can also be found in toolchains made by Buildroot (optional)
Regular
Hello World
4691 B
Busybox
287783 B
Inkscape
11397 KB
Best for tiny
stripped 2904 B (-38 %) 230408 B (-19.9 %) 9467 KB (-16.9 %) executables!
sstripped 1392 B (-70 %) 229701 B (-20.2 %) 9436 KB (-17.2 %)
46
Library Optimizer
http://libraryopt.sourceforge.net/
Contributed by MontaVista
Examines the complete target file system, resolves all shared
library symbol references, and rebuilds the shared libraries with
only the object files required to satisfy the symbol references.
Can also take care of stripping executables and libraries.
However, requires to rebuild all the components from source.
Would be nicer to achieve this only with ELF manipulations.
47
Compiler space optimizations
Regular compiler optimizations simplifying code also reduce size
You can also reduce the size of executables by asking gcc to
optimize generated code size:
gcc Os o husband husband.c
Os corresponds to O2 optimizations except the ones increasing
size, plus extra sizespecific ones.
Os is already used by default to build busybox.
Possible to further reduce the size by compiling and optimizing all
sources at once, with the fwholeprogram combine gcc
options.
See http://gcc.gnu.org/onlinedocs/gcc/OptimizeOptions.html
for all gcc optimization options.
48
Simple gcc optimization benchmark
Executable size
250000
200000
None
150000 O2 (generic)
O3 (speed)
Os (size)
100000
50000
0
Busybox Dropbear
49
Restartable applications
When RAM is scarce, can be useful to abort applications that are
not in use (for example hidden graphical interfaces).
Better to do it before the Linux Kernel OOM
(Out Of Memory) killer comes and makes bad decisions.
You can use the “Linux Checkpoint / Restart” project to have the
Linux kernel save the state of a running application so that it can
later resume its execution from the time at which it was
checkpointed.
See http://www.linuxcr.org/ for details.
50
Compressing filesystems
Can significantly increase your storage capacity
MTD (flash or ROM) storage: use UBIFS
or JFFS2 for small partitions.
Block storage: use SquashFS (http://squashfs.sourceforge.net)
instead of CramFS for readonly partitions. It compresses much
better and is much faster too.
51
Merging duplicate files
Software compiling and installing often create duplicate files...
Check that your root filesystem doesn’t contain any!
dupmerge2: http://sourceforge.net/projects/dupmerge
Replaces duplicate files by hard links.
clink: http://freeelectrons.com/community/tools/utils/clink
Replaces duplicate files by symbolic links.
Example: saves 4% of total space in Fedora Core 5.
finddup: http://www.shelldorado.com/scripts/cmds/finddup
Finds duplicate files.
52
Reducing power consumption
53
Tickless kernel
Kernel configuration: NO_HZ setting in Processor type and features
To implement multitasking, the processor receives a timer interrupt
at a given frequency (every 4 ms by default on Linux 2.6). On idle systems,
this wakes up the processor all the time, just to realize there is nothing to do!
Idea: when all processors are idle, disable the timer interrupt, and reenable it
when something happens (a real interrupt). This saves power in laptops, in
embedded systems and with virtual servers!
2.6.24: supports x86, arm, mips and powerpc
54
PowerTOP
http://www.lesswatts.org/projects/powertop/
A very nice utility showing the top 10 sources of power consumption.
Requirements: a tickless kernel (> 2.6.21), and a “mobile” (x86) CPU.
Best run on laptops running on battery (to estimate battery savings).
PowerTOP measures the number of wakeups,
and counts the time spent in lowpower modes.
It detects issues in both kernel space (drivers) and user space.
It already uncovered several bugs in today's distributions.
Embedded systems: non Intel processors probably not supported yet.
You can still run your apps on a PC to check their behavior.
See also http://www.lesswatts.org for other useful resources.
55
PowerTOP in action
56
cpufreq
Configuration: CPU_FREQ in Power management options
Allows to change the CPU frequency on the fly
Supported architectures (2.6.20):
i386, sh, ia64, sparc64, x86_64, powerpc, arm (i.MX only).
Usually controlled from userspace through /sys by a user
configurable governor process, according to CPU load, heat, battery
status... The most common is cpuspeed:
http://carlthompson.net/software/cpuspeed/
Saves a significant amount of battery life in notebooks.
57
Suspend hidden GUIs
Idea: suspend hidden user interfaces to save CPU and power.
Send a suspend (stop) signal:
kill SIGTSTP <pid>
Send a continue signal:
kill SIGCONT <pid>
58
Software suspend
http://www.suspend2.net/
Lots of great features for notebook users, such as RAM suspend
or hibernate to disk.
Unfortunately, restricted on some Intel compatible processors
and targeting only machines with APM or ACPI (rarely found in
non PC embedded systems!).
Not addressing the requirements of embedded systems (support
for other CPUs, voltage reduction...).
59
Global perspective
Cost and combined optimization effects
60
Combined benefits
Speed increase RAM reduction Power reduction Cost reduction

More speed CPU can run slower or stay longer Slower, cheaper
in power saving mode CPU
Less RAM Faster allocations Fewer / smaller RAM chips: less Fewer / cheaper

Less swapping dynamic and standby power. RAM chips
Sometimes less CPU with less cache: less power CPU with less
cache flushing cache: cheaper
Less space Faster application Less RAM usage Fewer / smaller storage chips: less Fewer / cheaper

loading from storage power storage
and in RAM.
Sometimes, simpler,
faster code.
Less power Cheaper batteries
or cheaper AC/DC
converter
61
The ultimate optimization tool!
We have seen many ways to optimize
an existing system.
However, nothing replaces a good design!
So, first carefully design and implement
your system and applications with their requirements
in mind.
Then, use the optimization techniques to further improve
your system and the parts that you reused (kernel and
applications).
62
Related documents
All our technical presentations
on http://freeelectrons.com/docs
Linux kernel
Device drivers
Architecture specifics
Embedded Linux system development
How to help
You can help us to improve and maintain this document...
By sending corrections, suggestions, contributions and
translations
By asking your organization to order development, consulting
and training services performed by the authors of these
documents (see http://freeelectrons.com/).
By sharing this document with your friends, colleagues
and with the local Free Software community.
By adding links on your website to our online materials,
to increase their visibility in search engine results.
Linux kernel
Linux device drivers
Free Electrons
Board support code Our services
Mainstreaming kernel code
Kernel debugging
Custom Development
System integration
Embedded Linux Training
Embedded Linux demos and prototypes
All materials released with a free license! System optimization
Unix and GNU/Linux basics Application and interface development
Linux kernel and drivers development
Realtime Linux, uClinux Consulting and technical support
Development and profiling tools Help in decision making
Lightweight tools for embedded systems System architecture
Root filesystem creation System design and performance review
Audio and multimedia Development tool and application support
System optimization Investigating issues and fixing tool bugs

Embedded Linux Optimizations

Uploaded by

Copyright:

Available Formats

Embedded Linux Optimizations

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Embedded Linux Optimizations

Uploaded by

Copyright:

Available Formats

Embedded Linux optimizations

C program Compiled with shared libraries Compiled statically

Speed increase RAM reduction Power reduction Cost reduction

Less RAM Faster allocations Fewer / smaller RAM chips: less Fewer / cheaper

Less space Faster application Less RAM usage Fewer / smaller storage chips: less Fewer / cheaper

You might also like

Embedded Linux Optimizations

Uploaded by

Copyright:

Available Formats

Embedded Linux Optimizations

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Embedded Linux Optimizations

Uploaded by

Copyright:

Available Formats

Embedded Linux optimizations

C program Compiled with shared libraries Compiled statically

Speed increase RAM reduction Power reduction Cost reduction

Less RAM ­ Faster allocations ­ Fewer / smaller RAM chips: less ­ Fewer / cheaper

Less space ­ Faster application ­ Less RAM usage ­ Fewer / smaller storage chips: less ­ Fewer / cheaper

You might also like

Less RAM Faster allocations Fewer / smaller RAM chips: less Fewer / cheaper

Less space Faster application Less RAM usage Fewer / smaller storage chips: less Fewer / cheaper