Difference between revisions of "Frandom"

From ArchWiki
Jump to: navigation, search
(Example)
(Example)
Line 75: Line 75:
  
 
Summary: 320GB in approx. 2.67 hours
 
Summary: 320GB in approx. 2.67 hours
 +
 +
[http://stackoverflow.com/questions/556405/what-do-real-user-and-sys-mean-in-the-output-of-time1?answertab=active#tab-top What does 'real', 'user' and 'sys' mean?]

Revision as of 20:19, 25 February 2012

This template has only maintenance purposes. For linking to local translations please use interlanguage links, see Help:i18n#Interlanguage links.


Local languages: Català – Dansk – English – Español – Esperanto – Hrvatski – Indonesia – Italiano – Lietuviškai – Magyar – Nederlands – Norsk Bokmål – Polski – Português – Slovenský – Česky – Ελληνικά – Български – Русский – Српски – Українська – עברית – العربية – ไทย – 日本語 – 正體中文 – 简体中文 – 한국어


External languages (all articles in these languages should be moved to the external wiki): Deutsch – Français – Română – Suomi – Svenska – Tiếng Việt – Türkçe – فارسی

frandom is a fast alternative to /dev/urandom. It can be used wherever fast random number generation is required, eg for randomising large hard drives prior to encryption.

From the frandom page: "The frandom suite comes as a Linux kernel module for several kernels, or a kernel patch for 2.4.22. It implements a random number generator, which is 10-50 times faster than what you get from Linux' built-in /dev/urandom."

Does frandom generate good random numbers? Refer to the frandom page for this and other technical info.

Installation

Frandom is available as a package from the AUR.

Once the daemon has been started, it is available from . It is run in the normal way:

# /etc/rc.d/frandom {start|stop|restart}

Or if you prefer, it can be started at boot by adding it /etc/rc.conf:

DAEMONS=(... frandom ...)

Wiping a drive/partition

Use the following dd command. This will wipe all the data on the specified device, take care!

# dd if=/dev/frandom of=/dev/sdx1

Example

1) On a 1.73 GHZ Thinkpad T43 with 2 GB ram:

# time dd if=/dev/frandom of=/dev/sdb2
 dd: writing to `/dev/sdb2': No space left on device
 587384596+0 records in
 587384595+0 records out
 300740912640 bytes (301 GB) copied, 12844.6 s, 23.4 MB/s
 real    214m4.620s
 user    3m34.693s
 sys     77m28.660s

Summary: 300 GB in approx 3.5 hours


2) On a 2.4 GHZ (T8300 Core 2 Duo) Thinkpad T61 with 2 GB ram:

# dd if=/dev/frandom of=/dev/sdb bs=1M
  dd: writing `/dev/sdb': No space left on device
  476941+0 records in
  476940+0 records out
  500107862016 bytes (500 GB) copied, 5954.52 s, 84.0 MB/s

Summary: 500 GB in approx 1.65 hours


3) On a 2.8 GHz (Athlon2 X4) with 4 GB ram:

# dd if=/dev/frandom of=/dev/sdc3 bs=1M seek=100KB
  dd: writing `/dev/sdc3': No space left on device
  1807429+0 records in
  1807428+0 records out
  1895225712640 bytes (1.9 TB) copied, 20300.3 s, 93.4 MB/s

Summary: ~2TB in ~5.64 hours. However, on the same machine:

# dd if=/dev/frandom of=/dev/null bs=1M count=1000
  1000+0 records in
  1000+0 records out
  1048576000 bytes (1.0 GB) copied, 7.81581 s, 134 MB/s

versus

# dd if=/dev/urandom of=/dev/null bs=1M count=1000
  1000+0 records in
  1000+0 records out
  1048576000 bytes (1.0 GB) copied, 144.296 s, 7.3 MB/s

This makes frandom 10-20 times faster on this machine, meaning it would take approx 50-120 hours (2-5 days!) to randomize 2TB using urandom.

4) On a 2.70GHz (i7-2620M) ThinkPad x220 with 8GB Ram:

# time dd if=/dev/frandom of=/dev/sdc
  dd: writing to `/dev/sdc': No space left on device
  625140336+0 records in
  625140335+0 records out
  320071851520 bytes (320 GB) copied, 9618.12 s, 33.3 MB/s
  real    160m18.126s
  user    1m8.916s
  sys     36m16.401s

Summary: 320GB in approx. 2.67 hours

What does 'real', 'user' and 'sys' mean?