LinuxQuestions.org
Visit Jeremy's Blog.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - General
User Name
Password
Linux - General This Linux forum is for general Linux questions and discussion.
If it is Linux Related and doesn't seem to fit in any other forum then this is the place.

Notices


Reply
  Search this Thread
Old 03-10-2016, 02:04 AM   #1
rng
Senior Member
 
Registered: Aug 2011
Posts: 1,077

Rep: Reputation: 43
Speed comparison of different Linux distributions


I ran a small program (in R statistical programming language) meant to find prime numbers on a desktop which had 7 Linux distributions installed. The program runs for 10 seconds, produces continuous terminal output of steps and finally lists all the prime numbers found. The program was run multiple times to confirm that the results were very similar. Following are the number of prime numbers found in 10 seconds (indicating speed) in different distributions:

Code:
Slackware14.1-32bit-KDE	   27785
Salix14.1-32bit-KDE	   26896
Manjaro-64bit-KDE	   25116
Debian8(stable)32bit-KDE   23890
Mageia5-32bit-KDE	   20569
Salix14.1-32bit-Mate	   14747
Manjaro-64bit-Mate	   14730
In addition to comparison of distributions, it also seems that the Mate desktop is slower in terms of calculations &/or terminal output than KDE desktop (even though Mate distributions need much less RAM and other resources to run). Are these findings as expected?

Last edited by rng; 03-10-2016 at 03:26 AM.
 
Old 03-10-2016, 07:35 AM   #2
sundialsvcs
LQ Guru
 
Registered: Feb 2004
Location: SE Tennessee, USA
Distribution: Gentoo, LFS
Posts: 9,078
Blog Entries: 4

Rep: Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175Reputation: 3175
No, I personally don't find sufficient basis to think that, say, Manjaro, if equally(!) configured, is somehow "twice as fast" as, say, Slackware. I predict that there is an as-yet unaccounted-for bias in these statistics.
 
Old 03-10-2016, 07:37 AM   #3
rtmistler
Moderator
 
Registered: Mar 2011
Location: MA, USA
Distribution: MINT Debian, Angstrom, SUSE, Ubuntu, Debian
Posts: 7,779
Blog Entries: 13

Rep: Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272
Interesting test.

I personally would expect them all to be largely close in value as opposed to the ones with MATE showing a large degradation in speed.

And what can I say .. Go Slackware!
 
Old 03-10-2016, 08:30 AM   #4
pan64
LQ Guru
 
Registered: Mar 2012
Location: Hungary
Distribution: debian/ubuntu/suse ...
Posts: 12,797

Rep: Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019Reputation: 4019
I have no idea why, but this is definitely misleading. You measured something else, like cache/buffer/swap/io/gui/whatever size/speed/whatever/else.
 
Old 03-10-2016, 11:10 AM   #5
rng
Senior Member
 
Registered: Aug 2011
Posts: 1,077

Original Poster
Rep: Reputation: 43
Quote:
Originally Posted by sundialsvcs View Post
No, I personally don't find sufficient basis to think that, say, Manjaro, if equally(!) configured, is somehow "twice as fast" as, say, Slackware. I predict that there is an as-yet unaccounted-for bias in these statistics.
The numbers mean higher the faster. Hence, Slackware is about twice as fast as Manjaro (and Salix) with Mate desktop in this test.

Last edited by rng; 03-10-2016 at 11:13 AM.
 
Old 03-10-2016, 12:11 PM   #6
salasi
Senior Member
 
Registered: Jul 2007
Location: Directly above centre of the earth, UK
Distribution: SuSE, plus some hopping
Posts: 4,070

Rep: Reputation: 897Reputation: 897Reputation: 897Reputation: 897Reputation: 897Reputation: 897Reputation: 897
You didn't make the mistake of installing them all at once, did you?
 
Old 03-10-2016, 12:41 PM   #7
rtmistler
Moderator
 
Registered: Mar 2011
Location: MA, USA
Distribution: MINT Debian, Angstrom, SUSE, Ubuntu, Debian
Posts: 7,779
Blog Entries: 13

Rep: Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272Reputation: 3272
Quote:
Originally Posted by salasi View Post
You didn't make the mistake of installing them all at once, did you?
Quote:
Originally Posted by rng View Post
... on a desktop which had 7 Linux distributions installed.
Salasi,

I'm thinking the answer to your question is "Yes"
 
Old 03-10-2016, 04:38 PM   #8
jefro
Moderator
 
Registered: Mar 2008
Posts: 19,024

Rep: Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881Reputation: 2881
You should have built your system for this test. I would have used Gentoo maybe. Using any stock distro is hard to determine what all the possible kernel and other settings could be. Common distro's are made to be rather generic and to suit a wide range of hardware and software.

I have no idea if the findings are correct. We have no idea the test hardware and setttings and what all was installed or running at the time.

From a very general test on this one system using some setup on some test the numbers may be as expected.

Last edited by jefro; 03-11-2016 at 07:23 PM.
 
Old 03-10-2016, 04:53 PM   #9
OregonJim
Member
 
Registered: Feb 2016
Posts: 98

Rep: Reputation: Disabled
Quote:
Originally Posted by rng View Post
Are these findings as expected?
The findings are not even close to being what I would call an informed conclusion.

Are all the distributions running the same kernel? Same modules? Same daemon processes? Same executable test program compiled with the same version of libraries?

Statistics can be interpreted to say whatever you want. You need to define your goal and eliminate as many variables as possible. The playing field here seems to be about as level as the Swiss Alps.
 
Old 03-10-2016, 05:52 PM   #10
suicidaleggroll
LQ Guru
 
Registered: Nov 2010
Location: Colorado
Distribution: OpenSUSE, CentOS
Posts: 5,573

Rep: Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133Reputation: 2133
The choice of DE should have absolutely no effect on the raw computation speed in a command line program, barring low memory issues. Something is up.
 
Old 03-10-2016, 06:41 PM   #11
fogpipe
Member
 
Registered: Mar 2011
Distribution: Slackware 64 -current,
Posts: 550

Rep: Reputation: 194Reputation: 194
Yay team!
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
Comparison chart of Server Distributions josegmay Linux - Server 6 05-24-2012 04:23 AM
[SOLVED] Linux distros comparison (stability, speed, security) alanv Linux - Newbie 24 11-09-2010 11:23 PM
LXer: A Comparison of IRC support between GNU/Linux distributions LXer Syndicated Linux News 0 01-18-2010 09:30 AM
speed comparison alvi2 Linux - Networking 3 04-02-2005 11:50 AM
speed comparison alvi2 Fedora 1 04-02-2005 11:42 AM

LinuxQuestions.org > Forums > Linux Forums > Linux - General

All times are GMT -5. The time now is 01:48 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration