LinuxQuestions.org
Help answer threads with 0 replies.
Go Back   LinuxQuestions.org > Forums > Other *NIX Forums > *BSD
User Name
Password
*BSD This forum is for the discussion of all BSD variants.
FreeBSD, OpenBSD, NetBSD, etc.

Notices

Reply
 
Search this Thread
Old 11-30-2005, 08:52 PM   #1
Synesthesia
Member
 
Registered: Jan 2004
Location: the abyss
Posts: 205

Rep: Reputation: 30
[OS performance comparison]


If you are not aware of these OS benchmarks, you might want to check them out:
http://bulk.fefe.de/scalability/

I didn't realize OpenBSD has such performance issues (in comparison to the rest). Has anyone here experienced these performance differences when running openbsd in comparison with any of the other OS's (especially for a server)? FreeBSD really kicks the crap out of it, and so does netBSD in some tests...
 
Old 12-01-2005, 09:42 AM   #2
cnjohnson
Member
 
Registered: Nov 2002
Location: Nashville
Distribution: FreeBSD, Linux, OS-X
Posts: 544

Rep: Reputation: 30
Re: [OS performance comparison]

Quote:
Originally posted by Synesthesia
If you are not aware of these OS benchmarks, you might want to check them out:
http://bulk.fefe.de/scalability/

I didn't realize OpenBSD has such performance issues (in comparison to the rest). Has anyone here experienced these performance differences when running openbsd in comparison with any of the other OS's (especially for a server)? FreeBSD really kicks the crap out of it, and so does netBSD in some tests...
I use freebsd (started with 4.8) for both servers and desktops.

It is important to remember what the design goals for an OS are. For OpenBSD it was (and remains) high security. Whether they have achieved it is open for discussion, and I, for one, remain sceptical. However, that goal will always impose performance penalties; there is simply no way around that. So, while I find the results more than acceptable (I use freebsd afterall ), I am not put off by Open's results. I will be most interested in how they fare over time. Clearly they have a lot to do, but getting security right is not easy.

Cheers--
Charles
 
Old 12-04-2005, 04:35 AM   #3
rehab junkie
Member
 
Registered: Nov 2003
Location: /var/local/pub/bar
Distribution: OSX 10.4.9
Posts: 259

Rep: Reputation: 30
Quote:
Originally Posted by cnjohnson
For OpenBSD it was (and remains) high security.
Not quite... their ultimate goal is correctness in everything they do, through things like constantly auditing all code. From this, increased security will naturally follow.
 
Old 12-04-2005, 03:28 PM   #4
cnjohnson
Member
 
Registered: Nov 2002
Location: Nashville
Distribution: FreeBSD, Linux, OS-X
Posts: 544

Rep: Reputation: 30
Since one of the core openBSD developers is a member of the Nashville Linux User Group (of which I am immediate past president ), and I have had to sit through more than one obsd lecture on high security coupled with integrated cryptography, I'll stand by my assertion.

HTH

Cheers--
Charles
 
Old 12-05-2005, 07:51 PM   #5
SteveK1979
Member
 
Registered: Feb 2004
Location: UK
Distribution: RHEL5/6, Solaris 10/11, NetBSD, OpenBSD, FreeBSD, MacOS
Posts: 221

Rep: Reputation: 40
I obviously don't know anyone involved with OpenBSD like Charles, but what I would say is that it's a test that was last updated two years ago, and for example OpenBSD on this test is 3.4, a version that is now more than 2 years old and not supported any longer.

I'm fairly sure these tests have been discussed here before, but personally, I prefer to test out the operating systems with the applications etc that I want to use rather than relying on benchmarks such as these. When I last tried out FreeBSD, my perception was that it was sluggish, using a lot more memory than I expected and I just wasn't happy using it. I ended up putting Solaris 8 backon the machine (it's an UltraSparc CPU) and now it's the most solid machine I own. And to be honest, Solaris is probably the OS I've seen handle an absolute thrashing better than any other (think load averages in the 50's every day for months). But it does make an interesting read, and the 2.6 kernel did kick some ass in this particular instance

Anyhow, what does all that mean? Nothing except I'd caution making judgements based too heavily on benchmarks such as these.

Cheers,
Steve

Last edited by SteveK1979; 12-05-2005 at 07:52 PM.
 
  


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 Off
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Why does linux performance seem so terrible in comparison to windows? hermaphromike Linux - Newbie 17 04-30-2005 08:37 PM
comparison izza_azhar Programming 4 01-19-2005 11:57 PM
Comparison introuble Programming 1 01-18-2005 02:57 PM
Comparison halobungie Linux - Newbie 4 08-10-2003 03:55 PM
*comparison in C simbo Programming 2 04-22-2003 11:05 PM


All times are GMT -5. The time now is 05:56 PM.

Main Menu
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
identi.ca: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration