[TriLUG] Slightly OT: Which *BSD?

Brandon Newport bnewport at appws.com
Mon Sep 20 06:31:13 EDT 2004


I would agree that is a great site for comparing benchmarks.  I was not
stating that one was better than the other.  I think all have their place.
All I said was that during our testing FreeBSD performed better than Linux.
The interesting part was we all expected Linux to win hands down.  Since we
already knew Linux and how to tweak it a little, it took us a while to learn
the FreeBSD generic (kernel) files.  When we were done with everything FreeBSD
just benchmarked better.  Again this was over 2 years ago and I firmly believe
that hardware support has a LOT to do with it.

-brandon

Aaron S. Joyner (aaron at joyner.ws) wrote:
>
> Brandon L. Newport wrote:
>
> >The main thing we noticed was the network performance...it did not seem
> >to really slow under high loads.  However overall it just tested
> >better...disk IO/mem/CPU/etc  We tested with databases, web servers,
> >ftp, etc.  Over all it just did better.  That was a little over two
> >years ago.  We will probably test again next year.  One interesting
> >thing to note for anyone who has the on board promise IDE raid
> >cards...they work 200% better in FreeBSD than Linux....you can find a
> >dozen google search that talk about it.  It is kinda interesting.  Our
> >corporate web server uses IDE drives.  Other than we dont have enough
> >memory in it right now it does great!!! (shoe maker kids)
> >
> >-brandon
> >
> >
> >
> Okay, I feel I have to set the record straight on this one.  Depending
> on your application, it's entirely possible that you'd see equal or
> better performance in some applications with OpenBSD.  But to say that
> performance across the board is better is at the least, misleading.  As
> opposed to attempting to clarify all of the results here, I'll just post
> a link to the definitive comparison (at least in my mind) of the various
> open source *NIX solutions.  Check it out here:
> http://bulk.fefe.de/scalability/  In summary, some of the OSes are
> better at some things than others, and none of them have a uncontested
> crown in all circumstances.  This study deals particularly with how an
> OS scales under higher loads, but if read carefully also provides
> insight into which core functions of each OS have been optimized in
> which ways.
>
> Aaron S. Joyner
> --
> TriLUG mailing list        : http://www.trilug.org/mailman/listinfo/trilug
> TriLUG Organizational FAQ  : http://trilug.org/faq/
> TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
> TriLUG PGP Keyring         : http://trilug.org/~chrish/trilug.asc
>




More information about the TriLUG mailing list