Re: Performance of freeBSD 5.1

From: Petri Helenius (pete_at_he.iki.fi)
Date: 08/27/03

  • Next message: Alexander Leidinger: "Re: Performance tests I did with FreeBSD, Linux and PostgreSQL" 
    Date: Wed, 27 Aug 2003 09:27:36 +0300
To: TeJun Huh <tejun@aratech.co.kr>

    Did you disable INVARIANTS and WITNESS when running the tests?

    Pete

    TeJun Huh wrote:

    > Hello,
    >
    > I've been conducting performance tests using simple HTTP proxies and
    >web polygraph <http://www.web-polygraph.org/>. Web polygraph is
    >believed to generate realistic Internet load (high number of sessions,
    >many idle, req rate independent of resp rate, and so on). I built
    >several equivalent HTTP proxies using various concurrency mechanisms
    >and comparing those on linux and freeBSD.
    >
    > The following result is obtained on p4 xeon 2.4g (UP kernels).
    >Custom version of Kqueue is used on Linux. FreeBSD 5.1 is updated to
    >the latest cvs tree three weeks ago (due to thread stability issues)
    >and compiled with GENERIC option.
    >
    > FreeBSD Linux
    > Select Kqueue Select Kqueue
    >rps 400 500 500 2500
    >
    > I feel something is very wrong with FreeBSD kqueue result. Used
    >tuning parameters are
    >
    >* Loader tunables
    >
    >kern.maxusers="512"
    >kern.ipc.maxsockets="200000"
    >kern.ipc.nmbclusters="51200"
    >kern.ipc.nmbufs="102400"
    >net.inet.tcp.tcbhashsize="65536"
    >
    >* Sysctls
    >
    >sysctl -w net.inet.ip.portrange.first=1024
    >sysctl -w net.inet.ip.portrange.last=65535
    >sysctl -w kern.ipc.somaxconn=1024
    >sysctl -w kern.maxfiles=200000
    >sysctl -w kern.maxfilesperproc=200000
    >sysctl -w kern.threads.max_threads_per_proc=100000
    >sysctl -w kern.threads.max_groups_per_proc=8000
    >
    > As soon as the test machine becomes available, I'm gonna run tests on
    >freeBSD 4.8 and compare the results. Any suggestions are welcomed.
    >
    >P.S. Please don't forget to CC me. TIA.
    >
    >
    >

    _______________________________________________
    freebsd-performance@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-performance
    To unsubscribe, send any mail to "freebsd-performance-unsubscribe@freebsd.org"

  • Next message: Alexander Leidinger: "Re: Performance tests I did with FreeBSD, Linux and PostgreSQL"

    Relevant Pages

    • Re: FreeBSD 4.x Opteron Question
      ... the FreeBSD developers told everyone that 5.3 was da ... initially over linux not because there's a bunch of good guys on the ... My tests measure kernel performance; ... > a networking device is a key performance indicator. ...
      (freebsd-questions)
    • Re: Newbie Experience
      ... I've only been around since FreeBSD 5.4 ... FreeBSD kernel too. ... always sunshine and linux is farts. ... in the hey day of AT&T Unix I'm ...
      (freebsd-questions)
    • Re: Review of FreeBSD 5.4
      ... but not less problems compared to FreeBSD. ... If you like to have a bleeding edge system using debian --- just go ... > the linux kernel suffers. ... When the kernel suffers, everyone who uses ...
      (comp.unix.bsd.freebsd.misc)
    • Re: FreeBSD & Linux distro
      ... as a FreeBSD advocacy the tone of the article should be neutral and all ... do not like Linux and more over I have never used it in my life but I ... Statement of the type BSD appears more stable than Linux is ... fewer FreeBSD advocates make claims like that, however, is part of the ...
      (freebsd-questions)
    • Re: Compare Linux and Freebsd Redux
      ... >> Perhaps the most unsettling 'misinformation' about the FreeBSD vs. ... >> Linux situation was the spin that Linux was somehow more 'open', ... >> the official Linux tree... ... your own distribution is that the FreeBSD distribution is/was already ...
      (comp.unix.bsd.freebsd.misc)