Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 25

Thread: Debian: kFreeBSD 9.0 Kernel Competing Against Linux 3.2

  1. #11

    Default

    Interesting benchmarks in itself but it feels meaningless:
    Quote Originally Posted by BagOfMostlyWater View Post
    Can you also perform the tests on identical hardware?
    and identical filesystems? Not all the benchmarks appear to be a kernel-only test or am I wrong?

    Quote Originally Posted by elanthis View Post
    Some analysis as to what caused those huge differences in performance would be welcome.
    Yes. I get it, the KFreeBSD kernel is generally slower but occasionally remarkably faster. Why? Were these results expected? Can you make the data accessible to noobs?

  2. #12
    Join Date
    Feb 2012
    Posts
    111

    Default

    Different hardware? sorry, but this can't be called a benchmark ... or at least, you can't compare both stuff, *again*, when it came to comparing *BSD to a Linux, you did a bad developed benchmark.
    Well, at least you didn't make the "mistake" of comparing ZFS vs ext4, and instead used ufs2
    Last edited by vertexSymphony; 03-07-2012 at 01:47 AM.

  3. #13
    Join Date
    Feb 2012
    Posts
    111

    Default

    Quote Originally Posted by Michael View Post
    Ditto. Same hardware, but Debian GNU/kFreeBSD doesn't properly expose multiple sockets for PTS to read.

    Whoops, sorry then ... I remember your postings in the mailing lists ( was it freebsd-hackers or freebsd-current? ) asking about a proper way of reading hardware data.

  4. #14
    Join Date
    Jan 2011
    Posts
    1,287

    Default

    Quote Originally Posted by kraftman View Post
    I hope Debian will get rid of kfreebsd kernel and simply focus on Linux and systemd, so it will be easier to Ubuntu to switch. The benchmark results are great. It's good to see the same GCC version was used.
    And why would Debian care about how easy or hard is it for Ubuntu?

  5. #15

    Default

    Quote Originally Posted by mrugiero View Post
    And why would Debian care about how easy or hard is it for Ubuntu?
    Maybe because Ubuntu propagates Debian's way when comes to few things like apt, debs. Ubuntu and Linux kernel (and kernel related things like systemd) are far more important for Debian than bsd crap. I don't like Debian cares about crap like kfreebsd and llvm as GCC replacement which just makes things more complicated. They're free to do what they want, but they do stupid things sometimes.

  6. #16
    Join Date
    Jan 2011
    Posts
    1,287

    Default

    Quote Originally Posted by kraftman View Post
    Maybe because Ubuntu propagates Debian's way when comes to few things like apt, debs. Ubuntu and Linux kernel (and kernel related things like systemd) are far more important for Debian than bsd crap. I don't like Debian cares about crap like kfreebsd and llvm as GCC replacement which just makes things more complicated. They're free to do what they want, but they do stupid things sometimes.
    Don't get me wrong, I don't think caring about an experimental GNU/FreeBSD is better or worst than caring about another GNU/Linux distro (in fact, if they care for Ubuntu, better for me, since it's my main distro), nor about switching the main compiler. But I really don't understand why would they care about Ubuntu. Even when they propagate their way in such things, I don't think that following up Ubuntu decissions (like switching to systemd) can make sense from that point. If they do that, they're in fact following the Ubuntu way now, not the other way around like you suggest.
    With that criteria, Debian should start using Unity for default, so Ubuntu can benefit of the manpower...

  7. #17

    Default

    Quote Originally Posted by mrugiero View Post
    But I really don't understand why would they care about Ubuntu. Even when they propagate their way in such things, I don't think that following up Ubuntu decissions (like switching to systemd) can make sense from that point. If they do that, they're in fact following the Ubuntu way now, not the other way around like you suggest.
    With that criteria, Debian should start using Unity for default, so Ubuntu can benefit of the manpower...
    In my first comment I didn't say they should care about Ubuntu and you can ignore my second one, because I was thinking about something else. I read somewhere it will be easier to Ubuntu to go for systemd if Debian will switch to this init system. While Debian ships kfreebsd which doesn't support systemd it can may some impact on their decision about switching or not. I'd love to see systemd in Ubuntu, but they say something like there's no reason to switch yet etc. So it's not that Ubuntu have impact on what Debian does, but it's clearly opposite in my opinion. If Debian will switch it will influence Ubuntu to do the same, because they will have less work than go alone with Upstart.

  8. #18
    Join Date
    Jan 2011
    Posts
    1,287

    Default

    Quote Originally Posted by kraftman View Post
    In my first comment I didn't say they should care about Ubuntu and you can ignore my second one, because I was thinking about something else. I read somewhere it will be easier to Ubuntu to go for systemd if Debian will switch to this init system. While Debian ships kfreebsd which doesn't support systemd it can may some impact on their decision about switching or not. I'd love to see systemd in Ubuntu, but they say something like there's no reason to switch yet etc. So it's not that Ubuntu have impact on what Debian does, but it's clearly opposite in my opinion. If Debian will switch it will influence Ubuntu to do the same, because they will have less work than go alone with Upstart.
    Sorry, I misread. I missunderstood and thought you meant Debian devs actually could take that into account at the time of making a call. Now that you clarified it, I can say I agree with you :P

  9. #19
    Join Date
    Mar 2012
    Posts
    4

    Default Windows reference

    Sometimes it would be nice to have Windows in there as a reference ..

  10. #20
    Join Date
    Jan 2011
    Posts
    1,287

    Default

    Quote Originally Posted by brunis View Post
    Sometimes it would be nice to have Windows in there as a reference ..
    That's not relevant here. The point of the benchmark is not to make an objective speed scale, but just to compare kFreeBSD with Linux kernel (even though there are some doubtful conditions and such), so the Windows reference is unneeded here.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •