Re: [Lse-tech] [patch] sched-domain cleanups, sched-2.6.5-rc2-mm2-A3

From: Ingo Molnar (mingo_at_elte.hu)
Date: 03/30/04

  • Next message: Marc-Christian Petersen: "Re: Linux 2.6.5-rc3" 
    Date: 	Tue, 30 Mar 2004 10:45:01 +0200
To: Nick Piggin <nickpiggin@yahoo.com.au>

    * Nick Piggin <nickpiggin@yahoo.com.au> wrote:

    > You're probably mostly right, but I really don't know if I'd start
    > with the assumption that threads don't share anything. I think they're
    > very likely to share memory and cache.

    it all depends on the workload i guess, but generally if the application
    scales well then the threads only share data in a read-mostly manner -
    hence we can balance at creation time.

    if the application does not scale well then balancing too early cannot
    make the app perform much worse.

    things like JVMs tend to want good balancing - they really are userspace
    simulations of separate contexts with little sharing and good overall
    scalability of the architecture.

    > Also, these additional system wide balance points don't come for free
    > if you attach them to common operations (as opposed to the slow
    > periodic balancing).

    yes, definitely.

    the implementation in sched2.patch does not take this into account yet.
    There are a number of things we can do about the 500 CPUs case. Eg. only
    do the balance search towards the next N nodes/cpus (tunable via a
    domain parameter).

            Ingo
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  • Next message: Marc-Christian Petersen: "Re: Linux 2.6.5-rc3"

    Relevant Pages

    • Re: Cluebook Conundrum
      ... 'Balancing the Scales' and 'Pentology' were both put out by Origin after being bought by EA. ... make sure you get ones that have color maps listed- not all versions of MD came with them. ...
      (rec.games.computer.ultima.dragons)
    • RE: Delete scheduler SD_WAKE_AFFINE and SD_WAKE_BALANCE flags
      ... >>benefit with zero aggressiveness. ... type of workload has detrimental effect on another workload and vice versa. ... > of the zero balancing case, then that would be preferable I think. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: HARD: Cant Go....:o(
      ... But to you and the other decent people balancing out the scales in New ... Orleans, I apologize if I hurt your feelings. ...
      (alt.smokers.cigars)
    • Re: weight distribution: road vs. mountain bike
      ... I tried doing a track stand while balancing the Allez on ... two bathroom scales, but that didn't work out too well. ... I suggest you get someone ELSE to read the scales, ...
      (rec.bicycles.misc)
    • Re: [sched, patch] better wake-balancing, #3
      ... Ken hasn't measured the effect of wake balancing in 2.6.13, ... Ken's workload, we can actually make it stronger. ... Without wake balancing we basically have a completely random ... In any case, more measurements are needed. ...
      (Linux-Kernel)