ntp daemon wakeup/sleep patch by RedHat - NTP

This is a discussion on ntp daemon wakeup/sleep patch by RedHat - NTP ; Hello list, Can anyone please tell me if the patch from RedHat to prevent the NTP daemon to wake up every second has been or will be committed to a mainline release? Patch was written as a response to the ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: ntp daemon wakeup/sleep patch by RedHat

  1. ntp daemon wakeup/sleep patch by RedHat

    Hello list,

    Can anyone please tell me if the patch from RedHat to prevent the NTP daemon
    to wake up every second has been or will be committed to a mainline release?
    Patch was written as a response to the powertop program which identifies the
    top power consuming applications on systems.

    RedHat patch to ntp:
    http://cvs.fedora.redhat.com/viewcvs....patch?rev=1.1

    PowerTOP's description of NTP's
    http://www.linuxpowertop.org/known.php

    Thanks in advance.
    _______________________________________________
    questions mailing list
    questions@lists.ntp.isc.org
    https://lists.ntp.isc.org/mailman/listinfo/questions


  2. Re: ntp daemon wakeup/sleep patch by RedHat

    NTP Power Consuming? Compared to what? I run a 300MHz Geode based NTP box
    and hardly see *any* CPU usage at all (0.01 maybe?). That machine consumes
    maybe 10W of power... 15W peak if I'm compiling stuff.

    I looked on the linuxpowertop site and they don't give any sort of numbers
    on usage or savings with the patch.

    Jason


    >Hello list,
    >
    >Can anyone please tell me if the patch from RedHat to prevent the NTP

    daemon
    >to wake up every second has been or will be committed to a mainline

    release?
    >Patch was written as a response to the powertop program which identifies

    the
    >top power consuming applications on systems.
    >
    >RedHat patch to ntp:
    >http://cvs.fedora.redhat.com/viewcvs...l/ntp-4.2.4-sl

    eep.patch?rev=1.1
    >
    >PowerTOP's description of NTP's
    >http://www.linuxpowertop.org/known.php
    >
    >Thanks in advance.


    _______________________________________________
    questions mailing list
    questions@lists.ntp.isc.org
    https://lists.ntp.isc.org/mailman/listinfo/questions


  3. Re: ntp daemon wakeup/sleep patch by RedHat

    In article <001501c7adf1$81109660$0a00a8c0@Inspiron>,
    jason@extremeoverclocking.com (Jason Rabel) writes:

    >NTP Power Consuming? Compared to what? I run a 300MHz Geode based NTP box
    >and hardly see *any* CPU usage at all (0.01 maybe?). That machine consumes
    >maybe 10W of power... 15W peak if I'm compiling stuff.


    The target is clients rather than servers. Consider a battery
    powered laptop. A lot of work has gone into this area recently.

    I don't have hard numbers and they will depend upon what you
    are doing, but doubling battery time would be my ballpark sort of
    guess. You can probably get much better than that is some
    interesting cases.

    One idea is to turn off the CPU between keystrokes and packets.
    (DRAM does self refresh.) Not idle/sleep, but off, as in remove
    the power.

    It takes 10-100 ms (handwave) to power up and go back to sleep.
    Idle loops that wakeup every second but don't do anything useful
    can use a lot of battery.

    There is probably a mechanism to synchronize that sort of loop
    when it is really needed. All N jobs get to run with one power up
    cycle rather than bouncing on/off N times.



    I haven't looked at the code yet. If it looks reasonable, I'd
    encourage folding it into the main codebase.

    --
    These are my opinions, not necessarily my employer's. I hate spam.


  4. Re: ntp daemon wakeup/sleep patch by RedHat

    Halim,

    It should be widely understood that patching out the 1-s will completely
    disable any verions of NTP. That done, no polls will be sent, the
    frequency will not be disciplined, and on and on.

    Dave

    Halim Issa wrote:
    > Hello list,
    >
    > Can anyone please tell me if the patch from RedHat to prevent the NTP daemon
    > to wake up every second has been or will be committed to a mainline release?
    > Patch was written as a response to the powertop program which identifies the
    > top power consuming applications on systems.
    >
    > RedHat patch to ntp:
    > http://cvs.fedora.redhat.com/viewcvs....patch?rev=1.1
    >
    > PowerTOP's description of NTP's
    > http://www.linuxpowertop.org/known.php
    >
    > Thanks in advance.
    > _______________________________________________
    > questions mailing list
    > questions@lists.ntp.isc.org
    > https://lists.ntp.isc.org/mailman/listinfo/questions
    >


  5. Re: ntp daemon wakeup/sleep patch by RedHat


    >It should be widely understood that patching out the 1-s will completely
    >disable any verions of NTP. That done, no polls will be sent, the
    >frequency will not be disciplined, and on and on.


    I haven't looked at the patch, but I assume it is peeking ahead
    to find out when the next poll will happen and does one long
    sleep rather than N sleeps for 1 second.

    If the poll interval is 256 seconds and you have 4 servers configured,
    that's 4 wakesup vs 256. I guess it would be 8 if the responses are slow
    enough.

    --
    These are my opinions, not necessarily my employer's. I hate spam.


  6. Re: ntp daemon wakeup/sleep patch by RedHat

    Hal,

    There are two reasons why this is a terrible idea. In your example you
    assume that only one sleep per server, rather than one sleep per second.

    The first problem is that the frequency is disciplined once each second.
    While in principle it would be possible to recompute the frequency at
    longer intervals, the code would become extremely complex and fragile.
    The maximum interval between frequency updates depends on the time
    constant, which changes with poll interval.

    The second problem is that several other things are going on at the
    seconds sleep. These include interface scan, statistics recording,
    leapseconds counter, garbage collect expired keys, manage the
    huff-n'-puff filter and write the frequency file.

    In priniciple a conventional multiple-timer facility could be written
    and this might make everybody happy. In fact, once upon a time such a
    facility was used, but it became overwhelmingly complex and fragile, so
    was replace by the current crudity.

    So, if Linux has a patch, the patchers need to address the issues in
    this message.

    Dave

    Hal Murray wrote:
    >>It should be widely understood that patching out the 1-s will completely
    >>disable any verions of NTP. That done, no polls will be sent, the
    >>frequency will not be disciplined, and on and on.

    >
    >
    > I haven't looked at the patch, but I assume it is peeking ahead
    > to find out when the next poll will happen and does one long
    > sleep rather than N sleeps for 1 second.
    >
    > If the poll interval is 256 seconds and you have 4 servers configured,
    > that's 4 wakesup vs 256. I guess it would be 8 if the responses are slow
    > enough.
    >


+ Reply to Thread