[ntp:hackers] A stop-gap authenticated time service

Magnus Danielson magnus at rubidium.dyndns.org
Wed Nov 11 19:47:01 UTC 2015


Greg,

TDEV isn't particularly useful on itself here, since NTP do some min 
processing to filter out some of the worst noise, and then a minTDEV 
matching that filtering is more relevant. Plotting minTDEV for various 
sample window sizes can be quite telling.

Similarly could a minADEV and minMDEV be plotted, if you want frequency 
stability rather than phase stability.

Network "noise" breaks a bunch of assumptions, such as being stable.

Cheers,
Magnus

On 11/11/2015 04:22 PM, Greg Dowd wrote:
> Jitter isn't the best metric for analyzing this data if you are looking at performance over time.  Some type of variance metric like tdev is good.  Jitter is more white.  A millisecond of jitter on measurements a day apart is far different than the same measurement 1s apart.  For those who find statistical analysis of the type interesting, maybe we could pull some of these datasets together and see if we can extract some useful conclusions.  The open source ntp does move the Allan intercept around based on jitter and variance.  I tend to not like staying in the pll mode as it steers the clock pretty aggressively in quiet networks and gets trapped in network path changes since it keeps the poll interval smaller.
>
> On Nov 11, 2015 7:09 AM, brian utterback <brian.utterback at oracle.com> wrote:
> EXTERNAL EMAIL
>
>
> On 11/11/2015 3:32 AM, Poul-Henning Kamp wrote:
>> --------
>> In message <20151111065510.GT11550 at localhost>, Miroslav Lichvar writes:
>>
>>> What has changed significantly since the old days is the network
>>> jitter. It went down by orders of magnitude, so the Allan intercept is
>>> much shorter than before and ntpd is too slow now.
>> Do you have measurements to back this up, and if so, are they
>> representative of the median network paths rather than one particular
>> installation ?
>>
>
> I'm with PHK on this. The minimum RTT has fallen and with it the minimum
> jitter, but my experience is that jitter across longer network paths has
> gone up if anything. I don't have any generally applicable data to back
> that up and your mileage may vary, but I haven't seen a decrease of
> jitter on my networks.
>
> --
> Oracle <http://www.oracle.com>
> Brian Utterback | Principle Software Engineer
> Phone: +1 6038973049 <tel:+1%206038973049>
> Oracle Systems/RPE Solaris Network
> 1 Oracle Dr. | Nashua, NH 03062
> ------------------------------------------------------------------------
> All working systems eventually start to exhibit their own agenda
> ------------------------------------------------------------------------
> Green Oracle <http://www.oracle.com/commitment> Oracle is committed to
> developing practices and products that help protect the environment
> _______________________________________________
> hackers mailing list
> hackers at lists.ntp.org
> http://lists.ntp.org/listinfo/hackers
> _______________________________________________
> hackers mailing list
> hackers at lists.ntp.org
> http://lists.ntp.org/listinfo/hackers
>


More information about the hackers mailing list