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

Magnus Danielson magnus at rubidium.dyndns.org
Sun Nov 15 09:24:58 UTC 2015


Poul-Henning,

On 11/14/2015 09:38 AM, Poul-Henning Kamp wrote:
> --------
> In message <56452BAF.7060206 at rubidium.dyndns.org>, Magnus Danielson writes:
>
>> That's where I think the Allan intercept method might not be
>> completely useful in this case.
>
> And just in case people forget:
>
> The "Allan intercept" is a rule of thumb it has no formal mathematical basis.

Actually it does, but you have to dig around to figure it out.
You do the same thing in the phase-noise world. It's the cross-over 
point from two different noise types.

For NTP it relates back to an article by Judah Levine on the ACTS 
system, which got lifted into the NTP world without validating the noise 
properly. My point refers back to the later, but as in the original 
context it was more of a valid analysis than it is now.

The trouble with using the Allan intercept is that was originally used 
for random noise types, but as soon as you introduce systematic noises 
such as network noise (which defies several random noise properties) it 
is not a helpful concept anymore.

So, it has no real validity in out case anyway.

Cheers,
Magnus


More information about the hackers mailing list