[ntp:questions] NTP absolute accuracy?

Juyong Do jdo at stanford.edu
Mon Jul 2 17:34:27 UTC 2007


Thanks Jason.

I guessed so since it wouldn't be possible under the network environments as
you pointed out.

Then, my question is how to detect a case with absolute time inaccuracy
beyond a certain limit? Is the Root Dispersion the best indicator? How about
Offset and RTT? Also, I'm wondering whether there is any way to find out the
status of network connection such as software and hardware delays---are
there any parameters about them?

Basically, I'm trying to find or combine parameters to detect a certain
outage case where absolute time inaccuracy exceeds my limit. Let me know if
anyone has experience on this.

Thanks,
Juyong

"Jason Rabel" <jason at extremeoverclocking.com> wrote in message 
news:004b01c7bb2f$e0c51d60$7b01a8c0 at Inspiron...
> Juyong,
>
> You can't really give an "absolute" number when there is so much 
> uncertainty
> and variety in every NTP implementation. Factors such as system hardware,
> software, and everything network related, eliminate any possibility of 
> doing
> apples to apples comparisons. Even the smallest change can mean large
> differences in performance numbers.
>
> Keep in mind only a couple platforms will support a PPS signal.
>
> Read up on Root Dispersion, I think that number might lead you in the 
> right
> direction on your quest.
>
>
>> Hi,
>>
>> Does anyone know any survey or paper measuring absolute time accuracy 
>> over
>
>> NTP? I saw statistics about offsets, RTT, and dispersion but am not sure
> how
>> well those parameters reflect or are correlated with absolute accuracy of
>> each NTP servers and clients.
>>
>> To my mind, the only way to measure the absolute accuracy (difference 
>> from
>
>> UTC or GPS time) would be to physically bring a GPS receiver to each NTP
>> clients and compare the timing bases. It is certainly not feasible for a
>> large scale survey but I guess there should have been someone who've done
>> this before. Let me know if there is.
>>
>> Thanks,
>> Juyong
>
> _______________________________________________
> questions mailing list
> questions at lists.ntp.isc.org
> https://lists.ntp.isc.org/mailman/listinfo/questions
> 





More information about the questions mailing list