[TICTOC] Re: TICTOC NTP ensembling

Attending the WSTS07 last week, I was left with the impression that people
considered one of the good things about NTP is the ability to form a clock
from an ensemble of master clocks. However, I see this as another
potentially serious source of noise which may swamp the output noise budget
for many telecom applications. I can see that having an ensemble could
offer good performance, but I think that would only be acheived if the
ensemble was a stable collection. If master clocks can appear and
disappear, then their individual biasses will pull the output phase around
a little. For the generally-accepted accuracy of 1-10ms, this effect is
probably not visible, but when the application requires +/-500ns accuracy
then I think this problem will dominate. I am not an NTP expert, but I can
imagine this problem is more difficult to solve than PDV-filtering. I would
also like to know what the carriers think of this - do they consider this
approach to be "carrier-class"?

Dave

Semtech Ltd
Office address:  2-3 Park Court, Premier Way, Abbey Park Ind Est, Romsey,
SO51 9DN
Tel: +44 (0)1794 527600           Fax:  +44 (0)1794 527601

Registered office: 4th Floor, Saltire Court, 20 Castle Terrace, Edinburgh,
EH1 2EN
Registered in Scotland 53259
VAT No. GB572685604

_______________________________________________
TICTOC mailing list
TICTOC@ietf.org
https://www1.ietf.org/mailman/listinfo/tictoc