[ntp:questions] Clock and Network Simulator

David Woolley david at ex.djwhome.demon.invalid
Wed Jun 30 21:40:59 UTC 2010


unruh wrote:
> On 2010-06-30, David Woolley <david at ex.djwhome.demon.invalid> wrote:

>> Real world NTP networks don't behave like that, and most of the things 
>> that annoy people relate to the real world behaviour.  Real networks are 
>> subject to diurnal and near step changes in frequency.
> 
> ? It seems you are discussing the behaviour of the clocks, not of the
> network. (the network has no frequency). His simulator works with real

The links have no frequency.  The nodes have frequencies.  A network is 
the combination of the two.

> clocks which have exactly the behaviour you describe. Now it may be that
> the network model is not the best ( random "long" delays in one way trip
> time due to network overload).
> 

If he really has distinct nodes, he will have links between them, even 
if they don't use standard technologies.  I'm not entirely convinced 
that he is not running a discrete event simulation.

Even if he is running real nodes in real time, I somehow suspect he is 
going to take steps to eliminate frequency variations, rather than 
making them, and their statistics part of the model.  It is also very 
likely that all the nodes are in the same thermal environment.




More information about the questions mailing list