[ntp:questions] More Granularity in the US in the NTP Pool

Richard B. Gilbert rgilbert88 at comcast.net
Wed Sep 12 01:47:48 UTC 2007


Hal Murray wrote:
>>Suppose a ntpd client can pick up 5 random pool servers, and
>>periodically (say once a day) replace the 'worst' server
>>by randomly picking a new one.
> 
> 
> I like that suggestion.  But, ignoring implementation details,
> how do you decide which of two servers is "better"?
> 
> Is a server with 20 ms of jitter all of the time better or
> worse than a server with 10 ms of jitter most of the time but
> 25 ms of jitter for an hour a day?
> 

How about the same way ntpd picks them; lowest synchronization distance 
wins!  Or, since synchronization distance tends to vary with the time of 
day (or how busy the network is) the lowest average synchronization 
distance over a period of 24 or 48 hours.





More information about the questions mailing list