[ntp:questions] Proposed NTP solution for a network

David Woolley david at ex.djwhome.demon.co.uk.invalid
Tue Mar 3 08:40:31 UTC 2009

Jason wrote:

> "precision" (that is, all the hosts should be within a few 10s of 
> microseconds, but they could be as much as a small hundreds of 

That is going to require very careful engineering. You may find it 
better to link the sites only via GPS, rather than have to prioritise 
NTP traffic on the network.

It's somewhat unusual to have data centres that don't use Windows these 
days, as with Windows, you would be talking about 10ms+ at the 
application level.

> microseconds off of UTC).

> 2. Stratum 1 level: Each Appliance has an output at Stratum 1 via the 
> Ethernet connection. Each appliance should be a peer to the other 

I am a bit queasy about peering without having enough servers to outvote 
the peers.

More information about the questions mailing list