[ntp:questions] NTP architecture recommendation

Jason Rabel jason at extremeoverclocking.com
Sat Aug 18 00:07:14 UTC 2007

> Well, I need to clarify "datacenter": actually these are not OUR
> datacenters, we use co-location at some provider, so we don't have a
> lot of control there (some of them will not even allow us to feed
> another cable into their server room). So even using a "remote
> antenna" (on the roof) might not be possible in some areas. It really
> looks like my approach will have to be: get a NTP-appliance with a
> reference clock (like the Meinberg devices) as close to the servers in
> the different sites as possible (ideally via a LAN).

Just a thought, have you talked to the datacenters yet to see if they
already have a local NTP server for their clients to use?

Again, the Meinberg devices are not going to really help you unless you
experience frequent network outages. The NTP version's oscillator is
disciplined via whatever the network accuracy is. It's not going to be any
better or worse than a regular PC (unless your network goes out and can not
sync to outside sources, then the higher grade oscillator will give you more
accurate time longer than a PC would).

What kind of root dispersion are you seeing on your servers currently? Have
you tried to find other NTP servers near your location or just picked some
at random (or using the NTP pool)?

In Europe I don't think there is much CDMA usage, but there are similar GSM
NTP devices.

> Actually I will have to keep the servers in sync for 24 hours in the
> near future (for now 12 hours is ok; we are in financial business and
> will extend our operations to other markets soon, which basically will
> extend our "business hours" to 23 hours a day).
> And btw: we are not using any Windows servers :-).

What do you mean by "in sync"? If you are running NTP then it should always
be "in sync" within a certain degree of accuracy.

More information about the questions mailing list