[ntp:questions] Advice about architecture

Marco Marongiu brontolinux at gmail.com
Fri Nov 4 13:04:42 UTC 2011

Hi all

I am going to revise the design I usually use for our synchronization
subnet. The plan is to take IPv6 in consideration, manycast, and to use
autokey. Anyway, before going any further, I'd like to ask you what you
think about how I _currently_ organize my synchronization subnets.

Normally, in each location I use four multicast servers with symmetric
keys. Each location has very small boundaries, both geographically and
network-wise: it could be a datacenter, or a building, with a limited
number of network switches/routers. In such an environment, network
variables such as delay and latency are usually quite stable, and I feel
that many of the downsides of using multicast are almost negligible.

I have four independent stratum-2 servers, each one syncing with
three/four public stratum one; there is no stratum one in common to any
pair of servers. Plus, three stratum-2 out of four peer with each other.

The goal of having some of them peer with each other is to provide some
kind of "cross reference" among them in case of loss of connectivity to
the upstreams (e.g.: the entire datacenter loses connectivity due to an
outage), or if some of them loses one or more upstreams between the
periodic sanity checks I do on all my servers (I normally check the
status of the upstreams once every six months in each synchronization

I leave one out of this peering because I still like to have a fully
independent source.

Is this set-up sensible? Do you see any other good pro, or any really
bad con?

thanks a lot in advance

-- bronto

More information about the questions mailing list