[ntp:questions] long linear network, time accuracy, and ntp strata
unruh at wormhole.physics.ubc.ca
Tue Oct 19 12:37:10 UTC 2010
On 2010-10-18, Miernik, Jerzy (Jerzy) <jerzy.miernik at alcatel-lucent.com> wrote:
> I have a question about how to use ntp in a long linear network. Assume there are 50 nodes with the one most in the West having GPS, and every node running ntpd:
> node1 --- node2 --- node3 --- ... --- node50
What does "long linear network" mean? Does it really mean that all
network traffic between node 1 and node 50 must pass through eachof the
other machines ( eg because all have two network cards and the only
connection from node i to i-1 is via one card and from i to i+1 is via
the other network card on machine i?)
If so, you will have a horribly slow network no matter what you do.
Almost always direct connection is more accurate-- the outbound and
inbound network times are measured for each packet. Not assumed.
But as said by others, you want each of those nodes requesting time
directly from node 1. ( also note that a chain of 50 machines each
asking time from the previous one exhauses the possible stratum numbers
and it will not work. )
> Which approach to synchronization would be better in terms of the time accuracy in nodes:
> 1. configure 'server node1 ...' in nodes 2, 3, ..., 50; or
> 2. use manycast for automatic client / server self-organization?
> Approach 1 seems to me to lead to inferior accuracy due to increasing distance between a client in nodex and server in node1, where routing in nodes node2, node3, ..., node(x-1) would add to jitter. Yet we would have one node stratum1 and all others stratum2, it would seem...
> I would think approach 2 with manycast would end up in smaller offsets (if any) between clocks, but would lead to clocks of stratum 16, 17, ..., 50, and I am not sure how ntpd would react in node16, node17, ..., node50.
> Could an expert please offer a comment on which approach would be preferable?
More information about the questions