[ntp:questions] Re: Philosophical question about strata

Pete Stephenson pete+usenet at heypete.com
Fri Oct 7 15:24:35 UTC 2005

In article <1128694706.317164.261890 at o13g2000cwo.googlegroups.com>,
 "Lee Sailer" <lee.sailer at direcway.com> wrote:

> I had this thought about strata and I am hoping someone might have some
> ideas to contribute.


> Suppose you have two ntp clients that get their time from multiple
> strata 1 servers.  They should be very accurate statum 2 servers as a
> result.


> Now later, you are configuring a new client, and for some reason you
> have a choice.  You can use one of the stratum 1 servers, or the
> stratum 2 server.  (I know, make another choice.  This is a theoretical
> question, ok?)
> It seems to me that the right choice is the stratum 2 server.  Am I
> wrong?

That is correct. Assuming they're on the same network, it's more likely 
that the third client could get more accurate, less-jittery time from 
the two Stratum 2 servers you have running there already. Additionally, 
if there were to be a network outage and all the servers lost 
connectivity to the internet, the two Stratum 2 servers would continue 
to keep time according to their internal clocks and each other (assuming 
they're peers of each other). All the clients, who are querying the two 
Stratum 2 servers, would remain syncronized to those servers.

While the servers will drift over time and not reflect the true, 
accurate time, all the systems on your network will remain in sync with 
each other. Once the network comes back online, the Stratum 2 servers 
will then adjust themselves (and their clients) back to "real" time.

> However, if you configure this new client so that it uses one stratum 1
> the stratum 2, I think the algorithm will choose the stratum 1.  Again,
> am I wrong?

It depends...assuming the Stratum 2 server has a more stable network 
path between the client than the client has with the Stratum 1 server, 
the client may prefer the Stratum 2 due to more favorable network 

> Basically, my thought is that it is possible for a stratum 2 source to
> be better than a stratum 1 source, and likewise for a stratum n source
> to be better than a stratum m source, where n > m.

To the best of my knowledge, it is not possibly for a higher-stratum 
(i.e. 2, 3, 4, etc.) to be more accurate with the actual time than a 
lower-stratum server. However, to the client, the higher-stratum server, 
if located on a more stable network segment, may provide less jittery 
time to the client.

Also, there's the whole "politeness" thing; many Stratum 1 servers are 
extremely swamped with traffic, leading to network congestion. A fair 
amount restrict access in a multi-computer environment to only the 
Stratum 2 local timeservers for that environment, and require that all 
other computers on the network query the local machines. It helps keep 
network traffic and jitter down, and helps with providing more accurate 

If we're talking theory though, I'd really like a rubidium or cesium 
clock here in my basement. :)

Pete Stephenson

More information about the questions mailing list