[ntp:questions] How to keep Linux server in Chicago and Mumbai in sync to within 5 microseconds

Chris Albertson albertson.chris at gmail.com
Wed Jan 12 19:00:43 UTC 2011


On Wed, Jan 12, 2011 at 9:31 AM, Rick Jones <rick.jones2 at hp.com> wrote:

> Is there one big CDMA timespace that encompases the planet, or are
> there really several discrete CDMA timespaces that are more "loosely"
> coupled?

The CDMA towers all have GPS receivers on them and use GPS to get UTC
and then rebroadcast that.  So yes there is one big "time space"
although there is loss of precision in the translation and processing.

It turns out that towers get replaced and upgraded and there is a
large cottage industry in China re-selling used cell tower equipment
on eBay.  That is currently the best way to get some really
sophisticated timing equipment at very low cost.  These guys are
selling rubidium oscillations for as low as $100   The equipment being
sold is typically working at the "tens of nanoseconds" level while the
CDMA signals are "tens of microseconds".  About a factor of 1000
difference.  The GPS receiver I'm using as an adjustment to account
for speed of light delay down the antenna feed cable, It is quite a
significant delay at the nanosecond level.  The cell towers internally
work at that level.  So thanks to GPS they are quite well sync's with
each other

A the level we are talking about here "same time" is a complex subject
. The speed of light is very slow at these time scales.   The OP has
asked for two computers to be synchronized at a level  smaller than
the speed of light delay between the two locations.  So if a single
observer where to watch both computers the distant computer would
appear "slow".  The definition of "simultaneous" depends of the
location of the observer.  Yes we actually are getting down to that
level here.

-- 
=====
Chris Albertson
Redondo Beach, California



More information about the questions mailing list