[ntp:questions] frequency adjusting only

Unruh unruh-spam at physics.ubc.ca
Mon Apr 28 18:14:41 UTC 2008


hal-usenet at ip-64-139-1-69.sjc.megapath.net (Hal Murray) writes:

>>Sorry it's actually difficult for me to precise this topic for
>>confidentiality reason of course and also because I'm an intern and thus
>>there for a short period. Hence I don't know exactly why myself. This is
>>just one of the requirement of my project.

>Keeping a handful of machines synchronized to within 10 microseconds
>might be possible.  It won't be easy.  It's unlikely to be the cheapest
>overall solution to some problem.

>I suggest you tell your boss that he has given you a hard to impossible
>problem.  You are not likely to get 10 uS by just poking a few magic
>parameters into your ntp.conf.  Your target is not a reasonable intern
>project.

Actually he only wanted 50us from what I recall. And it will depend on the
state of the switches, of the network cards, of the temp cycling of the
system, etc. chrony will quite easily achieve 50us. ntp is more difficult,
a) because it throws away 7/8 of the data (which could of course be
compensated by just querying more often) and b) throwing away the data one
it has once been plugged into the feedback loop. (Ie, the system is
essentially markovian) But even with ntp, I think 50us is possible. 

But all this speculation is a bit pointless. Do the experiment.
And if you really want to do it properly, get a gps or two and hook it up
to the machines to measure the actual deviation from true UTC of various of
the machines, including the server. Then you can subtract the two ( machine
-server) offsets to see what the relative accuracy is. 
In a situation like this, experiment is always better than theory.
(Note, I am a theoretical physicists)




>[That's just my opinion.  I could be way off.]

>-- 
>These are my opinions, not necessarily my employer's.  I hate spam.




More information about the questions mailing list