ntpd: program structure

[Eric S. Raymond]

Hal Murray <hmurray at megapathdsl.net>:
> 
> Eric said:
> > I concur.  I would need to see actual measurements before I've convinced that
> > ntpd with even a thosand client connections is a performance-degrading load. 
> 
> I think there are two cases: popular public servers like NIST and everybody 
> else.
> 
> I agree that the load on most servers is not a problem.  The load on the NIST 
> servers is huge.  I don't know if the limit is network or CPU.  When the 
> government screwup gets fixed, I'll see if I can get some data.
> 
> If you think we should retain the lockclock code, then we should pay attention 
> to performance.
> 
> I think a small server-only mode could be a good fit for both cases.

I'm still going to take convincing.  Like, with actual load numbers from
one of those big servers.

We've just seen that a 2.8Ghz system with old, slow memory only goes
to 3% load handling 18K clients.  Load is going to rise by number of
clients probably a bit superlinearly, but this still suggests that
even 180K clients would be quite far from saturating even that
hardware, let alone a modern system.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

My work is funded by the Internet Civil Engineering Institute: https://icei.org
Please visit their site and donate: the civilization you save might be your own.