[ntp:questions] improving ntpd performance on Windows

Dave Hart davehart at gmail.com
Thu Mar 5 18:50:22 UTC 2009


On Mar 5, 1:59 pm, Martin Burnicki <martin.burni... at meinberg.de>
wrote:
> >> Unfortunately under W2k the new code does not seem to work as good during
> >> the first minutes after startup
>
> Dave Hart wrote:
> > 20000 seconds is 5.5 hours, the length of the roller-coaster ride it
> > took as I read it.  I see inflection points which I'm guessing are 512
> > seconds apart where the counter frequency estimate was changed.  I'd
> > like to see the ntp.log or event log entries for the first few hours
> > of that w2k run, if available.  I'm particularly interested in the
> > (frequent) messages showing the ctr frequency observations and running
> > average.
>
> I'll send you the event log file and the raw loopstats via private mail.

Thanks, I received them.  The counter frequency tuning was reasonably
well behaved, using 90 ppm nearly every time, with a few instances of
111 peppered here and there.  That leaves me unable to explain the
roller coaster effect and days-long frequency correction.  I'm far too
impatient and short on test subjects to let ntpd hunt for that long.

I'm currently testing averaging 6 512-second samples and it seems to
perform better on my test machine with a 3.5 MHz counter frequency,
which has been the most challenging in general for this frequency
correction exercise.

Cheers,
Dave Hart




More information about the questions mailing list