[ntp:questions] Garmin 18 LVC: whether to fudge

Unruh unruh-spam at physics.ubc.ca
Sun Feb 15 18:04:36 UTC 2009


Hans =?iso-8859-1?Q?J=F8rgen?= Jakobsen <hjj at wheel.dk> writes:

>On Sun, 15 Feb 2009 04:07:33 GMT, Danny Mayer wrote:
>> Richard B. Gilbert wrote:
>>> Chris Adams wrote:
>>>> Once upon a time, Richard B. Gilbert <rgilbert88 at comcast.net> said:
>>>>> My bet would be that there is an asymmetry in your ADSL link!   If I'm 
>>>>> not mistaken, the "A" in ADSL stands for asymmetric!
>>>> The asymmetry in ADSL is in bandwidth, not path or latency.  More
>>>> frequency space is used for downstream (ISP->end user) communication
>>>> than for upstream, but both travel the same path.
>>>>
>>>> There may be asymmetric routing going on (very often the case when you
>>>> talk to a host across the Internet, especially if either your ISP or the
>>>> remote host's ISP are multihomed), but it is highly unlikely it is
>>>> happening between the end user and his own ISP.
>>>>
>>> 
>>> The difference in bandwidth would mean that a packet of N bytes would 
>>> take more time in one direction than the other!  I don't know if that 
>>> would be sufficient to account for the observed behavior but I can't 
>>> think of another explanation.
>>
>> Since a regular ntp packet is just 48 bytes without extensions, the idea
>> that ADSL would make a difference is rather unlikely. The asymmetry is
>> related to the size of the packets but it affects TCP rather than UDP.

>There ARE asymetric delays even for ntp packets.
>For my 14M/1.5M VDSL line I see an offset of 7-800 microseconds:
> ntpq -p
>     remote           refid      st t when poll reach   delay   offset  jitter
>==============================================================================
>-ns.tele.dk      .GPS.            1 u   48   64  377   22.450    0.891   0.096
>-tix.ns.tele.dk  .GPS.            1 u    7   64  377   22.338    0.698   0.183
>-puma.tele.dk    .GPS.            1 u   19   64  377   27.048    0.784   0.145
>+GPS_HP(0)       .GPS.            0 l   10   16  377    0.000   -0.776   6.230
>oPPS(0)          .1PPS.           0 l   10   16  377    0.000    0.012   0.015
>+GPS_ONCORE(0)   .GPS.            0 l    8   16  377    0.000    0.035   0.006

Something is really weird here. .776ms from a GPS refclock is horrible.
That is about a factor of 300 worse than you would expect. In fact all of
your GPS refclocks show bad time What are you doing to your clocks?


>20.6 ms of the roundtrip time are spend on the DSL line.

How do you know it is exactly 20.6ms?


>Some interleaving has been configured. It's better to have delay than dropped
>packets.
>The VDSL line runs at a distance from the CO where ADSL have better performance,
>at least for download.
>/hjj




More information about the questions mailing list