[ntp:questions] NTP vs RADclock?

unruh unruh at invalid.ca
Fri Jun 8 21:07:19 UTC 2012


On 2012-06-08, Rick Jones <rick.jones2 at hp.com> wrote:
> unruh <unruh at invalid.ca> wrote:
>> This is a rare thing (Ie once in a 100 packets has round trip delay
>> of 10ms, vs 10us usually.) I have no idea if this is interrupt
>> coalescing or what. Since our network switched to Gigabit switches
>> and some of my systems use gigabit cards, the timing behaviour has
>> become really bad.
>
>> It used to be that I got consistant 140us round trip delays ( with a
>> scatter of about 10-20us) Now, although that is still the minimum,
>> all have a scatter of about 50-100us, some have a bivalent round
>> trip of either around 150us but also around 300 us. Some have a
>> wildly varying ( a few packets around the 150, most around 300us,
>> with scatter up to 10ms.) round trip. It is a total mess. It really
>> seems like the GB switches are a disaster for timing. Of course
>> running them at 100 would not help. It is a problem with the switch
>> design I believe.
>
> Do you see the issue with classic 100Mbit/s NICs connected to your
> gigabit switch?  Or do you only see the issues with 1 GbE connected
> systems?  If so, then go ahead and disable the interrupt coalescing
> and see what that changes.
>
> Earlier I said that switches wouldn't coalesce - I should say
> "traditional" (my use of terminology) switches wouldn't - if the
> "switch" is really a Linux instance running bridging software then...

No, the switch is a Cisco 32 port 
>
>> I have no idea where this horror is taking place-- the NICs, the
>> switches, ....
>
> But you are sure it is gigabit the technology... :-)

I am sure that it started when we switched from 100Mb technology to Gb
technology, yes. Other places to look for the problem would be
appreciated. 

Look at 

http://www.theory.physics.ubc.ca/chrony2008-11-01/chrony.html

(pre Gb switch) 

vs
http://www.theory.physics.ubc.ca/chrony2012-04-28/chrony.html

esp at charge (100Mb card) or info (Intel Gb card)  

At that time string, a 100Mb carded computer was the server. 

the current page (which is a mess because of the changes I am making,)
http://www.theory.physics.ubc.ca/chrony/chrony.html
 is using info (1Gb card) as the ntp server. Note the vastly worse
behaviour of all of the systems in the round trip times (magenta) and
the offsets.

So yes, I think it is the Gb technology that is causing trouble. 




More information about the questions mailing list