[Nottingham] "ntpq -p" displayed flags

Martin martin at ml1.co.uk
Fri Jan 27 21:08:07 UTC 2012

On 27/01/12 20:58, Camilo Mesias wrote:
> Hi,
> On Fri, Jan 27, 2012 at 5:30 PM, Martin <martin at ml1.co.uk> wrote:
>> The claim for PTP is in microseconds. I just wonder if NTP achieves
>> similar results for a cluster synchronised across a LAN?
> I think PTP can out-perform NTP in the case where it makes use of
> hardware timestamping - it avoids the error introduced by device
> drivers and the OS. This is beyond what NTP was designed to do. PTP

Is that hardware timestamping in the NIC or network switches and
inserted directly into the data packets?

As in... Special hardware is required?

> should enable ethernet to be used for telecoms applications that
> require strict synchronisation, which would otherwise be provided over
> dedicated (expensive) lines...
> I've seen some lab hardware that demonstrates this but I'm not working
> on it at the moment :)

Mmmm... This is all building up to an interesting talk ;-)


The "Clock of the Long Now" must rate as one of the most expensive and
long lived follies ever but still is a "must see" at both the Science
Museum London and at the sites over in America.

The Chronophage is rather good fun also :-)


Martin Lomas
martin at ml1.co.uk

More information about the Nottingham mailing list