[ntp:questions] .1 Microsecond Synchronization

David Woolley david at ex.djwhome.demon.co.uk.invalid
Thu Jun 4 21:26:27 UTC 2009


ScottyG wrote:

> To do this the order in which market data was received and trades 
> transmitted need to be maintained. I do know from their current log 

Remember that London and New York are, maybe 15 milli-lightseconds 
apart.  Even before allowing for cable routing and velocity factors, 
you don't need 100ns accuracy to prove non-causality.

> files that 1 ms is not fine enough for this and that on occasion .1 
> ms is not good enough. They currently are using a feature of the 
> processors that seems to return clock tick on the microprocessor 
> (Some assembly language instruction). They have an algorithm for 

RDTSC.  ntpd/Linux/FreeBSD kernels use this as one of their methods to 
interpolate hardware clock ticks.

> controlling the skew that occurs using this method. This seems to 
> meet there needs in a single server scenario but when going across 
> machines this will obviously not work.




More information about the questions mailing list