Re: Post processing of NTP data...

vschmidt@xxxxxxxxxxxx (Val Schmidt) · Tue, 27 Sep 2005 14:48:11 GMT

I've embedded some further questions below.
By the way, thanks for the insight.
On Sep 27, 2005, at 5:05 AM, Brad Knowles wrote:
At 5:08 PM -0400 2005-09-26, Val Schmidt wrote:
 I want to log several things with time stamps on the order of ~ . 
1ms -

 maybe less.

    Most modern OSes don't allow you to directly achieve better  
than 10-20ms accuracy at the level of an individual event.  Some  
real-time operating systems (RTOSes) may allow you to achieve finer  
resolution at that level, but I don't know if any of them are going  
to let you get down to the level you want.

Can you help me understand why?
Setting aside for a moment the issue of ntp itself and the roping in  
of the system clock, is the crux of the problem that the processing  
of interrupts is too unpredictable?
In my simplistic way of thinking of things, assuming the clock is  
sufficiently accurate, it seems to me the ability to time-stamp an  
event appropriately should be directly proportional to the time it  
takes to recognize that the event has occurred and retrieve a time  
stamp. When an event occurs and an interrupt is registered, the speed  
of the processor or more accurately, the speed in which it can be  
preempted seems to me to be the limiting factor. Is this the correct  
way to think about things?
Then returning to the system clock, would making data logging system  
itself a stratum 1 server provide sufficient stability for ~.1ms  
stable long-term time-keeping? It occurs to me that even a stratum 1  
server is at the mercy of interrupt processing when the reference  
clock registers a new signal of some kind. I don't suppose ntp does  
any kind of benchmarking of the system on which it's running in some  
attempt to determine and correct for the interrupt latency of the OS.
-Val
------------------------------------------------------
Val Schmidt
CCOM/JHC
University of New Hampshire
Chase Ocean Engineering Lab
24 Colovos Road
Durham, NH 03824
e: vschmidt [AT] ccom.unh.edu
m: 614.286.3726

_______________________________________________
questions mailing list
questions@xxxxxxxxxxxxxxxxx
https://lists.ntp.isc.org/mailman/listinfo/questions
.

Re: Post processing of NTP data...

Relevant Pages