[ntp:questions] .1 Microsecond Synchronization

Joseph Gwinn joegwinn at comcast.net
Sat Jun 6 21:25:42 UTC 2009


In article <2PmdnYgdgrSWZbrXnZ2dnUVZ_uidnZ2d at giganews.com>,
 ScottyG <scottg at pepex.net> wrote:

> Hello.
> 
> The company I am working for needs to be able to record timestamps in a 
> trading 
> system logs down to a .1 microsecond accuracy.
> 
> We will have servers located in London, New York and Chicago. There will be a 
> dedicated resilient link between London and New York.
> 
> Searches on the web have made claims that NTP can achieve this accuracy. 
> Unfortunately the sales rep for the NTP server we looked at told me that the 
> best I 
> could expect is 2-5 ms synchronization across servers.
> 
> Has anyone had any experience doing this? Can anyone suggest how to achieve 
> this accuracy?
> 
> We do have some budget but this but if I need to spend a whole lot on this I 
> need 
> to get in front of my management with the reasons. 
> 
> Thank you in advance for any help or suggestions you can give me.

It's not obvious that what the company wants to do is even physically 
possible.  

As many others have mentioned, getting 100 nanosecond synch everywhere 
in a network 10,000 km in diameter isn't easy, and isn't possible 
without expensive specialized hardware.  

However, even with the special hardware, there are very deep problems 
with the whole idea.  

The classic reference is:

"Time, clocks, and the ordering of events in a distributed system", 
Leslie Lamport, Communications of the ACM 21,7 (July 1978) pages 558-565.

Paper 26 in 
<http://research.microsoft.com/en-us/um/people/lamport/pubs/pubs.html>.


Joe Gwinn




More information about the questions mailing list