[ntp:questions] Time sync on a windows domain

David J Taylor david-taylor at blueyonder.not-this-bit.nor-this-part.co.uk
Mon May 28 22:37:56 UTC 2007

Ryan Malayter wrote:
> On May 28, 1:29 am, "David J Taylor"
> <david-tay... at blueyonder.not-this- bit.nor-this-part.co.uk> wrote:
>> What is the URL for your graph?
> It was linked via the first URL I posted; here is a driect link:
> http://www.malayter.com/Win2003NTPperformance.gif
>> If you can automate the software deployment it may not be too bad.
> Yes, but considering Windows Time Service can be controlled with no
> scripting at all (via group policy), and is already installed, widely
> depolying the real NTP on Windows machines can be a tough sell unless
> the extra precision is required. I know, because I have had "deploy
> real NTP to Windows machines" on my own potential project list for
> several years. But I have never been able to justify the added
> complexity of ntpd on 250+ Windows servers and workstations to myself.
> Testing ntpd on such a diverse hardware and software base isn't
> trivial. w32time works well enough for most of our needs, is already
> installed, and can be easily configured from a central location
> without scripting.

Thanks for the URL - I had checked but not found it.

It's your call on the effort required versus the precision needed, of 
course.  Be aware that the perfomance of W32time on versions of Windows 
prior to Windows 2003 SP1 - particularly on Windows 2000 and XP - may not 
be as good as that in your graph.


More information about the questions mailing list