martin at jonesey.net
Mon Jul 20 16:25:35 UTC 2015
As another data point, I have been running a stratum 2 pool server in
Digital Ocean's KVM based cloud for a couple of months now and it performs
well. It stays within 1 ms of the stratum 1 servers it follows. That is
more than good enough for the ntp pool.
I now have it registered in the pool as a 1Gbps server but ntp traffic is
only using about 15% of my monthly transfer allowance of 1TB.
I see no good reason why a service provider should differentiate between
ntp and any other type of IP traffic while there is an agreed allowance
covered by the monthly fee.
On 20 Jul 2015 16:40, "Michael Rathbun" <time at rabendary.net> wrote:
> On Mon, 20 Jul 2015 01:49:37 -0700, Ask Bjørn Hansen <ask at ntppool.org>
> >But for most actually deployed servers isn’t that almost as true when
> you’re not virtualized? Modern kernels and CPUs do all sorts of tricks to
> save power and divide resources. Most servers aren’t configured to run at
> full speed just in case it makes the time keeping better.
> FWIW, my FreeBSD server instance is running in a VirtualBox VM on a Windows
> 7 platform. Before I had to consolidate some hardware, it was on its own
> FreeBSD dedicated hardware. The Windows environment also runs its own NTP
> server (Meinberg).
> Oddly, from the standpoint of measured offsets, the best performer
> according to the monitor is the VM instance; the worst is the Windows
> Sometimes half-ass is exactly the right amount of ass.
> -- Wonderella
> pool mailing list
> pool at lists.ntp.org
More information about the pool