Hi.

Yesterday I gathered some statistics about (IPv4) servers in the pool.
I'm posting them here, since at least the offset part seems on topic.
Offset (absolute value) measured against my stratum 2 in Prague. I used
ntpdate, so it might not be that accurate.

Out of 2348 servers:

+-----------+--------------+-----+
|   offset  | # of servers |  %  |
+-----------+--------------+-----+
| <= 1 ms   |     1105     | 47% |
| <= 2 ms   |     1441     | 61% |
| <= 5 ms   |     1925     | 82% |
| <= 10 ms  |     2148     | 91% |
| <= 25 ms  |     2269     | 97% |
| <= 50 ms  |     2311     | 98% |
| <= 100 ms |     2330     | 99% |
+-----------+--------------+-----+

Out of 2348 servers:

+-----------+--------------+-----+
|  stratum  | # of servers |  %  |
+-----------+--------------+-----+
|     1     |      142     |  6% |
|     2     |     1592     | 68% |
|     3     |      578     | 25% |
|     4     |       26     |  1% |
+-----------+--------------+-----+

Out of 1044 servers:

+-----------+--------------+-----+
|     OS    | # of servers |  %  |
+-----------+--------------+-----+
|   Linux   |      658     | 63% |
|    BSD    |      213     | 20% |
|  Windows  |       30     |  3% |
|   SunOS   |       25     |  2% |
+-----------+--------------+-----+

More then half of the servers didn't return info about their OS, so the
last table might be off quite a bit.

Matej Snoha

On Thu, Aug 2, 2012 at 9:07 AM, [email protected] <[email protected]> wrote:

> On Wednesday, August 1. 2012, 16:43:16 Matej Snoha wrote:
> > Hi.
> >
> > A few months ago I did a sweep over servers listed at
> > http://support.ntp.org/bin/view/Servers/StratumOneTimeServers .
> >
> > Out of the 183 reachable servers,
> > 122 (67%)  had offset <= 5 ms
> > 156 (85%) had offset <= 10 ms
> > 174 (95%) had offset <= 20 ms
> > 9 had offset > 20 ms
> > 1 had offset 10 s
> > (measured against my stratum 2 in Prague).
> >
> > Your server currently shows as -24.8 ms.
> >
> > Also, the drop in your graph could just be a result of a network
> topology
> > change.
> > The offset could be a result of asymetric internet connection.
> >
> > Hope this helps,
> > Matej Snoha
> Thanks that's really interesting and helps to manage my expectations. I
> hope the calibration other mentioned will resolve the offset.
>
> Best regards,
>
> Klaus
> _______________________________________________
> pool mailing list
> [email protected]
> http://lists.ntp.org/listinfo/pool
>
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to