[ntp:questions] More Granularity in the US in the NTP Pool

Harlan Stenn stenn at ntp.org
Thu Sep 13 17:43:49 UTC 2007


Henk,

Please add thoughts like this to someplace like:

 http://support.ntp.org/Dev/NewNtpConfFormat#Allow_configuration_of_up_to_N_I

Past experience shows that posting thoughts like these to either the
newsgroup or a mailing list leads to good ideas that do not get implemented
because of the "needle in the haystack" principle.

H
--
>>> In article <fcbh2a$o9k$1 at prometheus.cs.uu.nl>, Henk Penning <henkp at cs.uu.nl> writes:

Henk> In <74CdnW83I6w0uHrbnZ2dnUVZ_obinZ2d at megapath.net>
Henk> hal-usenet at ip-64-139-1-69.sjc.megapath.net (Hal Murray) writes:
>>> Suppose a ntpd client can pick up 5 random pool servers, and
>>> periodically (say once a day) replace the 'worst' server by randomly
>>> picking a new one.
>>  I like that suggestion.  But, ignoring implementation details, how do
>> you decide which of two servers is "better"?

Henk>   I assumed that ntpd already ranks servers, so finding the worst
Henk> server should be feasible. Finding the 'best' replacement algo- rithm
Henk> requires further study, but I think something 'reasonable' can be used
Henk> to start supporting poolclients.

Henk>   Here is what I think would be beneficial to the pool project:
Henk>   -- A ntpd config option like
Henk>        poolclient pool.ntp.org 5 86400
Henk>      meaning :
Henk>        initially pick 5 pool-servers from pool.ntp.org ; every 86400
Henk> seconds (24 hours) pick a random pool-server to replace the least
Henk> useful pool-server currently in use.
Henk> ...




More information about the questions mailing list