[apops] Re: anycast roots

[Randy Bush <randy@psg.com>]

> Well, either my probes don't pick up the jitter, or I'm guessing the
> naming convetion for j wrongly.
> 
> I see
> 
> jns1-hgtld.j.root-servers.net
> jns2-hgtld.j.root-servers.net
> jns3-hgtld.j.root-servers.net
> jns4-hgtld.j.root-servers.net
> jns5-hgtld.j.root-servers.net
> jns6-hgtld.j.root-servers.net

for purposes of tcp, that may be one flavor of jitter that alarmed
the verisign folk.  but see joe's pointers to
   http://www.isc.org/pubs/tn/isc-tn-2004-1.html
and
   http://www.isc.org/pubs/tn/isc-tn-2003-1.html

but i suggest we wait until we have data from widely distributed
edges before starting to form tentative conclusions.

randy