To:
mohta@necom830.hpcl.titech.ac.jp (Masataka Ohta)
Cc:
hardie@equinix.com, mohta@necom830.hpcl.titech.ac.jp, liman@sunet.se, dnsop@cafax.se
From:
hardie@equinix.com
Date:
Thu, 8 Jul 1999 09:00:47 -0700 (PDT)
In-Reply-To:
<199907080221.LAA13889@necom830.hpcl.titech.ac.jp> from Masataka Ohta at "Jul 8, 99 11:21:27 am"
Reply-to:
hardie@equinix.com
Subject:
Re: Single Origin
Masataka, > You wrote: > > To those peers, the organization announces a route to the > network containing the shared-unicast address of the root name > server. > > which means multiple organizations, each of which having its own > AS numbers, are the multiple origins of the shared address. I believe I begin to see the source of your confusion. Look back in the abstract: These practices presume that a single entity remains administratively and operationally responsible for each of the distributed servers. This should probably be updated to say something like "distributed server groups", because I mean that the whole paper presumes a single organization handles all of the servers associated with a single shared unicast address. So ISP FOO might handle z.root-servers.net, with distributed copies at various places in its network topologies. I initially used the language "each of the distributed servers", because I presumed that the practice might be common for operators of root servers. Ted Hardie