[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


To: dnsop@cafax.se
From: bmanning@ISI.EDU
Date: Mon, 10 May 1999 08:19:03 -0700 (PDT)
In-Reply-To: <199905060622.PAA06280@necom830.hpcl.titech.ac.jp> from "Masataka Ohta" at May 06, 1999 03:22:26 PM
Posted-Date: Mon, 10 May 1999 08:19:03 -0700 (PDT)
Reply-To: dnsop@cafax.se
Sender: owner-dnsop@cafax.se
Subject: Re: Root Servers

> 
> Any comments?
> 
> 						Masataka Ohta
> ---
> 1. Motivation
> 
>    For the stability of the domestic Internet, it is critical for each
			      --------  What does this mean?
>    country that there are sufficiently many DNS root servers operating
>    at various places of the Internet in the country.

	You have postulated a lema w/o evidence. Why is this critical
	for Internet stability and what evidence is there that the current
	system is unstable.

>    However, the number of unicast IP addresses of root servers is
>    limited.  Thus, with the legacy operation of DNS, the number of root
>    servers in each country (including US) must be equal to the number of
					    ----
>    unicast IP addresses of root servers divided by the number of
>    countries (some weight may be given according to the number of
>    Internet hosts in each country).

	Why is this a MUST?

>    Given the current number of countries and IP addresses of root
>    servers, each country (again, including US) will be able to have 1/20
>    root servers, which definitely is not sufficiently many.

	I do not parse this well.  Are you proposing that the Internet
	topology is organized on political boundaries? Your segmentation
	seems based on country desigantion.

>    Similar operation described in this memo may be applicable to gTLD
>    servers but it is outside the scope of this memo.

	It would seem that this tactic, if useful would be best targeted
	at the gTLDs, since most of the dynamic occurs there.

--bill

Home | Date list | Subject list