CO ISOC > Blog > Input Needed > Your Input Needed: Root Server System Scaling Study

Your Input Needed: Root Server System Scaling Study

We noticed an email sent to us from Bill Manning via NANOG today requesting input on DNS scaling and the impact on the root of adding IPv6 records, new TLDs (Top Level Domains), security records and other impending changes to the system.  Mr. Mannning explains it quite well:

With resolution 2009-02-03-04, the ICANN Board asked the Root
Server System Advisory Committee (RSSAC), the Security and
Stability Advisory Committee (SSAC), and the ICANN staff to study
the potential impact on the root zone stability that might arise
when IPv6 address records, IDN top level names, other new TLDs,
and new records to support DNS security are added to the root zone.

The Board has expressed interest in hearing of the impact of the
distinct changes, but also their aggregate effect on root zone
operations. The Board also asks that the study address the
technical and operational concerns regarding expanding the DNS
root zone that have been expressed on this topic.

A study team has been assembled and is composed of the following:

Jaap Akkerhuis
Lyman Chapin  – Chair
Patrik FC$ltstrC6m
Glenn Kowack
Lars-Johan Liman
Bill Manning

The goal of the study is to construct a complete, definitive, and
authoritative model of the root server system (including all of
its provisioning and query components) that shows how the
different parts are related, and how changing the value of a
variable in one part would affect each of the other parts. It will
be as quantitative a model as possible given the relatively short
time frame of the study. The idea is not to try to identify all of
the bad things that might happen if (for example) the number of
TLDs in the root zone were to increase by several orders of
magnitude during the same time period in which new RRs were added
for DNSSEC (b^@^how many entries in the root zone is too many?b^@^]), but
to construct a model that demonstrates what the effects would be
throughout the system of changing the value of one or more
variables either individually or in combination (e.g., the rate at
which new TLDs are added to the root, or the frequency of key
rollover events in a signed root). This model will provide an
unbiased factual and analyticalb^@^Trather than anecdotalb^@^Tbasis for
recognizing and evaluating concerns about root server system
security and stability as the root grows. As a tool for policy
makers, it will show the expected impact on the root server system
of policy decisions that would change various root zone parameters.

The study team has a couple of methods for collecting inputs and
encourages interested parties to provide the study team with their

There is a public mailing list:
There is a comment site:

If there are others whom you think would be interested in this work or
have opinions/ideas, please pass along this note.


Please take his invitation seriously and comment if you have one!

Alternatively, if you are a member of CO ISOC or you work/reside in/near Colorado and have some input but do not want to contribute directly; please feel free to send the information to us.  We will agregate all comments recieved and provide the study team with the resulting report.  You can send these (and any other) comments to our general mailing list: coisoc at coisoc dot org

June 13th, 2009 by | Posted in Input Needed | Tags: , , , , , , , , , , | 1 Comment

One Response to Your Input Needed: Root Server System Scaling Study

  1. Randy Bush says:

    flooding the root with icann’s greed is simply stupid and insane.

Leave a Reply