Re: Class B vs. lots of class C's

Dan Long <long@nic.near.net> Tue, 09 February 1993 00:04 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa20317; 8 Feb 93 19:04 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa20313; 8 Feb 93 19:04 EST
Received: from merit.edu by CNRI.Reston.VA.US id aa26529; 8 Feb 93 19:04 EST
Return-Path: <long@nic.near.net>
Received: from nic.near.net by merit.edu (5.65/1123-1.0) id AA03329; Mon, 8 Feb 93 18:59:42 -0500
Message-Id: <9302082359.AA03329@merit.edu>
To: Vikas Aggarwal <aggarwal@r2d2.jvnc.net>
Cc: njm@merit.edu
Subject: Re: Class B vs. lots of class C's
In-Reply-To: Your message of Mon, 08 Feb 93 18:44:05 -0500. <9302082344.AA04571@r2d2.jvnc.net>
Date: Mon, 08 Feb 1993 18:59:38 -0500
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Dan Long <long@nic.near.net>

	From: Vikas Aggarwal <aggarwal@r2d2.jvnc.net>
	Subject: Class B vs. lots of class C's
	Date: Mon, 8 Feb 93 18:44:05 -0500

	I remember this discussion a while ago, but don't remember the
	outcome...

	A site here has to make a decision between 3 class B's or 50 class C
	network addresses. Which shall the site apply for ? (routing table
	size vs. lack of class B's).

This was discussed at length at the MERIT seminar in Boulder.  The consensus
was that assigning a block of 50 (up to 64 or perhaps even 128) C's was
preferable IF they didn't anticipate needing to use most of them between now
and when Supernetting happens in mid '93.  If they do, then applying for B(s)
was the better move short term.  After mid '93, I assume that the answer would
almost always be to go with C's.

It was recommended that providers go get large blocks of C's but they hold off
doing that until the NIC is instructed about giving blocks out of the proper
part of the remaining class C space.  Not sure when that will happen.  Soon, I
hope.

The summary of the seminar's discussions was being compiled by Elise Gerich and
she or others can jog my memory if I got it wrong.

Dan