Re: making Noel happy......
Noel Chiappa <jnc@ginger.lcs.mit.edu> Tue, 02 March 1993 21:48 UTC
Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa09149;
2 Mar 93 16:48 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa09145;
2 Mar 93 16:48 EST
Received: from thumper.bellcore.com by CNRI.Reston.VA.US id aa23734;
2 Mar 93 16:48 EST
Received: by thumper.bellcore.com (4.1/4.7)
id <AA08029> for ietf-archive@nri.reston.va.us; Tue, 2 Mar 93 16:46:44 EST
Received: from ginger.lcs.mit.edu by thumper.bellcore.com (4.1/4.7)
id <AA08007> for /usr/lib/sendmail -oi -fowner-pip X-pip;
Tue, 2 Mar 93 16:46:40 EST
Received: by ginger.lcs.mit.edu
id AA08104; Tue, 2 Mar 93 16:46:39 -0500
Date: Tue, 2 Mar 93 16:46:39 -0500
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Noel Chiappa <jnc@ginger.lcs.mit.edu>
Message-Id: <9303022146.AA08104@ginger.lcs.mit.edu>
To: jnc@ginger.lcs.mit.edu, pip@thumper.bellcore.com,
tsuchiya@thumper.bellcore.com
Subject: Re: making Noel happy......
Cc: jnc@ginger.lcs.mit.edu
Except for adding provider selection, the initial (near-term) Pip routing
architecture is plain old vanilla hierarchical addresses......
Umm, I can't conceive of a scalable RA which doesn't have hierarchical
addresses! If anybody knows of one, please let me know! :-)
The real question, of course, is what your guidelines are for creating the
addressing abstractions. It turns out addressing abstractions are really the
key to how well the routing will work (in terms of how optimal the routes
are); the actual algorithm used to calculate routes (i.e. DV, SPF, etc) is
basically irrelevant at that level.
Oh yeah, the fact that the top level routing algorithm in PIP is PV is very
significant (and one of the things I don't like :-).
Noel
- making Noel happy...... Paul Tsuchiya
- Re: making Noel happy...... Noel Chiappa
- Re: making Noel happy...... Paul Tsuchiya
- Re: making Noel happy...... Noel Chiappa