Hey everyone,

I recently submitted a new draft on an evaluation and analysis of
Strategy D in Herrin's Taxonomy.  It can be found here:

http://www.cs.ucla.edu/~jenster/strategyDeval-Jen.txt


If all goes well, I will present my results in detail at the upcoming
IETF.  As always, questions and comments are welcomed.  I intend to
polish up the draft a lot, so actually qs and comments are particularly
welcomed.

Here is an abstract:
The recent preliminary recommendation of the RRG lays out a set of
strategies to scale routing.  Much of the RRG effort has focused on
strategies A-C, while there is rough consensus that strategies E-F
should not be recommended by the RRG.  However, little attention has
been given to strategy D, which recommends that routers reduce FIB sizes
by storing only the information they need, and nothing more. 
The major criticism of this approach is that "there's no evidence to
support a conclusion that they'd aggregate well deep in the core".
However, no deeper investigation has been done on this direction.  In
this draft, I present the results of a quantitative analysis of the
effectiveness of strategy D for FIB reduction for a major Tier1 ISP.
The contribution of this evaluation is to aid the RRG in assessing the
merits of strategy D, hopefully allowing us to take important steps
towards converging on a recommended solution.


Thanks guys,


Dan Jen

_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to