This paper is easily summarized, as they simply attempted to infer the nature of AS connections from their behavior. Nothing terribly difficult, and not rocket science. Each AS behaves differently based on who they are connected to, forwarding packets if the user has paid, and generally not forwarding if they have not. With a slightly more nuanced view, they show that it's not that hard to guess the relationships.
As to the paper itself, it went into far too much depth on the theoretical view of such a simple argument. That should have been left for an appendix, if anything. For discussion topics, I find little value. The previous paper gave a greater depth to the discussion of the reasoning for the AS setup. The inference of the AS status is not particularly interested to me either.
As a quicker read, this paper did give a good background on AS and BGP.
This paper is redundant and doesn't add much to the syllabus. I would remove it for an earlier paper describing the discussions made to push BGP to where it is. I'm a sucker for those sorts of things.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment