Worth Reading: Higher Levels of Address Aggregation

Every now and then someone tells me how much better the global Internet would be if only we were using recursive layers (RINA) and hierarchical addresses. I always answer “that’s a business problem, not a technical one, and you cannot solve business problems by throwing technology at them”, but of course that has never persuaded anyone who hasn’t been running a large-enough business for long enough.

Geoff Huston is doing a much better job in the March 2022 ISP Column – read the Higher Levels of Address Aggregation, and if you still need more technical details, there’s 30+ pages of RFC 4984.


  1. I've just read Geoff's column. Somehow it doesn't offer much of an insight except from stating the problems that people have already known: the routing and address explosion. He didn't qualify what he meant by achieve: "Somehow, we’ve achieved this without regional aggregation", so that doesn't mean much.

    Most of all, since John Day the originator of RINA -- IMO a very knowledgeable, down to earth and straightforward person who knows more about Networking than any man alive -- is still around, I'd say maybe someone can contact him directly and ask him this specific problem and how he would detail a solution as an engineering topic. He's already provided the part of the answer in chapter 8 of his book, Patterns in Network Architecture: A Return to Fundamentals, on why and a little bit of how, topologically-protected addressing can address provider's sharing issues -- now I'm not sure if Geoff has read it -- but surely people who seek a deeper answer rather than just mindlessly consume dogma can ask him questions if they're genuinely interested in Networking problems.

Add comment