Re: [lisp] LISP EID Block Size

[Dino Farinacci <[e-mail]>]

> If consumers adopt LISP, we'll need 10,000,000,000 prefixes within the next ten years. That means each public PITR will have to announce 10,000,000,000 BGP-routes to itself if consumers use random PI-prefixes from the global unicast space. Multiply the number of public PITRs necessary for such a 

You are making it sound that LISP is requiring the need for 10 million prefixes. If there are 10 million new sites added to the Internet, no matter which technology you use to connect them, there will be 10 million more addressable prefixes. And if they are indeed PI prefixes, every core router will have to store them. With LISP the edges will store probably 6 orders of magnitude less. And if I'm off with 6, then say 5, but that is just a rounding error IMO.

> deployment with 10,000,000,000 BGP-routes each and you'll realize LISP will either break the internet and go into decline or just be a limited luxury for big companies. Using a dedicated PI-prefix which can handle 10,000,000,000 subnets will reduce the BGP-routes a public PITR has to announce to ONE. So using 24 or 32 bits is way to less for the expected growth.

The PITRs will not have to advertise those. If a large portion of those 10 million new sites are LISP sites then legacy sites may just want to talk to them. When that demand comes, the PITRs will on path of the tail legacy sites (the default path into the Internet), so those 10 million won't need to be advertised.

In fact, a PE router at an ISP would be a fine place to configure the PITR. But the choice is left the ISP to design how it would want to aggregate state, maybe 100 PE routers fan into 10 upstreams where that is where the PITRs go. But I think a PE router can handle 1 million routes today. They do it for the 1000s of VPNs they support. So this isn't a stretch.

Dino