Re: How do we handle 4 billion CRH-FIB entries ¿?

[Brian E Carpenter <[e-mail]>]

Pablo,
On 22-May-20 23:00, Pablo Camarillo (pcamaril) wrote:
...
> 1.- How does it scale for an operator to configure on CLI more than 4 billion node-local CRH-FIB entries for the entire network?

It doesn't. It's clear that the words "configure" and "CLI" must be removed from an operator's vocabulary in such a network. Everything must configure itself, autonomically.

> 2.- Assuming that some day in the future you would like to advertise this in the IGP: how is it going to deal with 4 billion node-local entries?

There will never be such a problem. I'm not sure that RPL is the answer to all questions, but you can forget about traditional IGPs completely in such a network.

> Also, you said:
> 
>> Somewhere in the network, there is an entity constructs the CRH and the list that it contains. That entity needs access to another data structure, that includes a global view of each node’s CRH-FIB.

Indeed, that will not happen at such scale. There needs to be a distributed data structure; there will be no "global view" except at the policy intent level.

This is why some people have been thinking about autonomic networking for some years, and if it sounds like research, some aspects are indeed still research.

That doesn't argue against 32 bits; it just tells you that 32 bits is a strategic, not a tactical, choice. Rather like choosing 128 bits for IPv6, when 64 would have been "enough".

Regards,
    Brian