RE: [EXTERNAL] Re: Extending a /64

["Manfredi (US), Albert E" <[e-mail]>]

-----Original Message-----
From: ipv6 <[e-mail]> On Behalf Of Brian E Carpenter

> On 16-Nov-20 11:29, Philip Homburg wrote:
>>
>> Assuming 16 billion people in the year 2100, then a 39 bit number for planes
has space for 34 planes for each person alive.

That's a little bit apples and oranges. You expect many, many more addressable devices, not to mention routers and subnets, on any large platform, than you would on a "person." I don't think it's logical to associate address space with number of people. Not directly, anyway. (Maybe eventually, when we all have hundreds on embedded medical sensors in our bodies, people and platforms will be more similar?)
 
>> Maybe that is overdoing it a bit. Or if we have that number of planes, maybe
a single prefix is a bit silly.

Amen.

> It's more than that. It's just wrong and doesn't match how Internet WAN routing works.
>
> We might as well admit that 3GPP/5G gets worldwide mobile connectivity right and the Internet doesn't.

Amen to that too. Although it doesn't go far enough, right? We're hauling along intranets now, not just one /64.

Also, I work with ships rather than planes. I can guarantee that a single ship would not be happy with one /60, if we are to assume that IIDs are 64 bits wide. Not even close, and this has next to nothing to do with how many souls are on board.

It's also not clear why a single platform should be fed off a single prefix. In fact, different onboard nets are administered by different organizations, so that the normal case would be, each organization assigns its own prefixen.

In practice, I'd say, we should not assume that excomm and interior communications would be managed the same way. In fact, they are not.

Bert