Re: [RRG] map-and-map (vs map-and-encaps) - further thoughts

[Brian Dickson <briand@ca.afilias.info>]

Brian E Carpenter wrote:
> > Each EID block gets mapped to a set of RLOC blocks, but each RLOC 
> > block only has one EID block mapped to it.
>
> Doesn't that mean there will be far more RLOC prefixes than EID
> prefixes?
It depends on your measure of "far more". :-)
But yes, it will be more.

Notably, the following should all be applicable:
- not every site needs an EID block, only multi-homed sites
- each RLOC block is assigned out of a PA block, and in the DFZ only 
should appear as the covering aggregate from the LIR
- the distribution of number-of-homes versus 
number-of-sites-with-that-number-of-homes, would likely be highly tilted 
towards the values 1, 2 and 3.
- EID blocks *may* number less than non-RLOC PA assignments, depending 
on the uptake of multihoming generally.

So, even in the extreme cases of some sites having lots of homes, I 
would expect the *average* RLOC:EID ratio is likely to be about 2 and 
change.

In short, I don't see it as being a big concern.

The main question will be, what are they typical block sizes for 
multihomed sites compared to non-multihomed sites?

That will impact the allocation assumptions being made by LIRs, if it 
differs significantly from the general case (non-multihomed).

However, I don't think there is any association between a sites desire 
to be multihomed, and the size of prefix it needs. The experience with 
PI assignments in IPv4 certainly suggests that small prefix multihomed 
is a very popular niche need.

Brian

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg