Re: [Bier] Questions on draft-eckert-bier-te-arch-01

[Caitlin Bestler <[e-mail]>]

> [Tony saiz:] sigh, no, in most generic sense the same routing underlay carries all sub-domains in the normal case.
>
> I am tad baffled by your seemingly unique interpretation of the architecture @ this point in time. We had tons of people looking @ it by now and no'one tried to replace SI with sub-domains or felt that the concepts are the same.
>
> I utterly fail to see how the UI towards bier is being complicated given what you need, as I said, if your SDK towards BIER on your platform does not need  MT or sub-domains, just don't expose it and use 0,0 for everything. SI is hidden in the BIER layer itself. All you need really then is two (three) calls:
>
> Socketopt(my_BSL);   // set bitstring length if not default
> recv_on_bier(my_bfer_bit_nr);  // you can even skip the bit_nr given you're a leaf and frame arriving means it's for you as BFER
> send_to_bier(bfer_dst_bitmask);   // that will replicate/SI in the driver
>
> the rest is lower layer/driver stuff.
>
> all done, forget all the other stuff like overlays/underlays/subdomains/MTs/BSL translations/unicast tunnels and anything else in the architecture that seems superfluous for your needs ...


This is more of an issue of document clarity than a request for an 
architecture change.

My areas of expertise are at the application layer, host stacks, NICs 
and L2 switching. Viewed through those perspectives
there is virtually nothing in the architecture document that explains 
why a sub-domain exists and what the implications of
using a sub-domain vs. the use of an SI-subset are. Section 6.1's 
discussion of adolescences alludes to a few specifics, but
does not present an explanation.

As it stands right now the reader is left with "sub-domains and Subnet 
Identifiers are different because we said so." You can
implement with that, because the differences are well explained. But the 
lack of an inline or even cited rationale is bothersome
from the perspective of someone who wants specs to be minimal and 
self-justifying.

>
>