Re: [core] Lars Eggert's Discuss on draft-ietf-core-sid-22: (with DISCUSS and COMMENT)

[Carsten Bormann <[e-mail]>]

Hi Lars,

I’d like to pick up a few comments that address the size of SID ranges (see below).

First, while you are correct that the range 0..2^63-1 is huge,
let me also point out that some SIDs are “better" than others.
In particular, the deltas for SIDs below 65536 can always be encoded in 16 bits.
So the idea of the allocations below was to give the IETF modules some prime real estate.

PYANG isn’t quite up to the task yet, but I get on the order of <20_000 SIDs from the YANG modules that rsync from rsync.ietf.org::everything-ftp/yang and seem to be from published RFCs and Internet-Drafts (not all of which will make it), and some <12000 more from the IANA-published modules.
(We probably should sort these allocations into yearly bins and do some projection on when we’ll hit 60_000.)

Since there is an actual economic cost for all the specification creation work that ultimately leads to each SID being consumed, I think that 1_000_000 is quite comfortably above that 60_000.

We wanted to put up a bit of the good space (5536 SIDs) and some overflow (34464 SIDs) for the experimental space.

I think we can allocate some more space (out of 100000 to 999999) for "RFC required” now, so modules can explicitly opt for allocating 4-byte ranges in order to conserve 2-byte space (the actual hit will be much lower than 2 bytes per item because of the YANG-CBOR SID delta encoding).

Grüße, Carsten

> On 2023-10-26, at 15:18, Lars Eggert via Datatracker <[e-mail]> wrote:
> 
> ## Comments
> 
> ### Section 6.3.3, paragraph 3
> ```
>     The initial entry in this registry is allocated to IANA:
> 
>     +=============+=========+============+===================+==========+
>     | Entry Point | Size    | Allocation | Organization      | URL   |
>     |             |         |            | name              |       |
>     +=============+=========+============+===================+==========+
>     | 0           | 1000000 | Public     | IANA              | iana.org |
>     +-------------+---------+------------+-------------------+----------+
> ```
> I would have expected the initial allocation to IANA (and the regions
> defined within) to be MUCH larger, given the overall size of the
> namespace.
> 
> ### Section 6.4.2, paragraph 11
> ```
>     +=============+=========+==========================+
>     | Entry Point | Size    | IANA policy              |
>     +=============+=========+==========================+
>     | 0           | 1,000   | IESG Approval            |
>     +-------------+---------+--------------------------+
>     | 1,000       | 59,000  | RFC Required             |
>     +-------------+---------+--------------------------+
>     | 60,000      | 40,000  | Experimental/Private use |
>     +-------------+---------+--------------------------+
>     | 100,000     | 900,000 | Reserved                 |
>     +-------------+---------+--------------------------+
> ```
> These seem very small as well, given the overall size of the namespace.