[Softwires] MAP and 4rd - Feature comparison table

[Rémi Després <[e-mail]>]

Le 2012-02-02 à 18:35, Alain Durand a écrit :

> Please, Remi, do build such a table! That would be very useful.

Here is what I got:

   +----+--------------------------------------+-----+-----+-----+-----+
   |    | Feature                              | MAP | 4rd | MAP | 4rd |
   |    |                                      |  -T |  -H |  -E |  -E |
   +----+--------------------------------------+-----+-----+-----+-----+
   |  1 | Full Transparency to IPv4 DF bit     |  N  |  Y  |  Y  |  Y  |
   |    |                                      |     |     |     |     |
   |  2 | ISP can impose a Tunnel traffic      |  N  |  Y  |  Y  |  Y  |
   |    | class                                |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  3 | IPv6 port-based ACLs work for IPv4   |  Y  |  Y  |  N  |  N  |
   |    | packets                              |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  4 | IPv6 web caches work for IPv4        |  Y  |  Y  |  N  |  N  |
   |    | packets                              |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  5 | No constraint on subnet prefixes in  |  N  |  Y  |  N  |  Y  |
   |    | CE sites (V-octet format)            |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  6 | Number of excluded ports is flexible |  Y  |  N  |  Y  |  N  |
   |    | (GMA algorithm, 2 parameters)        |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  7 | Migration from DS routing to         |  N  |  Y  |  N  |  Y  |
   |    | IPv6-only is possible without        |     |     |     |     |
   |    | changing CE addresses and/or         |     |     |     |     |
   |    | prefixes (DMR may apply to CEs)      |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  8 | Automatic support of all protocols   |  N  |  Y  |  N  |  Y  |
   |    | having ports at their usual place    |     |     |     |     |
   |    | and a TCP-like checksum anywhere     |     |     |     |     |
   |    | (checksum neutrality)                |     |     |     |     |
   |    |                                      |     |     |     |     |
   |  9 | IPv4-options supported               |  N  |  N  |  Y  |  Y  |
   |    |                                      |     |     |     | (1) |
   |    |                                      |     |     |     |     |
   | 10 | Datagram reassembly avoided in BRs   |  N  |  Y  |  N  |  Y  |
   |    |                                      | (2) |     | (2) |     |
   |    |                                      |     |     |     |     |
   | 11 | Packet IDs from shared-address CEs   |  N  |  Y  |  Y  |  Y  |
   |    | cannot be confused in destinations   | (2) |     |     |     |
   +----+--------------------------------------+-----+-----+-----+-----+

   (1) Not in draft yet, but decided to be added (trivial to do)
   (2) Not in draft, but could be added if decided



> On Feb 2, 2012, at 10:56 AM, "Rémi Després" <[e-mail]> wrote:
> ...
>> A 4 columns table would be ideal. Also, It could have a sign identifying points that are N in current drafts, but  could easily become Y if the final consensus is that they are worth the additional complexity.
>> I can work on it if you are interested.