Mobility Implementation Question
Max Laier <[email protected]>
| Newsgroups | gmane.ietf.nsis |
|---|---|
| Organization | Universität Karlsruhe (TH) |
| Message-ID | <[email protected]> |
Hi, we are currently pondering about how to best implement mobility aware QoS-signaling. One of the first stumbling block I hit so far is that the mobility draft suggests that all signaling for the route-optimized case should be done using the CoA as the source/destination of the flow. i.e. signaling is done for the actual flow rather than for the logical one. This certainly makes sense, but any potential application that uses e.g. QoS-NSLP as a service will ask for a reservation for the *logical* flow as it is anaware of MobileIP (and should be). So the questions becomes where should the translation from logical to actual flow take place. I believe that mobility management should be taken care of within the NTLP layer as the NSLP should not have to be concerned with the actual message routing. Currently I have two possible sollutions to the problem which I'd like to hear feedback about: 1) Extend the GIST-to-NSLP API with a type of NetworkNotification that is able to convey Address-Binding-Information and let the NSLP figure out if this information is relevant to any of it's sessions. 2) Use a similar approach as the MRITS described in [1]. This would take place in the MN / CN respectively and translate the MRI according to the BindingCache (and/or the routing header). I feel that the second option might be easier to implement and could help to reduce the amount of duplication. i.e. this way only the NTLP has to interact with the mobility layer and the NSLP can be mostly oblivious of the fact that it is in fact mobile. The NTLP would obviously still have to send some kind of trigger in case of a BindingUpdate, but that could be a normal NetworkNotification for the HoA. Any comments greatly appreciated. P.S. How does the QoS-NSLP translate the packet classifier in a GaNAT environment. I'm not completely through the nattraversal-darf, yet and would appreciate any pointers. [1]http://tools.ietf.org/html/draft-pashalidis-nsis-gimps-nattraversal-05#section-5.3 -- /"\ Best regards, | [email protected] \ / Max Laier | ICQ #67774661 X http://pf4freebsd.love2party.net/ | mlaier@EFnet / \ ASCII Ribbon Campaign | Against HTML Mail and News _______________________________________________ nsis mailing list [email protected] https://www1.ietf.org/mailman/listinfo/nsis
signature.asc
(application/pgp-signature, 187 B)
-----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQBHYDZqXyyEoT62BG0RAok5AJ0UtdsmZtoVtsFXE1X+Sk7EJYG6wgCfS8xQ npon5pEpWbfMkj9cjVtD64g= =vwHJ -----END PGP SIGNATURE-----