Section 2.1.1 states: "current Mobile IPv6 / NEMO specifications do not allow the use of multiple home agents by a mobile node simultaneously"
What exactly is the technical limitation prohibiting this simultaneous use? As I understand it, there is no problem running two simultaneous home addresses each of which can be bound to the same care-of address. Section 2.1.4 suggests that the Home Agent switch can be used to trigger a change of HA when there are no ongoing sessions that need address continuity. How is the HA supposed to determine that this condition holds? The knowledge about ongoing sessions is inside the MN. Section 2.1.5 talks about flow mobility but the relationship to DMM is not clear. Section 2.2: "Service Data Gateway (SGW)" I think this should be "Serving Data Gateway". Also, you should talk about the tunnel between eNB and SGW. DMM will have the most impact if we can replace those tunneling protocols too. Section 2.2.1: "On the other hand, as soon as the mobile node moves, the resulting path starts to deviate from the optimal one." You should note that this situation may be acceptable as long as the session is short-lived, and a new address is used for new sessions. Section 2.2.2 compares X2 to the PMIP local routing. I don't think this is a good comparison. X2 is used to forward traffic for a single mobile node between a previous and new eNB. In contrast, the local routing solutions are about routing traffic between two different MNs. Section 2.2.3, you state that SIPTO has no mechanism for seamless session continuity if you move out of the coverage area of a local PGW. But really, there is no technical reason why you couldn't open up a GTP tunnel to the old PGW from the new network. Note also there is a double negative in the last sentence of Section 2.2.3: "cannot not survive". The draft seems to consider existing protocols only in isolation; however, I think there is something to be gained by considering combinations of the existing protocols. For example, a network-based mobility solution could be used for localized mobility management within a given domain (where the sub-optimality would be somewhat limited) and then a client-based mobility solution could be used for handling mobility outside of this domain. If a network-based mobility scheme is used to handle the client-based Care-of Address, then most of the disadvantages of the client-based mobility scheme (frequent updates over-the-air to the HA) disappear. -- Peter J. McCann Huawei Technologies (USA) [email protected] +1 908 541 3563 Rm. C-0105, 400 Crossings Blvd. (2nd floor), Bridgewater, NJ 08807-2863 USA _______________________________________________ dmm mailing list [email protected] https://www.ietf.org/mailman/listinfo/dmm
