Re: Draft on PMIP Localized Routing for traffic offloading
Peter McCann <[email protected]>
| Newsgroups | gmane.ietf.nemo |
|---|---|
| Message-ID | <5963DDF1F751474D8DEEFDCDBEE43AE716F52B15__12163.380199536$1375443049$gmane$org@dfweml512-mbx.china.huawei.com> |
Hi, Alex, I definitely think this technique could be used within a WiFi access network as the MN changes from one AP/AR to another. However, it is not clear whether the WiFi APs would be attached to the same I-BGP system as the cellular base stations. If they are, then great, I think it would work. However, if there are radically different backhaul topologies for WiFi and cellular it might be necessary to have two different I-BGP domains and use a global mobility protocol for preserving an address between cellular and Wifi (I think client-based MIP is good for this). I personally think the PMIP-based LR solutions are kind of ugly. -Pete Alexandru Petrescu wrote: > Hello, > > Seeing the presentation of Ryuji in intarea on the use of BGP to > realize mobility in the context of a cellular operator, draft- > matsushima-stateless-uplane-vepc-01, it makes wonder: > > - could this be used outside the context of a cellular operator? Or > within the cellular operator, but for WiFi offloading? > Because we have a draft for WiFi offloading or traffic control for > other reasosn, submitted as Localized Routing for PMIPv6: > > draft-boc-netext-lr-roext-05.txt > Alex