1. The Edge Geolocation Services EIDs have a 64bit H3-ID part of the location of the Vehicle being served
2. This is how we eliminate the need for an actual geospatial network. These are virtualized in the form of H3EIDs 3. The prefix of the service EID is the H3 location it serves, the suffix is the specific Geolocation Service: collection, parking .. 4. The vehicle EID is not related to location (H3) because its moving. It is however ephemeral to protect drivers geoprivacy 5. The only entity which knows the vehicle carrier assigned RLOC is its RTR. We can host those at the carrier (UPF) or bring your own 6. Both lisp-nexagon and lisp-fix do not change the protocol. But you do have to understand the map assisted overlay ucast/mcast 7. Maybe after more examples we can generalize how Compute First Networks use LISP to aggregate and reduce data on a need to know basis.. ..so that non lispers can follow a cookbook, but for now we need to do it example at a time i think. lisp-nexagon in publication lisp-fix-00 i think work group response was good, so by next ietf?? --szb Cell: +972.53.2470068 WhatsApp: +1.650.492.0794 > On Nov 12, 2021, at 14:25, Dino Farinacci <[email protected]> wrote: > > From my understanding of the use-case, I believe there are no changes to the > base protocol to support this use-case. Regarding EID assignment and > allocation, that is an admin/configuration function. > > H3-EIDs makes sense for assigning to nodes that provide services. But I > forget if you wanted to allocate an H3-EID to a moving vehicle. Because as > the EID moves, the H3 hexagon can change even the H3-ID is from resolution > 16. Unless you want to scope the mobility and when the vehicle moves, you > assign a new H3-EID. I hope you don't do thw latter or you lose the > EID-mobility features LISP has (or maybe don't need it in this case). > > Dino > _______________________________________________ > lisp mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/lisp _______________________________________________ lisp mailing list [email protected] https://www.ietf.org/mailman/listinfo/lisp
