Hi all,
after Monday's presentation of PIM deployment options
(draft-ietf-multimob-pmipv6-source), there was the discussion on further
optimization options. In detail, the idea was raised to reach MNs not
via their (permanent) HNP advertisements at the LMA, but directly at
their current MAGs. The latter would require a dynamic unicast routing
protocol in the access network.
I promised to talk to Sri about this (underlying unicast) approach. As
expected, Sri emphasized that the PMIP WGs intentionally do *not*
consider this a working option. The reason is that node mobility
typically is more intense and faster than unicast routing dynamics.
Advertising MN's HNPs throughout the access network would cause route
pollution and convergence problems and quickly lead to inconsistencies.
For the unicast case, this is the equivalent of pushing multicast
mobility management into multicast routing, which we equally avoid.
Consequently, the deployment of PIM-SM in the access network (at MAGs)
cannot be better than presented: The (MRIB) route to a (non-local) MN
must be via its corresponding LMA, and we actually do not gain anything
over the Proxy approach when using PIM-SM at MAGs. On the contrary, as
we discussed on Monday, the PIM self-organization (phase transitions)
will produce significant overhead for mobile sources (with likely
convergence issues) why I would not opt for using PIM-SM other than in
Phase 1.
Cheers,
Thomas
--
Prof. Dr. Thomas C. Schmidt
° Hamburg University of Applied Sciences Berliner Tor 7 °
° Dept. Informatik, Internet Technologies Group 20099 Hamburg, Germany °
° http://www.haw-hamburg.de/inet Fon: +49-40-42875-8452 °
° http://www.informatik.haw-hamburg.de/~schmidt Fax: +49-40-42875-8409 °
_______________________________________________
multimob mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/multimob
