Transcript MIP6: Food for thought

MIP6: Food for thought
B. Patil
Mobopts @ IETF68
March 22, 07
Background
• Mobile IPv6 (RFC 3775) published in
June 04
– Not exactly NKOB anymore
– Some amount of experience with
implementing the protocol has been
gained
– Deployment experiences would benefit
Route Optimization
• MIP6 RO is one of the more salient features
• When does RO get triggered?
– Type of traffic?
– Duration of session?
– Other criteria?
• Guidelines for when to trigger RO and
mechanisms for such would be useful and
help deployments
Signaling Compression
• The MIP6 signaling messages with various
extensions are starting to get bulky
• E2E compression, i.e between MN and HA
would possibly alleviate this issue
• Analysis of compression schemes and
applicability in various scenarios is worth
studying to improve MIP6 efficiency
Multicast
• Multicast in the Mobile IPv6 context has
not been completely addressed
• Acknowledged shortcoming in the base
specification (RFC3775)
• Mcast analysis and operation in MIP6
networks with/without RO is very much
needed
Handover enhancements
• With multi-interface devices/hosts becoming
more prevalent, the need for seamless
handovers at L3 is even more relevant
• Solutions that abstract L2 triggers and enable
MIP6 handovers between access types
needs to progress faster
– Simulations and implementation results needed
• Ongoing work based on 802.21 is a good
start
– Need to expand this to cover generic access
types
Host and network based mobility
• Co-existence of host and network based
mobility approaches in networks would help
– Transitive trust and authorization models when
the host is not involved need to be better
understood
– Scenarios and issues related to mobility between
networks that manage host mobility vs the host
managing its own mobility needed
– Impacts to applications/IP sessions in
transitioning between mobility types