RE: [netlmm] The term LMA

["Alper Yegin" <[e-mail]>]

> Alper Yegin wrote:
> > What's the definition of "local" and "global" anyways?!?
> 
> HEre's my personal definition:
> 
> "global" mobility is when one absolutely positively needs to use tunnels
> otherwise one threads the routes so much that they churn.
> 
> "local" mobility is when one can use route updates instead of tunnels,
> without provoking that route churn; precisely because the route updates
> are very "localized".
> 
> In that sense, tunnels are not needed in "local" mobility domains.

I think this is debatable. Like how much churn is significant churn to cause
a decision change, what is the definition of "very localized" (I think you
almost got a recursive definition there :-)

> 
> But that's just one oppinion, and is largely different than LTE and
> WiMax designs where Mobile IP tunnels are unnecessarily and
> overheadingly essential parts of local mobility domains.
> 
> > I agree with Raj.
> 
> Somehow yes, but one still needs to call that HA somehow an "enhanced"
> HA... because it's not just a plain vanilla rfc3775-HA.  The PMIPv6-HA
> should be able to work as a 3775-HA when the MN sends the BU (not the AR).
> 
> In my oppinion there is currently no other enhancement to 3775-HA that
> needed to call it differently than a 3775-HA.  Hence our problem...

A MIP4 HA may implement a number of extensions on top of the base MIP4 spec.
But we still call it an HA. Similarly, we call some other things a "router"
despite how many optional protocols they implement and differ from any other
"router." At a more detailed level, one needs to qualify an HA as the one
that implements RFC 3775 + RFCfoo + RFCbar....

Alper


 




> 
> Alex
> 
> >
> >> -----Original Message----- From: Chowdhury, Kuntal
> >> [mailto:[e-mail]] Sent: Thursday, May 10,
> >> 2007 7:07 PM To: Vijay Devarapalli; Basavaraj Patil; Giaretta
> >> Gerardo Cc: [e-mail] Subject: RE: [netlmm] The term LMA
> >>
> >> Another example is 3GPP SAE where mobility over S2a/b and S8
> >> interfaces will be based on PMIP6 protocol. These interfaces will
> >> be used for global mobility.
> >>
> >> So, there are plenty of examples where PMIP6 is applicable to
> >> global mobility as well. Hence, I would agree with Raj's proposal
> >> to rename LMA to HA.
> >>
> >> BR, Kuntal
> >>
> >>
> >>> -----Original Message----- From: Vijay Devarapalli
> >>> [mailto:[e-mail]] Sent: Thursday, May 10,
> >>> 2007 10:39 AM To: Basavaraj Patil; Giaretta Gerardo Cc:
> >>> [e-mail] Subject: RE: [netlmm] The term LMA
> >>>
> >>> Hello
> >>>
> >>> Why don't we take a specific example? I have always found the
> >>> terms global mobility and local mobility very vague when applying
> >>> to a particular SDO architecture.
> >>>
> >>> For example lets take WiMAX. I guess WiMAX forum intends to use
> >>> PMIPv6 between the CSN (where a home agent sits) and the ASN
> >>> (where the ASN GW sits). The CSN would have the LMA functionality
> >>> and the ASN gateway would have the MAG functionality. In this
> >>> case, it is absurd calling the home agent-link functionality in
> >>> the CSN, the LMA. :)
> >>>
> >>> So I do agree with Raj.
> >>>
> >>> Vijay
> >>>
> >>>> -----Original Message----- From: Basavaraj Patil
> >>>> [mailto:[e-mail]] Sent: Thursday, May 10, 2007
> >>>> 8:24 AM To: Giaretta Gerardo Cc: [e-mail] Subject: Re:
> >>>> [netlmm] The term LMA
> >>>>
> >>>>
> >>>> Hi Gerardo,
> >>>>
> >>>>
> >>>> On 5/10/07 10:11 AM, "ext Gerardo Giaretta"
> >>>> <[e-mail]> wrote:
> >>>>
> >>>>> Hi Raj,
> >>>>>
> >>>>> I disagree with this, mainly for two reasons.
> >>>>>
> >>>>> Based on the charter NETLMM is developing PMIP as a
> >>>> localized mobility
> >>>>> management protocol. Some other SDOs may use it as global
> >>>>> mobility management protocol, but we need to keep that in
> >>>>> mind: therefore,
> >> I
> >>>>> would like to have this reflected in the anchor name.
> >>>> Do you agree that the protocol as specified today can be used
> >>>> for global mobility as well? Why do we portray the work done as
> >>>> specifically being applicable only to local mobility? The
> >>>> protocol can be used to solve local mobility or global mobility
> >>>> and its better to acknowledge that than sticking our heads in
> >>>> the sand.
> >>>>
> >>>>> Secondly, I think a different name for the PMIP anchor (i.e.
> >>>>> LMA)
> >> is
> >>>>> very useful when you start working on MIP6/PMIP6 mixed
> >>>> scenario where
> >>>>> LMA may or may not be co-located with the MIP6 HA.
> >>>> Irrespective of co-location (which IMO is not an issue really
> >>>> since this is more of an implementation and deployment topic),
> >>>> I think what we
> >> have
> >>>> defined today as the LMA is an HA. So why not just refer to it
> >>>> as an HA. Do you believe the LMA is a specialized element being
> >>>> defined in Netlmm and the PMIP6 WG?
> >>>>
> >>>> I think it makes sense to get the scope and terminology right
> >>>> at this stage of the work in Netlmm.
> >>>>
> >>>> -Raj
> >>>>
> >>>>> Gerardo
> >>>>>
> >>>>> On 5/10/07, Basavaraj Patil <[e-mail]> wrote:
> >>>>>> Hello,
> >>>>>>
> >>>>>> The terminology that we have in the current PMIP6 I-D
> >>>>>> <draft-ietf-netlmm-proxymip6-00.txt> w.r.t LMA is not
> >>>>>> entirely accurate.
> >>>>>>
> >>>>>> LMA definition as per the I-D is: " Local Mobility Anchor
> >>>>>> is the home agent for the
> >>>> mobile node in the
> >>>>>> Proxy Mobile IPv6 domain.  It is the topological
> >>>> anchor point for
> >>>>>> the mobile node's home prefix and is the entity that
> >>>> manages the
> >>>>>> mobile node's reachability state.  It is important
> >>>> to understand
> >>>>>> that the LMA has the functional capabilities of a
> >>>> home agent as
> >>>>>> defined in Mobile IPv6 base specification [RFC-3775]
> >>>> and with the
> >>>>>> additional required capabilities for supporting
> >>>> Proxy Mobile IPv6
> >>>>>> as defined in this specification. "
> >>>>>>
> >>>>>> The protocol solution being discussed focuses on localized
> >>>>>> mobility. However this protocol also works equally well
> >>>> for the global
> >>>>>> mobility of the host. The term local mobility agent gives
> >>>>>> the perception that this entity is responsible only for
> >>>>>> local
> >>>> mobility and
> >>>>>> global mobility (if applicable) would be managed by a MIP6
> >>>> HA. While
> >>>>>> this deployment model is feasible, it is equally feasible
> >>>>>> to
> >> manage
> >>>>>> global mobility of the host with the same protocol (as
> >>>> long as certain
> >>>>>> assumptions about Proxy MIP capability in the access router
> >>>>>> that
> >> a
> >>>>>> host attaches to exist).
> >>>>>>
> >>>>>> IMO, we should change the LMA term to what it really is,
> >>>> an HA. If the
> >>>>>> HA is used in the context of localized mobility, it MAY be
> >>>> refered to
> >>>>>> as an LMA. But if it is providing global mobility, then it
> >>>> is just an
> >>>>>> HA. And since there really is no difference in the
> >>>>>> functionality
> >> of
> >>>>>> these entities, it would be better to just refer to it as
> >>>>>> HA. A PMIP6 HA = MIP6 HA + DSMIP6 capabilities + PMIP6
> >>>>>> enhancements
> >>>>>>
> >>>>>> -Raj
> >>>>>>
> >>>>>>
> >>>>>> _______________________________________________ netlmm
> >>>>>> mailing list [e-mail]
> >>>>>> https://www1.ietf.org/mailman/listinfo/netlmm
> >>>>>>
> >>>>
> >>>> _______________________________________________ netlmm mailing
> >>>> list [e-mail]
> >>>> https://www1.ietf.org/mailman/listinfo/netlmm
> >>>>
> >>> _______________________________________________ netlmm mailing
> >>> list [e-mail]
> >>> https://www1.ietf.org/mailman/listinfo/netlmm
> >> _______________________________________________ netlmm mailing list
> >>  [e-mail] https://www1.ietf.org/mailman/listinfo/netlmm
> >
> >
> > _______________________________________________ netlmm mailing list
> > [e-mail] https://www1.ietf.org/mailman/listinfo/netlmm
> >
> 
> 
> 
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________


_______________________________________________
netlmm mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/netlmm