Re: about QoS context

"James Kempf" <[email protected]> Tue, 20 Jul 2004 08:34:36 -0700
Newsgroups gmane.ietf.seamoby
Message-ID <[email protected]>
You are right, neither of these match well with context tranfer. Intsrv is
perhaps the closest, since the flow reservations for a moible undergoing
handover could be transferred from one access router to another, but Intsrv
is not widely deployed due to scalability questions.

There are two other ways QoS context transfer could be done:

1) Some Layer 2 protocols have specific QoS context associated with mobile
nodes. For example, in 802.11e, the HDF allows terminals to use TDM to
control channel access rather than CSMA. When a terminal moves from one AP
to another, the HDF reservation could be transfered.

2) A wireless service provider might set up their system with a kind of
firewall on the access router that would regulate the amount of traffic
allowed particular users. Some of the WLAN switches have this functionality
today. The user buys a certain amount of bandwidth, and if their traffic
exceeds that bandwidth, packets are dropped. Diffsrv treatments such as
Expedited Forwarding would be orthogonal, but related. A user might have to
buy EF also, and that would need to be recorded in context on the access
router. When the user moves from one access router to another, the bandwidth
allocation and EF treatment contexts would be moved to the new access
router. Alternatively, perhaps EF is afforded particular kinds of services,
and the user must subscribe to them. Whether or not the user subscribes (and
hence gets EF) would need to be transferred.

            jak




----- Original Message ----- 
From: <[email protected]>
To: <[email protected]>
Sent: Tuesday, July 20, 2004 12:09 AM
Subject: [Seamoby] about QoS context




Hi everybody,

I'm sorry for my intervention. I'm currently studying the context transfer
approach; I find that in most of the literature, QoS is mentioned as a
potential candidate service. As you know there are two well known different
approaches: Diffserv and Intserv, but neither of them seems to match the
idea
of the control transfer protocol. Diffserv doesn't keep any information
specifically related to the MN, on the other hand, Intserv does keep state
information but not only in the access routers but in the entire path.
I'd like to know if I'm missing or misunderstanding something. (or if I'm
totally lost)
I'll appreciate your comments.
Thanks a lot in advance for your time.

Christian G.

_______________________________________________
Seamoby mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/seamoby