Robert Elz wrote:
>
> Date: Wed, 26 Dec 2001 13:57:08 -0500 (EST)
> From: Scott Bradner <[EMAIL PROTECTED]>
> Message-ID: <[EMAIL PROTECTED]>
>
> | the whole point of class-based QoS is to not have to deal at the
> | port and protocol level
>
> But the information about the QoS desired has to come from somewhere
> doesn't it? We assume (as you say) that it doesn't come from the
> port/protocol. It can't come from the traffic class (DSCP), as that's
> explicitly been designated to be mutable - when it reaches any random
> router it could contain any value at all - not in any way related to
> the value it had when the packet left the originator.
>
> So, in your model, from where does an arbitrary router discover the QoS
> request desired for this particular packet?
The specific proposal for intserv usage is to signal the
flow label value in an RSVP Filter_Spec, as part of the
flow setup process.
The specific proposal for diffserv usage is to use
IANA-assigned values, specifically as per RFC 3140. As with
the whole of diffserv, out-of-band configuration is also
rqeuired.
I only mention these answers as existence proofs; there may be other
QOS mechanisms in the future with different answers.
Brian
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
