Hi Raju
I see that what could bind SLSs to diffserv is the use of the DSCP. But, this
DSCP is not necessarily related to the interior DSCP which is used to select
the PHB, at least this what draft-tequila-sls-00.txt states.
So, IMO it doesn´t matter whether a domain uses diffserv or not, as long as it
is able to met its negotiated SLSs. This is what I call a "general solution".
If this is true, then I think sls drafts shouldn´t reference diffserv (or other
technology), except for examples or appendices. But they refer to diffserv as
it would the target technology for SLS defintion and negotiation.
For example, draft-tequila-sls-00.txt:
" The basic assumption of this draft is that IP services will be
deployed over a public IP infrastructure, which will be (partly if
not completely) composed of diffserv-aware network elements ([RFC-
2475], [DS-MODEL]). These network elements are able to implement Per
Hop Behaviors (PHBs), including the Assured Forwarding PHB ([RFC-
2597]), and the Expedited Forwarding PHB ([RFC-2598]."
This doesn´t mean I have something against diffserv. I think that any reference
to specific technology should be removed from the drafts where their are not
necessary to preserve generallity.
Regards
Carlos Kamienski
Raju Rajan wrote:
> Hi Carlos,
>
> Could you expand a little bit what you see are the differences between
> "general
> solutions for SLS definition and negotiation or specific ones for diffserv"?
> From my understanding I can see two (relatively minor) differences:
>
> 1) The DSCP has to be communicated in the former instance. Typically, this
> would be one field in the service definition.
> 2) We may (as examples, or appendices) have to address specific services
> like EF and AS.
>
> What further differences in emphasis/content do you see?
>
> raju rajan
>
> ----- Original Message -----
> From: Carlos Alberto Kamienski <cak@cin.ufpe.br>
> To: sls-list <sls@ist-tequila.org>
> Cc: Raju Rajan <rajan@research.att.com>; Yves T'Joens
> <yves.tjoens@alcatel.be>
> Sent: Monday, February 26, 2001 4:06 AM
> Subject: Re: [tequila/sls] sls wg - proposed charter
>
> Dear Yves and Raju
>
> What is not clear to me is if slsu wg is going to present general
> solutions for SLS definition and negotiation or specific ones for
> diffserv.
> Previous discussions in this list let the impression people were looking
> for solutions for diffserv. Presented drafts also directed their focus to
> diffserv. Now, it seems the proposed charter is changing radically this
> vision:
>
> "As the architecture and negotiation framework for Service Level
> negotiation is not particularly bound to IP networks, it is to be
> understood that the procedures as discussed within the proposed WG are to
> be extensible to optical networks (o-sls)."
>
> So, if the outcomes of the group are not to be bound to any specific
> technology and diffserv is just instantiation, then I think this should
> be clear in the drafts. Otherwise, if it is to be bound to diffserv, then
> the charter should be changed.
>
> Carlos Kamienski
>
> > Dear sls-subscripts,
> >
> > please find attached a proposed charter for the sls work (with a little
> > delay, for which my apologies).
> > The objective of sending it around is to gather feedback on the listed
> > objectives and end-results, and as such all comments are warmly
> > welcomed. We have covered in the charter what we felt had support during
> > the last ietf meeting.
> >
> > Based upon the suggestions of this list, i will try to arrange a chat
> > with the ops area directors on the subject during the minneapolis
> > meeting.
> >
> > A new version of the framework draft, covering the technical remarks
> > during the san diego bof, will be made available before the minneapolis
> > meeting.
> >
> > further information can be found at
> > http://www.ist-tequila.org/sls.html
> >
> > yves & raju
This archive was generated by hypermail 2b29 : Wed Feb 28 2001 - 16:27:33 CET