RE: [Pce] draft-leroux-pce-of-00.txt

Hi Jerry, 

> -----Message d'origine-----
> De : ASH, GERALD R, ATTLABS [mailto:gash@att.com] 
> Envoyé : mercredi 14 mars 2007 21:17
> À : LE ROUX Jean-Louis RD-CORE-LAN; pce@ietf.org
> Cc : ASH, GERALD R, ATTLABS
> Objet : RE: [Pce] draft-leroux-pce-of-00.txt 
> 
> Hi JL, 
> 
> > -----Original Message-----
> > From: LE ROUX Jean-Louis RD-CORE-LAN
> > [mailto:jeanlouis.leroux@orange-ftgroup.com]
> > Sent: Wednesday, March 14, 2007 1:52 PM
> > To: ASH, GERALD R, ATTLABS; pce@ietf.org
> > Subject: RE: [Pce] draft-leroux-pce-of-00.txt
> > 
> > Hi Jerry,
> > 
> > > -----Message d'origine-----
> > > De : ASH, GERALD R, ATTLABS [mailto:gash@att.com] Envoyé 
> : vendredi 
> > > 9 mars 2007 05:28 À : LE ROUX Jean-Louis RD-CORE-LAN; 
> pce@ietf.org 
> > > Objet : RE: [Pce] draft-leroux-pce-of-00.txt
> > > 
> > > Hi JL,
> > > 
> > > > -----Original Message-----
> > > > From: LE ROUX Jean-Louis RD-CORE-LAN 
> > > > [mailto:jeanlouis.leroux@orange-ftgroup.com]
> > > > Sent: Tuesday, March 06, 2007 6:50 PM
> > > > To: ASH, GERALD R, ATTLABS; pce@ietf.org
> > > > Subject: RE: [Pce] draft-leroux-pce-of-00.txt
> > > > 
> > > > Hi Jerry,
> > > > 
> > > > Thanks for the feedback and comments.
> > > > 
> > > > Please see inline,
> > > > 
> > > > > -----Message d'origine-----
> > > > > De : ASH, GERALD R, ATTLABS [mailto:gash@att.com] Envoyé
> > > : > samedi 3
> > > > > mars 2007 01:54 À : LE ROUX Jean-Louis RD-CORE-LAN; 
> pce@ietf.org 
> > > > > Cc
> > > > > : ASH, GERALD R, ATTLABS Objet : RE: [Pce] 
> > > > > draft-leroux-pce-of-00.txt
> > > > > 
> > > > > Hi JL,
> > > > > 
> > > > > Looks like a good start.  
> > > > > 
> > > > > Looking through the IANA section, I don't see
> > registration of the
> > > > > objective functions (OFs) required in Section 5.1.17 of
> > > > > http://www.rfc-editor.org/rfc/rfc4657.txt:
> > > > > 
> > > > > "  The PCECP MUST support at least the following 
> > > > > "unsynchronized"
> > > > >    functions:
> > > > > 
> > > > >    - Minimum cost path with respect to a specified metric
> > > > >      (shortest path)
> > > > >    - Least loaded path
> > > > >    - Maximum available bandwidth path
> > > > > 
> > > > >    Also, the PCECP MUST support at least the following 
> > > > >    "synchronized" objective functions:
> > > > > 
> > > > >    - Minimize aggregate bandwidth consumption on all links
> > > > >    - Maximize the residual bandwidth on the most loaded link
> > > > >    - Minimize the cumulative cost of a set of diverse paths"
> > > > > 
> > > > > Shouldn't these standard OFs and their parameters be 
> registered 
> > > > > from the start?
> > > > 
> > > > Actually we prefer to keep this draft generic, and 
> define specific 
> > > > objective functions in other documents. For instance 
> > > > draft-lee-pce-global-concurrent-optimization-02.txt
> > > > requests for three
> > > > code points within the OF registry.
> > > 
> > > I guess I have the same question.  Which PCEP document(s) will 
> > > specify the missing 6 objective functions, listed above, 
> as required 
> > > in Section 5.1.17 of the PCECP Generic Requirements (RFC 4657)?
> > 
> > Two objective functions listed in 5.1.17 are specified in 
> section 5.1 
> > of draft-lee-pce-global-concurrent-optimization-02.txt.
> > We can add other synchronized objective functions in the 
> next revision 
> > of the gco draft.
> > 
> > > I don't think these should be
> > > specified in 'other' (unnamed) documents, they should 
> appear in the 
> > > main PCEP specification document 
> > > http://www.ietf.org/internet-drafts/draft-ietf-pce-pcep-07.txt
> > 
> > Hum, this would create dependency between the PCEP spec which is 
> > stable and this new OF draft.
> > 
> > > , or perhaps in draft-leroux-pce-of-00.txt.
> > 
> > We don't really like this option as we want to keep the OF draft 
> > generic.
> > 
> > One could write a short draft that would define a set of 
> > straightforward unsynchronized objective functions 
> including those we 
> > listed in 4657.
> 
> IMO an important missing PCECP requirement is identified in 
> http://www.ietf.org/internet-drafts/draft-ietf-pce-pcep-07.txt
>  Appendix A (Compliance with the PCECP Requirement Document):
> 
> "   Here is the list of currently unsatisfied requirements:
> 
>    o  Allow to select/prefer from advertised list of standard 
> objective
>       functions/options
> 
>    o  Allow to customize objective function/options
> 
>    o  Support "unsynchronized" & "synchronized" objective functions"
> 
> They all deal with the missing 6 objective functions required 
> in RFC 4657, and are an important omission in PCEP, so far.

In montreal we indicated that objective functions would not be covered in the base PCEP spec. 
There was no objection. This was recalled in San Diego, again no objection...

> 
> You say that PCEP is 'stable' so you don't want to add them 
> there?  

What I say is that this would create dependancy between PCEP and OF drafts which is not good. The base PCEP draft should not have any dependancy with other drafts.

> Does that mean no further changes to PCEP before 
> WGLC? 

Definitely no!

But, again, note that it was agreed that objective functions would not be covered in the base spec.

> If so, the omission would be brought up during WGLC as 
> an important gap in PCEP.

As far as I know in 4657 we don't say that all MUST requirements MUST be covered in the same document...

> 
> You don't want them in the OF draft, OK, your call.
> 
> Finally, once again, I don't think other new drafts (e.g., 
> draft-lee-pce-global-concurrent-optimization-02.txt) that 
> specify perhaps *new* PCEP requirements, should be satisfying 
> the original RFC 4657 PCECP requirements on PCEP.  

Why?

Draft-lee specifies two objective functions listed in 4657, where is the problem?

> You also 
> say 'We can add other synchronized objective functions in the 
> next revision of the gco draft.'  I don't think that's the 
> right place to satisfy the PCECP requirements.

Why? 

>  And 
> furthermore you say that yet another draft 'would define a 
> set of straightforward unsynchronized objective functions 
> including those we listed in 4657.'  It doesn't make sense 
> IMO, rather, all 6 of these objective functions should be 
> specified in the PCEP document.
> 
> It would be good to have other opinions on this.

Regards,

JL

> 
> Thanks,
> Regards,
> Jerry
> 

_______________________________________________
Pce mailing list
Pce@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/pce