[Nml-wg] Meeting notes Sep.24

[Guy Roberts]

Paola,

As per agenda item 1, the NSI and NML working group time slots on Tuesday have now been swapped.

Guy


-----Original Message-----
From: Paola Grosso [mailto:p.grosso at uva.nl] 
Sent: 24 September 2009 16:19
To: Network Markup Language Working Group
Subject: [Nml-wg] Meeting notes Sep.24

Hi all

These are the  notes of the meeting that just terminated few minutes ago.
Please send corrections to the mailing list if I got something wrong.
In particular re-read point 3 on transitional links 'cause I am not sure
I really got what people were saying.

Best regards, Paola


******* Meeting notes NML-WG Thurs. Sep 24 2009 *******

Present: Gigi, Martin, Aaron, John, Jeff, Evangelos, Paola and Jeroen.

Agenda items discussed:
    1) OGF27
    2) Current status of schema deliverable
    3) Transitional links
    4) Next meeting

1) OGF27

3 sessions requested. Currently on Tues. and Wed. morning. There is an
outstanding request to  move these sessions to the afternoon, so that
NSI can be held in the morning.
The last two sessions will focus on the _service_ and _adaptation_
elements in the schema.


2) Current status of schema deliverable

Works is in progress:
- Jeroen is incorporating changes mailed so far to the mailing list,
- Freek is focusing on the identifiers and usecase sections
- Jeroen will contact Lars as he is has been 'silent' for a while (Lars?)


3) Transitional links

John, Guy, Jeroen and Freek had a call yesterday with ITU people. They
discussed transitional links.

Jeroen says there are three main usecases for the NML schema:
(a) intra-domain provisioning, where detailed information is required by
the provisioning system
(b) monitoring applications
(c) inter-domain path finding

While the current schema covers well usecases (a) and (b), it is less
suited for (c) as this requires aggregated/abstracted views.

Evangelos proposes to use different identifiers for resources that are
in the 'physical' and 'aggregated' topologies, with mapping withing the
two sets of resources.  Jeroen points out this might not work when
dealing with multi-layer descriptions.

Martin says that domain 'collapsing' to node is already possible in the
current schema. A 'switching matrix' would be used to used to describe
the capabilities of a domain, in a similar manner as it is used for a
single node.

The consensus is that we need to provide a usecase describing an
aggregation/abstraction example.

John thinks transitional links are necessary to describe connections
between layers, and as such are needed by path finding apps. Jeroen and
Martin say that transitional link map 1:1 to NML adaptations. They are
an edge in the path finding graph.

We cannot reach a consensus, so we agree that:

- we need an usecase with multi-layer inter-domain path finding. This
might (or not) prove John's point we _do_ need transitional links;
 - we will might touch on this subject again at OGF27 when discussing
adapatations.

4) Next meeting.
To be decided at OGF.




_______________________________________________
nml-wg mailing list
nml-wg at ogf.org
http://www.ogf.org/mailman/listinfo/nml-wg