Hi, Some of it was my request I think, and to clarify, I meant that: Even though a concrete lightpath would obviously be a sequence (subclass of group :) of network elements, a path provisioning _request_specification_ could be structurally quite different. That specification could contain elements like: 1. Path must start at endpoint A 2.1 Path must go specifically through links F,G,H in that order 2.2 Path must go through whatever links you need to, just start from link K and get to link Q 3. Path must end up at endpoint Z I don't see a simple mapping to groups here... I guess ultimately it's up to NSI to define this specification, but both specs will be referring to the same objects (links, endpoints, domains, technologies, ...), so it'll be useful to have input from both WGs. On Oct 3, 2008, at 11:29 AM, Freek Dijkstra wrote:

Today was an interim meeting from the network service interface workgroup (NSI-WG) in Seattle:

The NSI-WG likes to be able to both describe the network itself, as well as the usage of the network. Thus, both describe the topology as well as a configured (or reserved) end-to-end path.

I think this is obvious, but this request was explicit, so I repeat it here.

I presume (but haven't discussed that yet) that also means that they want us to come up with names (path, circuit, network connection, or lightpath) and more advanced properties or distinctions (is a path unidirectional or bidirectional, what about multicast, or protected paths, etc.)

I did tell them it's all a subclass of a group, though :-)

Regards, Freek _______________________________________________ nml-wg mailing list nml-wg@ogf.org http://www.ogf.org/mailman/listinfo/nml-wg