Based on the discussion so far I have created the following. I will need to factor in modify, but at the moment this should do for the discussion. I have named the service elements: p2ps - This Point-to-Point Service element is used to specify a generic point-to-point service request in the NSI CS protocol. It can be the base for layer 1 services as well. ets - This Ethernet Transport Service element is used to specify a point-to-point Ethernet service request in the NSI CS protocol. evts - This Ethernet VLAN Transport Service element is used to specify a point-to-point Ethernet VLAN service request in the NSI CS protocol. In the schema vets extends from ets which extends from p2ps. I removed the "_v1" from the element names since the XML namespace identifies the version. John On 2013-07-11, at 11:29 AM, Henrik Thostrup Jensen <htj@nordu.net> wrote:
On Thu, 11 Jul 2013, Henrik Thostrup Jensen wrote:
However, ERO it not really something that is defines a network service. It is more like special instructions to an aggregator. If you specify an ERO or not, does not really change the service you get from a network perspective.
This is actually a good example of something that a network cannot provide, but a capability of an NSA(s). Should we consider putting this outside the service definition?
Thought a bit more about this on the way home. The source and destination STP are really not service requirements as well, but something the service requirements should hold for.
Should we consider something like seperating Path and ServiceRequirement. I.e., and object for the path (which can include an ERO), and a requirement which should hold for the path. However this is starting to get a bit hairy. I am overthinking this?
Best regards, Henrik
Henrik Thostrup Jensen <htj at nordu.net> Software Developer, NORDUnet
_______________________________________________ nsi-wg mailing list nsi-wg@ogf.org https://www.ogf.org/mailman/listinfo/nsi-wg