[Nml-wg] Notes for teleconference Feb 16
Jeroen van der Ham
vdham at uva.nl
Thu Feb 16 12:27:06 EST 2012
Below are the notes of the call we had today.
Attendance: Jason Zurawski (I2), Aaron Brown (I2), Freek Dijkstra (SARA), Yufeng Xin (Renci), Ilia Baldine (Renci), Roman Łapacz (PSNC), Jeff Boote (I2), Chin Guok (ESnet)
We go quickly through the NML schema document. Freek has updated it to the required syntax and added a short section on Identifiers.
Jeroen mentions that updates to the Introduction, Identifiers, and Example section are required, furthermore the definitions have to be checked against the latest schema diagram.
We will move the diagram first, Jeroen will write actionpoints for what has to be fixed.
Security section will be copied from the NM document and considered in light of current requirements.
Roman has worked on the Automated GOLE example and created an NML-ized version. The last few days we have had a discussion on this and he has further improved on the translation, adding the xCard description.
Jeroen notes that the NSI topology assumes bi-directional topology, while NML uses uni-directional topology. In time NSI will also support uni-directioanl requests, but this is not high on the list at the moment.
We decide to continue with both models and try to define a mapping between them.
next vs connectedTo: next is used for ordering in lists and should not be used outside of that context. We conclude that connectedTo is fine to be used, and does not clash with anything in the NML standard. This means that namespaced attributes don't seem to be necessary after all.
Freek mentions that it would be good to have a discussion between NSI and NML regarding topology approaches. He will set this up.
We discuss the concepts of STP and SDP.
STP: Service Termination Point, the point to which the service is provided. These are typically mapped locally to actual network endpoints.
SDP: is a pair of connected STPs which are in different domains.
The current NSI does not actively use the SDP, although in the future they may be used to define extra intermediate points of network connections.
Freek proposes to make a real NML version of the Automated GOLE example together with Roman and then compare the two approaches and see if we can map them.
Jeroen to go through document and set up action points
Jeroen to check the Security considerations of NM
Freek to contact NSI to set up topology discussion
Freek and Roman to create an NML version of the topology
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 235 bytes
Desc: Message signed with OpenPGP using GPGMail
More information about the nml-wg