[Nsi-wg] Minutes: GLIF NSI Implementation Group (Jan 22, 2014)

[John MacAuley]

Attendees (please let me know if I missed anyone):

John MacAuley, ESnet
Henrik Thostrup Jensen, NORDUnet
Radek Krzywania, Poznan Supercomputing and Networking Center
Joe Mambretti, Northwestern
Hans Trophert, SURFnet
Alex Moura, RMP
Freek Dijkstra, SURFsara
Jerry Sobieski, NORDUnet
Chin Guok, ESnet
Inder Monga, ESnet
Guy Roberts, DANTE

1. John presented the attached slide package discussing the mandate of the group, how we would like to operate, and how we would like to interact with existing NSI related groups.



Key point: Items will be discussed within the group only when a contribution has been made available 4 - 5 days before the meeting.  We need focus to avoid wandering discussions.

2. Discussed a list of topics we need to address in the short term.

a) Discovery Service - expand the existing discovery service proposal to cover new requirements from existing NSI implementations and changes in NS 2.0.  Action: John and Henrik to work together to revise the existing discovery service contribution for review.   Expecting prototype implementation to validate contribution.

b) Ethernet Service - we are using VLAN labels in the A-GOLE topology that are yet to be specified in a formal schema with a defined namespace.  We have agreed to capture the existing Ethernet parameters and formalize the definition.  Action: John to capture existing parameters and define in a formal schema definition for review.

c) Error handling - Expand existing NSI error set to include new errors values introduced through NSI 2.0 implementations.  Define required error feedback for taking corrective action on the errors (especially for path finding).  Action: John to write contribution as an initial proposal for discussion.

d) AAI - Hans has a proposal for end user authentication and authorization that will be running in the European production network, and we hope on the A-GOLE.  He will bring in a formal contribution for review.  Action: Hans to write a contribution for review.

e) Requirement for path finding - Formalization of topology requirements for pathfinding (both chain and tree).

f) Topology discovery - We have two primary use cases for topology exchange; support of dissemination of information for full tree based path finding, and summary path vectors for chain based routing.  Action: Chin to update his OGF contribution with input from Henrik, and review proposal.  Expecting prototype implementation to validate contribution.

3. Jerry related concerns that standardization activities need to occur in the OGF NSI-WG and not the GLIF NSI Implementation group.  Discussion occurred around providing implementation feedback on the standards into the GLIF NSI Implementation group so solutions can be vetted, and formal contributions defined for submission back into the OGF NSI WG.  Action: Inder, John, and Jerry to discuss concerns and propose resolution.

Please let me know if I have missed anything.

Next Wednesday is the OGF NSI WG meeting, and back to the GLIF NSI Implementation group meeting in two weeks.

Thank you,
John
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ogf.org/pipermail/nsi-wg/attachments/20140122/28e9c024/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: NSI Implementation GLIF-2.pptx
Type: application/vnd.openxmlformats-officedocument.presentationml.presentation
Size: 50140 bytes
Desc: not available
URL: <http://www.ogf.org/pipermail/nsi-wg/attachments/20140122/28e9c024/attachment-0001.pptx>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ogf.org/pipermail/nsi-wg/attachments/20140122/28e9c024/attachment-0003.html>