NWCRG @ IETF-97 (Seoul) minutes ------------------------------- Wednesday, Nov 16, 2016 Afternoon session I 13:30-15:00 Vincent Roca (present); Victor Firoiu (remotely) Slides available at: https://www.ietf.org/proceedings/97/slides/slides-97-nwcrg-01-chairs-slides-includes-01-02-05-00.pdf https://www.ietf.org/proceedings/97/slides/slides-97-nwcrg-03-fecframe-v2-01.pdf https://www.ietf.org/proceedings/97/slides/slides-97-nwcrg-04-network-coding-for-nfv-00.pdf * Welcome, administrative and general matters (5') -------------------------------------------------- This group last met during IETF93, July 2015, so more than one year ago. * taxonomy I-D update, Vincent Roca (10') ----------------------------------------- Previous version (-00) was submitted in Nov. 2014, two years ago. Current plan is to be ready for WGLC in IETF98 meeting, March 2017. => MILESTONE: start WGLC just after IETF 98 meeting * FECFRAME v2: adding convolutional FEC codes support to the FEC Framework, draft-roca-tsvwg-fecframev2-02, Vincent Roca (10') --------------------------------------------------------------------------- See the slides. Next steps will consist in: - updating the FECFRAME extension I-D. Note that the "version 2" tag will be removed and the document restructured as an extension to RFC 6363, without the need to obsolete it. Being smaller, it will be easier to read. Should be ready for IETF 98; - finishing the associated implementation by IETF 98 meeting; - submit a new RLC FEC Scheme I-D. Should be ready for IETF 98; WG Item adoption will be called in TSVWG for both I-Ds during IETF 98. * Network Coding for NFV, draft-vazquez-nfvrg-netcod-function-virtualization-00, Angeles Vazquez-Castro (15') ------------------------------------------------------------------------------------------------------------- See the slides (presented at NFVRG). The associated document obtained a good feedback at NFVRG and Angeles is invited to continue the work. Comment Blaine McDonnell (AT&T): no need to create new acronyms to VNF, but the concept makes sense. Q Vincent: could we benefit from the outputs of the 3 projects mentionned, as the use-case details and architecture that has been designed could be interesting to fuel the reflexions of the group. A Angeles: the H2020 project will soon finish and documents made public. The ESA projects have just become. Q Blaine McDonnell (AT&T): have you engaged discussions with the OPNFV and ??? organisations? Q Allison: What are the open research questions here? What about security? A Angeles: NC has very theoretical open questions but we can leave them aside. The main answers to what we need are here and the fundamental results are listed in the bibliography. Concerning security, we can reuse existing security to focus more on the enginering aspects that are Followed several questions and comments on the difficulty to maintain states as NFV functions are moved around. However those questions seem a bit marginal from the I-D main topic. * Proposals and open discussion, Vincent / Victor, all (20') ------------------------------------------------------------ Clarification of the group scope: we want to work on "coding techniques for efficient communications", which is broader than "network coding" stricto sensu (i.e. re-coding within transit nodes). The proposal consists in defining several topics of interest, identifying group of volunteers, and defining timing information and milestone(s). Examples of topics include: - "NC, from theory to practice" (already in charter); - "NC architecture" that will benefit from inputs from the previous topic; - "convolutional coding with and without recoding"; => We expect other topic proposals here, coming from the group. Comment Victor: We will contact previous contributers who probably moved to business activities. Comment Lars: If the community has moved on, it has moved on, and there's no community. Having people come to a meeting is fine, but this does not guarranty they will contribute in the long term. Comment Blaine McDonnell (AT&T): We are doing the optical access network by abstraction the physical hardware. This type of activity could be useful for us. General coomments (Lars and Allison): The group was basically closed, we thank you for trying to revive it, but it is not sure if there is still a group to work on the topics, although the work is interesting and fits the IRTF. Let's see what's the situation in 6 months from now, if there is an active community, and decide what to do. By that time, it is recommended to have a regular meeting (@ietf98) as well as an interim meeting, for instance co-located with an academic event. The idea is to attract new contributors likely to bring new attractive research results to the group.