Hi, Linda. On Aug 28, 2014, at 7:08 PM, Linda Dunbar <[email protected]> wrote:
> Suggestion of the wording: > > It would be more clear to have the third paragraph on "In pursuit of the > solutions.." right after the bullets under "The NVO3 WG will develop..." > because it is a very important deliverable of the WG. The 2nd paragraph, including those two bullets regarding architectural tenets, is meant to elaborate on the scope of the WG and define our work at the highest level. The 3rd through 6th (inclusive) paragraphs are meant to define specific areas of work: architecture, control plane, data plane, and use cases. I don’t think that merging paragraph 2 and 3 would change the meaning of the text. But it might reduce the clarity, by mixing two purposes together in a single paragraph. I would rather keep them separate as is done in the latest proposed charter text. > Instead of "In pursuit..", I would suggest the following wording: > > " To have a common baseline for all the solutions to be developed by NVO3, > the NVO3 WG will prioritize finishing the architecture document for network > virtualization within a data center environment" (because it is already an WG > draft. By the way I just noticed NVO3 architecture draft has expired). One of the goals of this recharter, which is agreed upon by the chairs and our ADs, is to get away from serialized process. In other words, we specifically want to get away from waiting for high-level documents before allowing more specific work on solutions. Rather, we do want for work on each specific area of work (architecture, control plane, data plane, and use cases) to proceed in parallel. > Some suggestions on the MILESTONES: > > I think application specific multicast is an important area, especially for > data centers whose underlay IP network don't support PIM or traditional L3 > multicast schemes. Many data centers fall into this category. I don’t particularly like the phrase “application specific multicast”, but I think I understand what you mean by it. And I agree with your statement that it’s an important area for the WG to address. I also think that multicast traffic can be taken into account when documenting the data and control plane protocols already enumerated in the milestones. It isn’t clear to me that there is any reason to call it out specifically, nor what such a milestone would look like. > Another important area is seamless VM mobility in Data Center environment to > avoid sub-optimal routing, as described by > draft-ietf-nvo3-vm-mobility-issues-03.txt. If there is WG consensus to publish the mobility issues draft then we can do so. We don’t have to specifically create a milestone for it. That being said, if there is still significant work to-be-done on the draft then the chairs will probably want to create a milestone (just for the sake of good practice, transparency, etc). I think we should decide how to handle this based on a WG discussion about the draft, instead of binding it to the rechartering. It might be worthwhile to have a brief presentation on the draft at the 11-Sep virtual interim meeting - can one of the authors lead this? Cheers, -Benson _______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3
