Hi Allan, Thank you for your support on publishing the draft and giving good comments/suggestions. Your comments and triggered discussions clearly show that the draft has some terms overloaded or not used properly, which needs to be fixed. Yes, we can improve the structure and elaborate the motivation for different use cases.
Thank you, Joe, Reith, and Joel for the discussions on the comments, which is helpful to clear the text problem in Section 4.2. If the WG gets a consensus for publishing this draft, we will revise the draft to address these issues; hope for you to review again. Thanks/Regards, Lucy From: nvo3 [mailto:[email protected]] On Behalf Of David Allan I Sent: Friday, August 12, 2016 6:17 PM To: [email protected]; Bocci, Matthew (Nokia - GB) Subject: [nvo3] FW: Call for interest on NVO3 use case draft Hi Matthew: I’ve read the draft, and as much as I believe it is a useful piece of work, I think it misses the mark in a few cases. I believe if they were addressed, the utility of the document would be improved IMO this is primarily in that certain technical requirements which in themselves could be considered use cases are presented without rationale…. This is primarily confined to section 4…. In 4.1 the ability to interconnect different technologies, fine… but then it diverges in discussing a tiered application and zone interconnection. Unrelated to the heterogeneous technology problem, and why I’d want interconnect between the web tier and the storage tier completely escapes me with or without a firewall… I’d recommend splitting these use cases, and providing either a different example of firewalled zone interconnect or a rationale for the example that elaborated a bit more on the motivations. 4.2 Why I would subnet my overlay could use some explanation. I normally think of subnetting as a convenient address summarization technique dependent on topology, and with an overlay I don’t have a topology. So the reason for introducing the complexity should be explained rather than simply offering this as a feature. Is this requirement elaborated on in any other drafts….? If I were to generalize where the use cases were going, I think the key themes are: 1) Separation of tenants (and rationale, although somewhat obvious ☺) 2) Rationale for multiple tenant networks vs. multiple zones in a single network and perhaps cases for both e.g Separation of zones within a tenant network for either technical or security reasons. Vs. hardwalling separate networks with no interconnect 3) Partitioning of tenant networks on the basis of L2 and/or L3….and a rationale for same. I think something along these lines would give the document a lot more structure and focus, and perhaps address some of the concerns vis-à-vis this as a standalone document vs. being folded into other documents. My 2 cents Dave
_______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3
