Re: [nvo3] Network Virtualization Overlay Control Plane Requirements posted

[Thomas Narten <[e-mail]>]

> Why was not this I-D submitted to dc?

I believe that the NVO3 effort, while related to 'dc', is separate
enough and scoped out enough that it shouldn't be part of 'dc'.

To me DC is a mailing list in search of a problem. 

In the case of both SDN and NVO3, I believe the efforts have done a
reasonable first cut at trying to carve out a specific problem area
that can stand on its own. By that, I mean can try to create a WG
based on a focused charter. Yes, both efforts still have work to do
before they can become WGs (if that is what eventually happens). And
yes, both efforts *relate* to data centers. But the way of the IETF is
to carve out self-contained WGs rather than trying to boil the ocean
in a big general area. If DC is ever to be successful, it needs to
tease out some concrete problems, with those individual problems at
some point leading to a focused WG. And such a WG would not be the "DC
WG", but a WG focussed on a much narrower problem.

But outside of SDN and NVO3, the DC "area" seems to be struggling with
coming up with concrete problems that the IETF should work on.

Thomas