Summary: while there are a few nits and the security considerations
need to be completely rewritten, there are part of the document that
are really excellent. One persistent issue is that the language
is not always clear in distinguishing between L2 and L3 transport,
and L2 and L3 overlays.
Melinda
--------
Introduction (section 1), paragraph 3: what does "cope with"
mean? (also can delete "in order to")
Could use a little more clarity around what's meant by L2
and L3 throughout document. Might be useful to distinguish
more clearly between L2/L3 transport and L2/L3 emulation
1.2: under ELAN definition, MEF is not defined, recursive
definition
1.3: top-of-rack should be defined in 1.2, as should DC GW
Rewrite last para of 2.1:
Core nodes utilize L3 techniques to interconnect NVE
nodes in support of the overlay network. These devices
perform forwarding based on an outer L3 tunnel header, and
generally do not maintain per-tenant service state.
However, some applications (e.g., multicast) may require
control plane or forwarding plane information that
pertain to a tenant, group of tenants, tenant service, or
a set of services that belong to one or more nvo3
tunnels. When such tenant or tenant service-related
information is maintained in the core, overlay
virtualization provides mechanisms to manage that
information.
3.1.5.1 second para "Centralized controllers provide a
facility to maintain global and distribute that state to the
network [ ... ]" Global ?state?
4.1 it's not just "miscommunication" that can lead to
inefficiencies - misconfiguration may, as well.
Don't really understand the point about hash-based load
balancing. Is the issue hash collisions or something else?
Section 4 is overall really nice.
The security consideration sections needs a rewrite. I keep
promising to help with that and not be able to find the
time, but I'll try to get to it in the next few weeks.
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3
