Next steps for GMPLS control of Ethernet

["Adrian Farrel" <adrian@olddog.co.uk>]

Hi CCAMPers and GELSters,

In CCAMP this time we explained the discussions and agreements that the 
chairs had with the ADs with respect to the way forward for developing
GMPLS extensions and procedures for the control of Ethernet devices. 
This email is to clarify the explanation and to give a base reference for 
the procedures that we expect to be followed.

- Work on procedures and extensions to GMPLS for the control
 of Ethernet devices will be done in CCAMP.

- This work is already covered by the CCAMP charter...
     The CCAMP working group coordinates the work within the
     IETF defining a common control plane and a separate common
     measurement plane for physical path and core tunneling
     technologies of Internet and telecom service providers (ISPs
     and SPs)
 However, if necessary, and if work is forthcoming, we will consider
 adding a specific work item an milestones to the charter. Such a
 charter change is not gating on this work, but may be useful for
 clarity and exposure of the work outside of CCAMP.

- Data plane specification is out of scope for the IETF. It must not
 be discussed on the CCAMP list. It may be discussed on the
 GELS list.

- CCAMP will not work on the specification of GMPLS except for
 data planes that are already approved by an SDO with responsibility
 for the Ethernet data plane. It is expected that the primary SDO
 in this case is IEEE.

We expect that you would use the following steps in order.

1. Decide upon the data plane behavior you wish to control

2. Ensure that the data plane you wish to control is approved
and documented by an SDO with responsibility for Ethernet
(e.g. IEEE).
This may be a null op in the case of existing standards, but
might require you to do work in the other SDO for new
standards for profiles of existing standards.

3. Start protocol work in CCAMP with an explicit pointer
to the approved data plane. If you feel the need to explain
your data plane in your I-D, then there is probably something
wrong with the specification in the other SDO and you
should return to step 2.

4. When it is clear that protocol work is progressing,
Deborah and I will liaise with the other SDO to get their
blessing for using a control plane for the nominated data
plane.


Note that this procedure can apply to multiple data planes at the same
time or separately, but we expect that the protocol encoding 
extensions will be flexible to all data planes.

Hope this is clear,
Adrian and Deborah