Dear all,
I most welcome the decision that the subject dealt in
Papadimitriou’s draft is within the scope of CCAMP charter.
Regarding the issue of VLAN switching, this is clearly not the feature
defined in IEEE 802.1Q. However, I do not think IETF should
wait until IEEE define new architecture. I am wondering
if the scope of CCAMP work exclude any technology requiring
modification on legacy hardware structure.
If it is not, considering industry needs on high-performance
Ethernet switch, I think it is desirable the scope of work
include both control plane and data plane. However, I also
stress that the extent of work must be discussed and defined
in a separate requirement document prior to architectural work.
I raised this issue in
http://www.ietf.org/internet-drafts/draft-jaihyung-ccamp-lesrequirement-00.txt,
with some discussion on other candidate technologies.
Unfortunately, I didn’t get enough response on the proposal
to date. Definitely, there are other technologies we can
consider for L2SC interface. I hope people pay attention
on the draft and also draft-jaihyung-ccamp-lesframework-00.txt
Regarding Papadimitriou’s proposal, I have some doubts on
compatibility issue. I think it may not be used with Ethernet
switches allocating VID for different purpose. I am wondering
how VLAN information of customer network can be delivered
across provider network.
Other point to comment is, personally I think it is desirable to
focus just on Ethernet interface. The draft is too heavy
compared to other Ethernet control protocols such as MSTP.
In fact, RSVP and IP routing protocols are all heavy to Ethernet.
Why should ISPs providing just metro-Ethernet service must
know all the unnecessary options devised for TDM and ATM?
I think it will be helpful to industry if CCAMP work on
dedicated, lightweight specification for Ethernet only network.
Thank you
Jaihyung Cho
Hi,
The only issue that I have is with VLAN switching. Since VLAN switching
is not a standard 802.1Q behavior, it can't be used with existing Ethernet
hardware. Therefore the scope of this draft is not limited to control-plane,
and requires new data-plane that is not defined in IEEE yet.
If the VLAN switching is removed from the draft, I support accepting it as
a WG draft.
Yours,
-Shahram
> -----Original Message-----
> From: owner-ccamp@ops.ietf.org [mailto:owner-ccamp@ops.ietf.org]On
> Behalf Of Adrian Farrel
> Sent: Sunday, January 23, 2005 6:46 AM
> To: ccamp@ops.ietf.org
> Subject: Layer 2 Switching Caps LSPs
>
>
> All,
>
> There is a draft
> (draft-papadimitriou-ccamp-gmpls-l2sc-lsp-03.txt) that we
> have discussed at several of the more recent CCAMP meetings, and have
> decided that the subject is within scope for our charter.
>
> The questions we have faced have been:
> - is the problem well enough articulated?
> - is this the solution that the WG wants to pursue?
> - is there a high enough level of interest in this work?
>
> If the answer to all three questions is "yes" then we can
> adopt the draft
> as a WG document and move forwards.
>
> Note: I think there are a large number of minor issues to
> clear up with
> this draft, but hopefully this is orthogonal to whether we
> make this a WG
> draft or not.
>
> Thanks,
> Adrian
>
>