RE: VLAN label range requirement

["Ong, Lyndon" <Lyong@Ciena.com>]

Hi dimitri,

No, that's a fair point - in practice we have been testing up to 21v,
had not looked into the ability to signal 256v.

I suppose if you did have a co-routed 256v this would require
256 32-bit SUKLM values, for 1024 bytes.

The issue with VLAN IDs came about when someone calculated that
4094 12 bit VLAN IDs would consume 6144 bytes worth of message
(or 5 messages) if you just sent each VLAN ID without compressing this.



Lyndon

-----Original Message-----
From: dimitri papadimitriou [mailto:dpapadimitriou@psg.com] 
Sent: Wednesday, July 12, 2006 6:53 AM
To: Ong, Lyndon
Cc: dimitri.papadimitriou@alcatel.be; Stephen Shew; Adrian Farrel;
ccamp@ops.ietf.org
Subject: Re: VLAN label range requirement

lyndon

- not sure to understand - the information part of the liaison question
was "how large is large" ?

eg. RFC 3946 support VC4-256v did you ever ask the question whether we
have an efficient encoding ?


Ong, Lyndon wrote:

> Sure, I understand.  I think the concern that was expressed at the
last
> OIF
> meeting was whether we might need to define an efficient way of
> describing
> a set consisting of a very large set of labels, since the range of
VLAN
> tags,
> for example, is quite a bit larger than the typical set of labels
> assumed for
> optical switches.  If OIF comes up with 
> any proposals along those lines, it would then liaise this to CCAMP
for
> its
> consideration.
> 
> Cheers,
> 
> Lyndon 
> 
> -----Original Message-----
> From: dimitri papadimitriou [mailto:dpapadimitriou@psg.com] 
> Sent: Wednesday, July 12, 2006 6:33 AM
> To: Ong, Lyndon
> Cc: Stephen Shew; Adrian Farrel; ccamp@ops.ietf.org
> Subject: Re: VLAN label range requirement
> 
> lyndon -
> 
> listing a set of labels as part of the response is something already
> covered in GMPLS
> 
> hence if i well understand the MEF.10 document the information
elements
> that needs to be considered is quite straightforward
> 
> so do we need this loop back and forth ? why not just produce a 2 page
> doc and close this ?
> 
> thx,
> - d.
> 
> Ong, Lyndon wrote:
> 
> 
>>That was my interpretation also, not sure if that came through on the 
>>slides.
>>
>>Lyndon
>>
>>-----Original Message-----
>>From: owner-ccamp@ops.ietf.org [mailto:owner-ccamp@ops.ietf.org] On 
>>Behalf Of Stephen Shew
>>Sent: Tuesday, July 11, 2006 8:09 PM
>>To: Adrian Farrel; ccamp@ops.ietf.org
>>Subject: RE: VLAN label range requirement
>>
>>You're welcome.
>>
>>The response from CCAMP on this question is very thorough, and I 
>>certainly appreciate the thought that went into it.  My interpretation
> 
> 
>>of the response is that that CCAMP encourages the OIF to design an 
>>encoding for the VLAN identifiers, and communicate it back to CCAMP.
>>
>>Stephen
>>
>>-----Original Message-----
>>From: Adrian Farrel [mailto:adrian@olddog.co.uk]
>>Sent: July 11, 2006 18:42
>>To: Shew, Stephen (CAR:Q840); ccamp@ops.ietf.org
>>Subject: Re: VLAN label range requirement
>>
>>Thanks Stephen.
>>
>>What is your opinion of the response to this question that CCAMP 
>>supplied in its return communication to the OIF?
>>
>>Adrian
>>----- Original Message -----
>>From: "Stephen Shew" <sdshew@nortel.com>
>>To: <ccamp@ops.ietf.org>
>>Sent: Tuesday, July 11, 2006 4:44 PM
>>Subject: VLAN label range requirement
>>
>>
>>In yesterday's meeting, there was a request for more information on 
>>the VLAN label range requirement.  From the original OIF liaison, the 
>>description was:
>>"We are investigating label formats to represent a list or range of 
>>VLAN identifiers, as used in MEF.11 bundling. In the case where a 
>>large number of non-consecutive VLAN identifiers is used for the same 
>>connection, we would like to keep the label to a reasonable size."
>>
>>The requirement to send a range of VLAN identifiers in signalling 
>>arises in the bundling feature in Ethernet services as specified in 
>>Metro Ethernet Forum (MEF) and ITU-T SG15 documents.  In the MEF, the 
>>MEF.11 spec "User Network Interface (UNI) Requirements and Framework" 
>>there is a UNI service attribute called "bundling" that is described 
>>as "A UNI attribute in which more than one CE-VLAN ID is associated
> 
> with an EVC."
> 
>>It is more fully described in MEF.10 "Ethernet Services Attributes 
>>Phase 1".  Both technical specs may be obtained free at:
>>http://www.metroethernetforum.org/TechSpec.htm
>>
>>Equivalent ITU-T Recs. are G.8011, G8011.1, and G.8011.2.  These are 
>>available to ITU-T members, or under a "3 free" capability on the 
>>ITU-T website.
>>
>>Of interest is the MEF Ethernet Link Management Interface (E-LMI) 
>>defined in MEF.16 that has an encoding for passing VLAN ranges between
> 
> 
>>the UNI client and network.  This in a structure called the "CE-VLAN 
>>ID/EVC Map information element".
>>
>>I hope this provides more clarity to the VLAN label range issue.
>>
>>Stephen Shew
>>Metro Ethernet Networks
>>Nortel
>>sdshew@nortel.com
>>Telephone: +1 613 763 2462 / ESN 393 2462
>>
>>
>>
>>
>>.
>>
> 
> 
> .
>