Hi Shahram,

 

Absolutely, the solution described in this draft does not aim at addressing all 
possible use cases.

Other solutions, like the ones described in your draft “Transporting PTP 
messages (1588) over MPLS Networks”, are also valid and may be more adapted to 
specific scenarios.

 

Note that, in theory, it might be possible to use a link local addressing 
solution with multiple timing domains, assuming that multiple instances of BCs 
or TCs (one per domain) are implemented in the intermediate network nodes.

I agree however that this may not be the simplest solution from the 
implementation perspective.

 

The major reason that I wrote this draft was to introduce this second flavor of 
solutions, and to highlight the fact that carrying PTP messages over MPLS 
networks is possible without necessarily using an MPLS tunnel.

 

Thanks.

 

BR,

 

Sébastien

 

De : Shahram Davari [mailto:[email protected]] 
Envoyé : mercredi 7 mars 2012 15:29
À : JOBERT Sebastien RD-CORE-LAN; '[email protected]'; '[email protected]'
Objet : Re: [TICTOC] TICTOC meeting at IETF-83

 

Sebastien,

You probably know that Your proposed method can't support tunneling multiple 
1588 flows belonging to multiple customers or domains. 

Regards
Shahram
 

From: [email protected] [mailto:[email protected]] 
Sent: Wednesday, March 07, 2012 01:51 AM
To: [email protected] <[email protected]>; [email protected] <[email protected]> 
Subject: Re: [TICTOC] TICTOC meeting at IETF-83 
 

Hi,

 

I have uploaded a draft which discussed the topic of PTP over MPLS networks: 

http://www.ietf.org/id/draft-jobert-tictoc-ptp-link-local-00.pdf

http://www.ietf.org/id/draft-jobert-tictoc-ptp-link-local-00.txt

 

Abstract

 

   This document introduces a method for transporting PTP messages over

   an MPLS network supported by an Ethernet physical layer. The MPLS

   layer itself is not used to carry the PTP messages with this method;

   instead, a link local Ethernet channel is used. Several advantages

   related to this method are highlighted in this document. The method

   targets in particular telecom applications requiring accurate

   phase/time synchronization, with "link-by-link" PTP architectures,

   where all the network nodes support a PTP function, such as Boundary

   Clock or Transparent Clock.

 

Any comment is welcome.

 

Yaakov, as I told you by email, I would like if possible to briefly introduce 
this draft during the TICTOC meeting (I guess that 10 min will be sufficient).

I think it is connected in a way to the topic discussed in the draft 
“draft-ietf-tictoc-1588overmpls-02”

 

Thanks.

 

BR,

 

Sébastien

 

De : [email protected] [mailto:[email protected]] De la part de 
Yaakov Stein
Envoyé : mardi 6 mars 2012 17:50
À : [email protected]
Objet : [TICTOC] TICTOC meeting at IETF-83

 

Hi all, 

 

Please note that the TICTOC session is early Monday morning, so make your 
travel plans accordingly.

 

I have uploaded a preliminary agenda to the meeting site 
http://www.ietf.org/proceedings/83/agenda/agenda-83-tictoc.txt

 

If you believe that additional items need to be discussed, please email the 
chairs.

 

Y(J)S

 

_______________________________________________
TICTOC mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/tictoc

Reply via email to