Hi the authors and all,


Some comments for clarification:



(1)     How to measure the link delay, loss… before the LSP is created? We know 
that the existing mechanisms for the measurement (delay, loss…) are based on 
in-service.



(2)     Why not take into account the node delay, loss? How much percentage 
between the node delay and link delay?



(3)     How much different between the component links for the delay or loss? 
How to advertise the TE links information when there is much different between 
the component links?



(4)     How much difference between the links (not only the component links 
between the same ends) in the networks? I mean if there is not much different 
(they could be approximate), the end-to-end delay can be simpily caculated as 
“number of the hops * delay (approximate constant value in the network)”.



(5)     Can this routing extensions also be used in GMPLS networks (e.g., TDM 
network or WSON)?








Thanks



Fatai





-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Acee 
Lindem
Sent: 2011年10月5日 5:45
To: OSPF List
Subject: [OSPF] OSPF TE Extensions for Delay, Loss, and Jitter (formerly known 
as OSPF Express Path)



We've discussed this draft at the last two IETFs and there seemed to be 
interest. We believe we sorted out the positioning/overlap with the MPLS and 
CCAMP WGs and are now ready to accept it as a WG document.

Does anyone have any objections?



Here is a link to the current draft: 
http://www.ietf.org/id/draft-giacalone-ospf-te-express-path-02.txt



Thanks,

Acee and Abhay

_______________________________________________

OSPF mailing list

[email protected]

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

Reply via email to