Hi Huaimo,

> You’ve computed the backup paths: [R4, R3, R6, R7] and [R5, R2, R9, R8].  
> This results in enabling flooding on (R3, R6) and on (R5, R2), (R2, R9), (R9, 
> R8).
>  
> [HC]: The enhanced algorithm enables the temporary flooding on (R3, R6) and 
> (R2, R9). It does not enable flooding on (R5, R2) or (R9, R8).


Thank you, yes, I missed that.


> Per the existing algorithm, the temporary additions would be (R3, R6) and 
> (R2, R9).  As it has less flooding, this seems like a better solution.
>  
> [HC]: It seems that the existing (rate limiting) algorithm may result more 
> flooding in some cases. For example, if there are links (R4, R8) and (R5, R7) 
> in Fig. 1, it seems that these two links may also be added to the FT for 
> flooding temporarily.


Ok, that’s possible.  However, if those links existed, wouldn’t your algorithm 
also make use of them?

I understand that your algorithm might pick only one path, but given that the 
path might not be functional, it seems like there would also be cases where a 
bit more flooding turns out to be the right answer.

Tony

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

Reply via email to