The MPLS WG Archive

["S.Matsushima" <satoru@ft.solteria.net>]

At 13:03 04/03/08 +0000, Adrian Farrel wrote:
> I am sorry, I am having
> some trouble understanding the requirement. Please see 
> below.
>  
> > > >
> >       (2) At an intermediate LSR,
> accounting of traffic through
> > > >
> >           LSPs for
> each pair of ingress to egress.
> > > > SP uses merging LSP and need per-lsp basis accounting,
> they will do it
> > > > over merging LSP.
> > >But downstream of a merge point you cannot tell which ingress
> the data 
> > >came from.
> > 
> > Yes.
> > Therefore, it should be a requirement.
> > More clearly, if the LSP operated as merged end-to-end, which
> ingress the 
> > data came from is only needed in the point of per-lsp accounting
> router
> > (maybe intermediate-lsr).
>  
> Consider the following network.
>  
> A--B
>    
> \         |--H
>     
> E--F--G--|
>    
> /         |--I
> C--D
>  
> Data flows from A and C to H and I. An LDP
> LSP is provided from B and D to G.
>  
> 1. When you say you want to account traffic
>    through an LSP for each pair of
> ingress 
>    to egress, do you
> mean
>    a. source and destination pairs
>       {A, H}, {A,
> I}, {C, H}, {C, I}
>    b. LSP ingress egress
> pairs
>       {B, G}, {C,
> G}


It is "b".


> 2. At which LSRs do you
> want to be able
>    to perform this
> accounting?
>    a. Ingresses B and 
> D
>    b. Egress G
>    c. Merge point E
>    d. Transit point
> F


Possibly all of these.
Where it is performed is where SP wants to perform it.

B and D should perform this accounting naturally, also E (maybe it depend
implementation) too.
In this case, what this requirement is that F and G needs something to
know which ingress the data
came from.


>  
> Thanks for helping me to understand what it
> is you want to achieve.
>  
> Adrian
>  

Thanks for your consideration too.


--
Satoru Matsushima