Cell Relay Archive

[erze9119@my-deja.com]

Hi folks!

Thank you for your comments on the N square problem.

Now I understand that the N square problem concerns
only networks with PVC connections and not SVC capable
networks like PNNI for example.

BTW, in  the IP over ATM (overlay model) is the assumption
PVC connected ATM network, or is it possible to use SVC
connected network as well, and thus avoid the N square
problem?

Regards,

erze


In article <8tb684$4da$1@nntp.teliafi.net>,
  jorma.mellin@telia.fi (Jorma Mellin) wrote:
> In article <8t9k0u$l7l$1@nnrp1.deja.com>, erze9119@my-deja.com says...
>  >I am reading some MPLS docs, and in one doc. they calim that
>  >MPLS avoids the N-square problem associated with the over-lay
>  >problem. What I don't understand is that ATM creates N*(N-1)/2
>  >switched virtual conns. (SVCs), between N edge routers. MPLS
>  >also creates equal nr. of LSPs; how are these two different?
>  >Doesn't MPLS also suffer from the same problem? Can somebody
>  >knowledgeable explain this to me please?
>
> Well, I don't see the problem here. If you use SVC's on your
> ATM network, why should you care how many connections there
> are active? Connections are setup and teared down automaticly,
> and you always follow the fastest path (with PNNI). The amount
> of SVC's possible on a NNI link is huge, if I remember correctly
> there are 14-bits available for SVC numbers (short amount is reserved
though).
>
> I am not an expert with MPLS, but I remember that you don't have
> all LSP's at all PE devices (provider edge). However, keep in mind
> that if you think using MPLS VPN's, the VPN doesn't reach the customer
> edge (CE). The VPN terminates at PE.
> With ATM the "VPN" goes all the way to the CPE.
> Both of these (VPN's) are non-crypted by def, although there might be
> a little more security when the data are in cells, rather than in
packets.
>
> Jorma
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.