Re: ISPACs

["Justin W. Newton" <[e-mail]>]

At 12:00 AM 12/6/96 -0800, Tony Li wrote:

>   A common
>   interconnect model does improve things, but that takes us back to a
>   provider based model where the people running the interconnect are a
>   provider, which leaves us at the current model, no different.
>
>I don't follow your leap here to the people running the interconnect being
>a separate provider.  They're providing a room and bandwidth and mebbe
>routing services under contract to the ISPAC (as a group).  Does that make
>them a competitor?  Is MFS a competitor because of MAE-East?

No, you are missing my point here.  The packets ned to get to the
interconnect somehow.  There are 2 ways that the packets acn get to the
interexchange (unless I am missing something very obvious)

1) The providers who are members of the ISPAC announce the aggregate
announcement and then route the specific traffic for individual ISPs to the
interconnect (or over the interconnect, whatever).

2) The interconnect itself advertisies the aggregate announcement and then
routes to the ISPAC members.

Setup 1) leads to the problem of dependancy on your competitors to get your
traffic from the Internet at large.

Setup 2) basically makes the interconnect provider your upstream provider,
the same as Sprint, MCI or UUNET would be.

What am I missing?  (Please use small words as I am obviously missing what
you are trying to teach me.)



>
>Tony
>
>
>

Justin Newton
Network Architect
Erol's Internet Services