Re: PPP over SONET

[Edward Henigin <ed () texas net>]

        I was thinking... as far as my limited knowledge allows..

        It seems that PPP or HDLC over OC-3 would be faster than
ATM over the same OC-3 circuit, and (at this stage) have fewer
interoperability problems.

        But you can configure PVC's with ATM, working with an ATM
cloud, similar to frame relay, no?  And if this is the case, then
a large backbone-type provider may desire to set up an ATM cloud,
making it easy to add POP's (just add them to the cloud), and allow
for more layer 2 routing (just configure PVC's between every POP you
have, building a full web, and a traceroute would now travel 2 or 4
hops instead of 12...?)

        instead of some sort of hierarchichal star topology, you'd
have a large web...

        and one more potential benefit to ATM -- won't it scale
very well to OC-12?  ie, when you want to upgrade your network to
OC-12, you'd have no changes to make?

        just thinkin...  and I love it when y'all correct me, I tend
to learn a lot :)


On 20 May 1996, Camil Samaha/TEIR/Thomson wrote:

> I've been doing a lot reading the last few days on the advantages of "PPP over
> SONET/SDH" vs "IP over ATM AAL5".  The available information bandwidth of
> roughly 150 Mbps out of 155.5 Mbps with STS-3c/STM-1 makes the PPP over SONET
> solution very attractive.
[...]

- - - - - - - - - - - - - - - - -