Doug, direct this question to fumble:
Currently we are transmitting Point to Point, on IP, on ATM, on Sonet. With the availability of channels , OWAD
capability and optical cross connects, it will likely be IP to Fibre directly and channel selection depending on the
QoS demanded by the client.
How will this scenario affect the demand for the next generation ATM switches.?
I am not an expert in this field, however you seem to oversimplified the subject thinking IP to Fiber is a solution for everything.
It will take years to archive Qos in pure IP world, if any, and in meantime big telcos are desperate for QoS broadband.
ATM can do IP over ATM infrastructure, can manage and scale in proven way.
If you add more complexity to IP to have Qos you end up with the same complexity as ATM, but ATM works were IP with QoS will need to prove itself.
Have you the seen?: SBC/Pronto is based on ATM architecture.
BT is basing his ADSL and IP services over ATM.
Sprint just yesterday stated that is moving toward voice over ATM.
These are big decisions which are realized as we speak.
We are talking about billions of $, which is not going to be reverse b/c some Jon will promise IP with QoS.
Big routers which can carry ATM cells and ATM switches with
IP are coming together to common platform.
Just my opinion.
Zbyslaw
|
