Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Ascend Communications (ASND) -- Ignore unavailable to you. Want to Upgrade?

To: Tera Bit who wrote (57584)	11/27/1998 6:31:00 PM
From: bucky89	Respond to of 61433

Traditionally, carrier switches are not "IP" aware. That is to say they are generally switching packets across DLCI/PVC "clouds". The IP intelligence (read that the ability to send a packet across the correct DLCI) is coming from a CPE router in most cases. My understanding is that in an MPLS world, CORE switches will run standard routing protocols and LDP. Daniel, My understanding is the China Post is not a service provider, but rather an enterprise. I believe they will be using IP, and the BSTDX9000 does have the ability to read IP and switch it across a DLCI/PVC. If the CP is using this feature/capability, then they will indeed be using the BSTDX to route IP across the ATM cloud. You are right that the core switches do need a label distribution protocol in order to do SVC's, but my understanding is IP routing will be handled at the edge switches. OSPF and/or BGP can be used to communicate IP connectivity among edge switches at the edges of the cloud, but this is transparent to the core. In a 100% CBR world, this would essentially be true. As you know,we don't live in that world. Good shaping of UBR/nrt-VBR /ABR and intelligent packet discard algorithms are non trivial and improtant factors that insure cores are not overwhelmed. Daniel, traffic shaping is the job of edge switches, which is precisely my original point. PNNI, OAM, packet discard, are just 3 examples of many complex tasks a core switch does. Save the AAL part, the core typically needs to do everthing the edge does but faster. Does that make it more or less complex? I don't know. Daniel, okay your point is taken that a core switch's job is not that simple. Thanks for pointing out my over-simplification. But my original point is that essentially anybody's ATM switches can be used at the core--as long as they run a basic LDP and maybe PNNI, without losing any of the multi-service capabilities and features of Ascend's edge switches. bucky89