Re: (ASCEND) 7.10.6 software for the DSL MAX 20

[Peter Lalor <plalor@infoasis.com>]

>From: Tony Ray <tony@turbonet.com>
>
>Currently I'm providing DSL (over Frame Relay from the DSLAM) and Frame
>Relay using Cisco equipment.  The feature Cisco has that I need and use
>is Integrated Bridging and Routing, irb.  I need to be able to bridge
>several DSL and Frame Relay pvc's together for telecommuting to offices
>and interconnecting offices between communities for customers.  I also
>need to place a routable interface on these bridge groups giving each
>their own subnet and access to the Internet.  Cisco's routers do this
>quite well, but I find I am limited to 120 dlci's per T-1 interface when
>the T-1's could easily handle three times their current traffic.  I'm
>also limited to 480 subinterfaces using irb.

I think you should be able to do this with what Ascend calls BIR and 
bridge groups. The various BIR configurations (and there are several) 
are detailed in the current release notes.

I haven't used bridge groups yet, but BIR is working fine. Your telco 
needs only to set up the PVC as bridged PDU.

>In a Cisco router, every dlci can be given it's own "interface" for
>routing purposes by creating serial subinterfaces, assigning each one
>its own dlci, assigning a (separate) bridge group to each serial
>subinterface (or group of subinterfaces) and placing a bridge virtual
>interface on the bridge group.  Considering the chain of things tied
>together to make this work in a Cisco router, I wouldn't be surprised if
>something like this isn't possible in the Ascend box but that is just
>isn't clear how to go about it from the documentation.

You can do this with BIR on bridged connections and with numbered 
interfaces on routed connections.

>Also, while the Terminator 100 appears to be able to handle 1000 dlci's,
>there is no mention of whether there are limits on the number of dlci's
>per T-1 interface.  I realize the size of the LMI packets places such a
>limit (but can't recall right off what that limit is).  Also, irb places
>a heavy load on the Cisco's processor so whether or not the Terminator
>100 can handle 1000 irb's (or equivalent profiles) is an equally serious
>question.

Although the Terminator was designed to support 1000 PVCs, I don't 
believe anyone has a Terminator in the field with 1000 PVCs. So the 
question is, how much did Ascend test the box under that kind of load?


Peter Lalor           Infoasis
plalor@infoasis.com   http://www.infoasis.com/
++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>