From: Howard C. Berkowitz (hcb@gettcomm.com)
Date: Wed Nov 02 2005 - 10:40:26 GMT-3
At 2:37 AM +0000 11/2/05, dusth@comcast.net wrote:
>Sound like the millionaire talking here:). spent $50.00 just for one
>question topic:) I'm not Mr. Bill Gates.
That is not an image I want to keep in mind, given I have become
increasingly nervous seeing airport signs "To All Gates".
>
>-------------- Original message --------------
>
>> The Williamson multicast book around page 464 really provided that
>> 'ah-ha' moment for me about the subject.
>>
>> Ben
>>
>> cciein2006@yahoo.com wrote:
>>
>> >Does PIM dense mode or sparse-dense mode work over frame relay? I
>>tried the ip
>> pim nbma-mode command on a multipoint frame-relay interface but it
>>generates an
>> error message saying that PIM nbma-mode is not recommended for dense mode or
>> sparse-dense mode?
>> >
> >Why is this?
More seriously and on point, I suggest that the goodness -- or not --
of fitting a given multicast model to a NBMA medium may take a bit of
abstraction, and then specific design.
Multicast (not broadcast) is inherently tree-structured. Highly
scalable multicast networks involve trees of trees.
Multicast routing over NBMA is, thus, an overlay problem. If you can
efficiently overlay the multicast trees over the L2/L2.5 trees, the
method fits. If it does not overlay, then look for a different method.
Admittedly, in real-world multicast application, some models may fit
better because you use more than one FR or other NBMA tree.
This archive was generated by hypermail 2.1.4 : Thu Dec 01 2005 - 09:12:04 GMT-3