The IP Over NBMA (ION) Archive

[Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>]

Grenville;

Your draft, draft-armitage-ion-cluster-size-00.txt, raises an
interesting point on the scalability of multicast.

But, though you mentioned CSRs in the draft, I don't think CSRs
are useful for cell-by-cell communication of best effort traffic.

So, I'm not worrying about the amount of state for multicasting
as long as it is equivalent to that for RSVP.

Then, though you wrote:

   A possible solution is for the CSR's underlying cell switching fabric
   to provide AAL_SDU-aware cell forwarding.  If segmented AAL_SDUs
   arriving from the source  Cluster  could  be buffered  and  forwarded
   in groups of cells representing entire AAL_SDUs, the CSR would need
   only a single SVC into the target  Cluster.

it will create a packet delay worse than that of packet based
relaying.

That is, if you transmit a 800byte packet over a dedicated 64Kbps
VC, packets from other sender are blocked 0.1 seconds.

If, you, instead, use 10 Mbps p-to-p VCs shared by all the best
effort traffic, packet delay is 640 usec and negligible.

For cell-by-cell multicast, it seems to me that it best to have
a separate p-to-mp VC for each sender.

Still, the amount of state necessary for VC maintainance is same
as that for RSVP.

						Masataka Ohta