The IP Over NBMA (ION) Archive

[Pankaj Aggarwal Dadu <dadu@bhel.ernet.in>]

[Please ignore, if this point has already been discussed before; in
 that case I'm sorry for wasting the bandwidth & time.]

...NHRP scalability...

Can the NHRP replies be based on the availability-based-on-load-
computation further down the line [where the querry is heading to].
i.e. before forwarding the NHRP querry, can we do the load computation
and then return the NHRP reply in affirmative to the originator ?

As probably happens in mailing-lists or news-groups whose subscribers
are present worldwide and are good-enough in number, that a new
request for subscribing to the list is handled by not putting the
new subscriber right at the "root/point-of-origination" of the 
mail-list/news-group but the subscriber can be added at a 
distribution-point/nearest-mail-coordinator to that subscriber. 
i.e. Once the request from the person 'A' travels towards the "root" 
point, then the e-mail address of 'A' can be added at say the person's
organisational-mail-gateway-machine.

The traffic/circuit-setting involved in initial transfer of request
for subscription towards the root, and facing a constaint in setting
up this circuit itself, will be eased out *in comparison to* a 
situation where not only the joining request but the rest of information
is also flowing from "root" to all subscriber. This is probably a
repetion of analogy of distributed server or tree-architecture concept

but the point is can NHRP protocol send reply on load-computation basis.
So that if the short-cut can't be made to the destination directly, at
least it could be made to *some-point(one/two/few hops short of 
destination)* along that path.

Regards