Re: [DNA] [Announce] DNA solution framework I-D

[Greg Daley <Greg.Daley@eng.monash.edu.au>]

Hi Erik, 

----- Original Message -----
From: Erik Nordmark <Erik.Nordmark@sun.com>
Date: Sunday, August 1, 2004 3:55 pm
Subject: Re: [DNA] [Announce] DNA solution framework I-D

> 
> [Catching up on email backlog]
> 
> > It may be worth mentioning the RS implosion problem.  How do we 
> allow 
> > nodes to solicit quickly when there is a possibility of 
> synchronzation 
> > between multiple nodes. The document talks about one of the tasks 
> being 
> > to get an RA quickly, and so dealing with this may be considered 
> to be 
> > part of the promlem of getting a fast RA, but maybe it's worth 
> mentioning.
> It would be good to understand how the L2 behavior can limit
> the RS synchronization.
> Does anybody have an idea how this would work with IEEE 802.11 and 
> cellulartechnology as an example of the L2 effects:
> - a train full of hosts move along the track
> - there are APs/base stations along the track
> 
> How much would the L2 handovers be spread out over time?
> If the hosts send a RS immediately when L2 provides a 'link-up' 
> indication,would that effect the L2 handover performance for the 
> other hosts?

I have an unpublished simulation result for 
30 simultaneous MN handovers and RS/RA on
802.11 wlan, but don't have the test results
for the impact (of and) on existing fixed
wireless hosts due to such an arrival.

There's a skewing of the L2 associations based
on probing and missing the probe response at L2.
there are loads of assumptions though about 
scenario.
Also, the exchange of base (in this case wep/null)
authentication serializes operations.

I've got some figs.

I can repeat the results for any number of MNs,
limited by the VMem on my simulation host.

Greg

>