[net2-wg] ttx slides draft

[Omprakash Gnawali]

> It would be good to quantify what caused those failures. We've also  
> seen experiments with 99%. So are those lost %s due to short-term  
> link variations, loops, or what?
> 
> Phil

For the 55 node experiment, there were 195200 pkts sent, 192468 pkts
received, that is 2732 lost pkts. There were 862 queue full events so
that many pkts lost due to queue overflow. There were no FAIL events
so no pkts were dropped due to bad link to the parent. Most likely
rest of the pkts were lost because of link failures - acked but not
received. I traced a few pkts in some other traces and saw this
behavior.

- om_p