Re: Location churn: empirical data

2008/5/14 Jano <[email protected]>:
> See mine attached. This node has not been steadly 24/7 up as of late, which can
> be more typical. It bounces a bit...

This is just a somewhat far-fetched theory, but I wonder if somehow
large changes in location cause a "cascade" effect in the network.

So, one node's location is reset, but instead of it returning to its
previous position after some time, it causes one or more of its
neighbors to change position - and the effect spreads through the
network.

If this was happening, one remedy would be to only count node's
locations in the swap algorithm if they haven't had a significant
change recently.

Of course, this is only a theory...  We should simulate node resetting
to see if we can observe this kind of effect.

Ian.

-- 
Email: [email protected]
Cell: +1 512 422 3588
Skype: sanity