On Mon, Dec 10, 2007 at 07:58:23PM -0600, David Sowder (Zothar) wrote:
> Matthew Toseland wrote:
> > On Tuesday 11 December 2007 00:12, David Sowder (Zothar) wrote:
> >   
> >> The most wandering node is darknet only and you know it as Zothar70. 
> >> The more calm node is a hybrid darknet/opennet node.  That being said,
> >> neither node appears to have a favorite location it comes back to. 
> >> OTOH, if I only look at the 48 hour graph rather than the week graph,
> >> the darknet only node appears to be favoring a particular location more
> >> than the hybrid node.
> >>
> >> In any case, I've got the data in an RRD for each node, if it's useful. 
> >> Perhaps if we had a "location of the datastore" metric formed by an
> >> average or something, I could export that via FCP and track that in the
> >> RRDs as well.
> >>     
> >
> > URL for current graphs?
> >   
> Graphs as generated in the last few minutes at
> http://freenet.xzwq.net/tmp593872/
> 
> The "1w" files are for a week, while the non-"1w" files are for 48 hours.

Something interesting to plot could also be the average (sum/neighbors) distance
to your peers (on the 0-1 circle). This will show how well the node is adjusted
with respect to the position of its neighbors in the network. It would indicate
whether the node swaps to a new location due to randomness (many unlikely swap
props but finally finding one which works) or due to that peers have also moved.

/vive
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20071211/05f2e5b4/attachment.pgp>

Reply via email to