On Monday 29 Aug 2011 20:34:01 Ian Clarke wrote: > On Mon, Aug 29, 2011 at 2:21 PM, Matthew Toseland <toad at > amphibian.dyndns.org > > wrote: > > > > > - What is the average load reported by responses this node > > forwarded, per > > > > remote node > > > > > > Ahhh, this one could be interesting - you could use it to penalise nodes > > which spam excessively. > > > > Actually, thinking about it, I don't think this will work. Requests are > > usually distributed evenly across the keyspace, and downstream nodes only > > know they are from this node - not which previous node they are from. So the > > pain will be shared across all the peers sending us requests, including our > > own local requests, and won't be identifiable as due to any single node. You > > have to do it by volume, not by reported load. > > But if a node is being abusive won't the nodes its requesting from tend to > have a higher than average load?
If its requests differ from the specialisation of the node in general, so are routed to a particular subset, then they might be different. But this is not necessarily true if it is a deliberate attack. > Perhaps you are right though, but this is > the kind of thought-process we need. To prevent a node from using an excessive proportion of our capacity, we need to *measure the proportion of our capacity that the node uses*. This much is straightforward IMHO. And that's the basis of fair sharing between peers. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20110831/ded4b5e9/attachment.pgp>
