>> Surely the routing will still work as long as a given node uses the same >> algorithm for routing inserts as it uses for routing requests? >No, because each node's perception of speed would be slightly different, >so the routing would be non-deterministic.
If the node's measure of the connection speed of each link is fixed (ie, it does not adapt the figures to take account of congestion), it will route the same insert to the same node every time. That's deterministic behaviour at the node level. As for deterministic behaviour at the network level, that emerges from deterministic node behaviour (we hope). Why do all the nodes need to agree on their connection speeds? They don't all agree on the location of any given document, but it can be found nevertheless because each node routes consistently. Over time, documents with similar keys cluster together *regardless of the measure of similarity*. I think the only necessary constraints on routing algorithms are those I outlined in my reply to Oskar: a node should route every insert to somewhere; a node should route any two inserts with the same key to the same place; a node should route an insert and a request with the same key to the same place. This rules out adaptive routing, but it doesn't rule out weighting the links with fixed weights. Michael _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
