On Monday 21 July 2003 12:04 am, Ian Clarke wrote: > > Basically taking probabilistic caching one step further. Come up with a > > ngrouting algorithm to perdict our incoming data requests, and if the > > item that is currently being processed is more likely to be requested > > then our current least likely to be requested data, it replaces it. > > I still don't understand your proposal. Please explain it in-terms of > what a node does when it receives a request.
It does nothing different with a request. When a reply comes back it needs to decide whether or not to cache it. Instead of having a fixed number of nodes caching or a set probability for each node caching a give reply, have an algorithm that predicts the probably we will be asked for that data again. If that is a higher probability than the least requested data we have, cache it. _______________________________________________ devl mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl