On Thu, 3 Aug 2000 01:32:48 -0700, freenet-dev-admin at lists.sourceforge.net wrote:
> >>I have been experimenting with a node for several days now and have been >>experiencing incredibly slow transfers through freenet particularly on >>larger files. It appears that several of the other nodes that I talk with >>are severely bandwidth challenged. Freenet is unlikely to be a viable >>online resource unless a method is built in to the routing process which >>lends prefference to connections with higher transfer rates. For those of >>you who are more technical than I am, would it be feasible to include a >>transfer rate metric in the data that is stored for each connection and >>could such a metric be used to direct traffic to faster servers. >If instead of the current best route strategy there would be a probability >routing, coupled with a speed value of routes. Storing the round-trip value >at each known route would add a lot more meaning the probability routing. >The probability of a routing into certain direction should be calculated >from closest routing found multiplied with that routings roundtrip value, >and the lower total the better. If a routing probability fails it should >either route into a random direction, or perhaps the second best should be >tried. Either way this should allow finding new routes to hosts, aswell as >routing more traffic into faster directions. > >I'm kinda sleepy so I hope you excuse my flood of typo's :-D It could be a client issue. The client could request the file to two or tree different nodes and then retrive it from the faster node-chain. -- Muzzle, Flatline, Zero on IRC; ICQ# 36124438 Key fingerprint: 5881 356B DDA7 115B 6FAF C529 34ED 70A8 7E52 D805 www.internations.net/it/muzzle "We are, we can, we will" _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
