On Fri, 14 Jun 2002, u Uler wrote: > When you say, "it should not be possible to stop Freenet by overloading," do > you mean that even if there are too few permanent nodes, the Freenet > protocol should be designed so that the network isn't stopped? Or do you > mean that we need more permanent nodes to make the network more stable?
What he's saying is "For some reason, it's taking 50-60 generally high end machines on dedicated connections to do the work of one 386-33 on a 9600 baud modem." Something is very broken, causing nodes to overload very quickly (and start timing out, thread loss, etc). A number of people are hoping it's blocking IO vs Non-Blocking, since that's fixable. For one, we can cut down threadcount a LOT if they don't have to sit on each job waiting for the other end to answer (who's waiting on someone else, waiting on someone else, etc). This is nowhere NEAR the problem, but just perhaps if everyone's not always overloaded, querys will start getting answered.... and we can start hunting for the real cause of breakdowns. --Dan _______________________________________________ devl mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl
