> I honestly don't know exactly what you are supposed to see. I think one or > more peaks would be good. I'm not sure that more than more one peak is any > worse than only one, but I will defer to the people who know more about how > the routing works. > > I was seeing two clear peaks on my node about 10 hours ago. Now it looks > flat :-(.
I've got 2 nodes running and I cant say I see any patterns.. Ian said in that lecture that is linked to from the webpage that the keys are compared based on lexical likeness. I'm not sure exactly what that is but wouldn't it involve more than the first nibble?? Node A is quite busy (It's on hawks ref list): Histogram of keys in in fred's reference table Nov 28, 2001 10:27:29 PM keys: 512 skipped: 0 0 |==================== 1 |=============================== 2 |================================== 3 |============================ 4 |====================================== 5 |======================================== 6 |====================================== 7 |============================ 8 |======================================== 9 |===================================== a |============================= b |========================= c |===================================== d |========================== e |=============================== f |============================== Node B is very alone in this world: Histogram of keys in in fred's reference table 28.nov.01 22:31:12 keys: 130 skipped: 4 0 |========== 1 |================ 2 |========= 3 |========== 4 |============ 5 |=========== 6 |========== 7 |======= 8 |== 9 |===== a |==== b |===== c | d |== e |=============== f |============ -- | Are ?rseth | elrond at idi.ntnu.no | http://ctemp.dhs.org/ | | tlf: 73886876 | mob: 410 00548 _______________________________________________ Devl mailing list Devl at freenetproject.org http://lists.freenetproject.org/mailman/listinfo/devl
