On Wed, May 30, 2001 at 09:53:59AM +0200, Francisco M. Marzoa Alonso wrote:
> On Wednesday 30 May 2001 03:35, you wrote:
> > Did you see the Chord paper? It's in the Freehaven docs... I can send you
> > a copy if you can't find it.
>
> Can you do it, please?
>
> Thank you very much!!!!!! :)
It's at http://theory.lcs.mit.edu/~karger/chord.ps. Also see the Plaxton
paper at http://www.cs.utexas.edu/users/plaxton/ps/1997/spaa.ps.
They're both very complex compared to my little experiment (and I'm now
hacking my way through the encryption and filesplitting in the client
proxy).
A word on Anarcast scalability: every node knows every other node. They
retrieve the node list from the inform server, and each address is (of
course) 4 bytes. Then they SHA hash them (now 20 bytes per node) and put
them in a tree. The inform server will be the bottleneck here--100
kilobyte transfers will be required for a 25,000 node network. (And
500K+ of node memory.) I consider this acceptable.
File splitting is still up in the air. Chunks should be pretty small,
maybe 64K. I'm not yet sure what scheme I will use to generate the check
blocks. Ah well....
--
"...it must be held that third-party electronic monitoring, subject
only to the self-restraint of law enforcement officials, has no place
in our society..." Mark Roberts | [EMAIL PROTECTED]
_______________________________________________
Chat mailing list
[EMAIL PROTECTED]
http://lists.freenetproject.org/mailman/listinfo/chat
