Hi, > Message-ID: <[email protected]> > From: Taral <[email protected]> > Date: Fri, 29 May 2009 23:21:26 -0700
> On Fri, May 29, 2009 at 10:49 PM, Kazuyuki Shudo <[email protected]> wrote: >> I recently measured the path lengths of routing results. >> >> Routing path length >> http://www.shudo.net/diary/data/20090506/path-length.png >> >> The result shows relatively long path length of Koorde instead of >> small routing table. > > Very interesting! Does this include the optimization where one expands > the neighbor set above $k$, or is it strict de Bruijn? I guess you mean base-k (k > 2) de Bruijn graph. Yes. In Koorde.java, // configuration and edge private final int digitBits; // k = 2 ^ digitBits In KoordeConfiguratoin.java, public final class KoordeConfiguration extends LinearWalkerConfiguration { public final static int DEFAULT_DIGIT_BITS = 1; // should be < 32 All the experiments I've done are based on base-2 de Bruijn graph. Kazuyuki Shudo [email protected] http://www.shudo.net/ _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
