[EMAIL PROTECTED] wrote: > On 27 Mar 2002 at 22:43, Eugene Leitl wrote: > > >>On Wed, 27 Mar 2002 [EMAIL PROTECTED] wrote: >> >> >>>I don't recall ever having read of this type of structure before, >>>but it seems so obvious that I'm sure it's been discussed before. >>>So is there a name for it? Does anyone use it? has it been >>>shown to be utterly worthless? >> >>You don't mean something like this: >>http://www.perfdynamics.com/Papers/Gnews.html do you? >> > > > Yeah, I think what I was describing was more or less what > they call a hypercube, or maybe just a cube. > I'm not one of those people that > can actually envision multidimensional structures, so I only > know "this is a 4-cube" if I see the coordinates. >
Ok, somebody has to say this, and I wish it were a topologist, mathematician, or somebody else working in the field, but it's not a hypercube or cube, except maybe in a fractal sense. I get the nagging feeling the following paragraph overly complicates something simple, but I can't think of any other way to describe your network in geometric terms. Anybody feel free to correct or elucidate. Here goes: You have to start one dimension lower, with a triangle. Each vertex also has branches to 295 additional nodes, that is, is a member of 3 fully- connected 100 node subnets, the other nodes of which are also each vertices of 2 other triangles. There are no edge nodes as in a lattice. The constraint is that any given node is shared by exactly three triangles which have no other nodes in common, which seems intuitive given that there are also no leaf (terminal) nodes. Again, I'm not not a mathematician, so I don't know exactly what you'd call this. Geodesic, Hettinga says, and he's right, but it's not any regular figure or solid I know. It has tree-like properties but is obviously not a tree, since there are links between nodes at what would be the same level. In terms of practical considerations, network diameter is 3, and minimum connectivity is 8 (if you count routes with common links) at the 3 hop level, which you'd probably want to use, with a fallback to longer routes on retries. Unless you're trying to discourage tracing or something.) It's a highly redundant, fault-tolerant network, and you're also right that duplicates aren't going to be a problem. However, there are only 297,000 links (if I'm counting right) among the 1 million nodes, and they're probably going to get saturated real quick. jbdigriz jbdigriz
