Talk:Content-addressable network

Has anyone implemented this on a reasonable scale? Specifically, are there any public file sharing networks that use this?


 * Not to my knowledge, but see distributed hash table for other DHT designs that are in public use. --Nethgirb 02:30, 26 January 2007 (UTC)
 * Does any implementation exist at all?--AndreasBWagner (talk) 02:12, 11 February 2009 (UTC)

Completely Logical?
Is "Completely Logical" being used as a technical term here? If so, could a definition be provided? If not, could the sense in which it's completely logical be clarified a bit?

I'm not sure if I should be reading it as "It totally makes sense to have a d-dimensional space, even if it doesn't make sense to you" or more like "this d-dimensional space is seperable, not Hausdorff, completely logical, and is isomorphic to..."

Bsradams 01:09, 26 July 2007 (UTC)


 * I like your first definition best. :-) However, what the author of that text meant is that the the d-dimensional space is completely virtual or abstract: the nodes are not physically arranged according to a d-dimensional space, but each node is given a "virtual location" in that space, and the pattern of overlay links (TCP or UDP connections) is constructed based on the virtual neighborhood of a node.  This use of the word "logical" to mean "virtual" (or "it's not the way things physically are, but it's a useful way to reason about them") is fairly common in CS research but I don't know where to find an authoritative definition. You are right that this should be rephrased for the Wikipedia readership. --Nethgirb 02:51, 26 July 2007 (UTC)


 * I agree with Nethgirb. "This d-dimensional coordinate space is completely logical." -- I'm pretty sure the original author was alluding to a logical address in some address space. I changed that sentence in an attempt to clarify. --68.0.124.33 (talk) 17:03, 6 November 2008 (UTC)

Graphics
I think a graphic explanation of a node joining and leaving the network would really help to understand CANs.--Littleant (talk) 13:35, 9 June 2008 (UTC)
 * Agreed. Wish I knew more how this works.  This page, found through Google for DHT's in general, seems to have some comparisons of various methods, and mentions Hypercubes (what I think CAN uses to organize the network) https://www.cs.cmu.edu/~dga/15-744/S07/lectures/16-dht.pdfEternalStudent07 (talk) 06:35, 20 September 2018 (UTC)

Where is the "content" and where is the "hash"?
From reading the article, I couldn't really see where the "content addressable" part fit. All I figured was that it is a way to dynamically map zones in this arbitrary coordinate space to IP addresses. How is this coordinate space used? How is the whole thing used to address content? —Preceding unsigned comment added by Helder Ribeiro (talk • contribs) 00:33, 23 July 2009 (UTC)

Wish there was a little math for comparison with other algorithms
Without reading all of this article, I don't have an easy way to compare this method to other DHT methods. Things like storage usage per node (how many connections must each keep track of) and typical/worst case search hops would be helpful.EternalStudent07 (talk) 06:37, 20 September 2018 (UTC)