Forgot your password?
typodupeerror
The Internet

Self-Assembling Networks 112

Posted by michael
from the sysadmins-out-of-a-job dept.
prostoalex writes "Researchers from Humboldt University found a way to build self-assembling networks. By emulating the behavior of ants and insects the team, which is led by Frank Schweitzer, demonstrated a simulation where agent-based architecture was able to quickly assemble itself into a network and quickly react to a broken link or damages. Schweitzer's research papers are available off his personal Web site. The scientific paper referred in the original article, Self-Assembling of Networks in an Agent-Based Model is available off Cornell server."
This discussion has been archived. No new comments can be posted.

Self-Assembling Networks

Comments Filter:
  • by Maradine (194191) on Thursday March 27, 2003 @09:06AM (#5606099) Homepage
    My network team looks *just* like a swarm of ants when the network goes down.
  • Enormous Benifit (Score:1, Interesting)

    by Anonymous Coward
    Imagine if you will being able to create and configure a LAN using technology like this? How long till we see it in Linux.

    Setting -> autolan configure -> select yes -> give network a name -> done!
    • That sounds more like a windows-networking-thing. In linux i guess it'll be a configure file with at least 100 lines, exluding the comments.
      • Yes - and the only one you need is at the top.
    • I think it has the best immediate potential for things like ad-hoc 802.11 networks that consist of live cars on the freeway, and buildings nearby.

      Traffic jams would be ironic...your data could be moving faster than you. :)
    • That resembles IPv6 router advertisment and node autoconfiguration to me. Actually you don't even need to select yes, just feed the router with a simple configuration (and it _is_ much less than 100 lines ;-).

      I think this article wasn't exactly focused on IP-layer networks, though.
    • --This could be a massive security hole, too. Imagine the government van in your neighborhood getting an IP in all the confusion, and you never know it.

      --To me, this kind of network assembly is irresponsible. Give me static internal IP's any day. Then at least you know exactly which boxes are on your net.
  • by routerwhore (552333) on Thursday March 27, 2003 @09:09AM (#5606117) Homepage

    "this network looks like a bunch of spiders having an orgy" has new meaning...
  • by wiggys (621350) on Thursday March 27, 2003 @09:10AM (#5606122)
    Taking this idea one step further, what if each computer node on the network was given a basic set of rules so that it emulated a bunch of brain cells. Would the network self-organise to create some sort of intelligence?
    • Would the network self-organise to create some sort of intelligence?

      Ignoring the ongoing debate about what is or isn't intelligence, they have been doing this sort of thing to a limited extent in software for a number of years. Neural networks are actually used for specific applications. Don't expect your network to suddenly gain an IQ of 200 though... ;)
    • by Neuronerd (594981) <konrad@@@koerding...de> on Thursday March 27, 2003 @09:24AM (#5606188) Homepage

      It turns out that already today all successful applications of socalled "artificial intelligence" are self assembling.

      In the first approaches to artificial intelligence [mq.edu.au] people used programming languages to obtain systems that generate intelligent or at least apparently intelligent behavior.

      All newer [utexas.edu] approaches to artificial intelligence start with a large number of very simple units that, learning from data from the real world, develop specific patterns of connections. Many models even develop their own structure in such a way.

      From my perspective is intelligence as well as artificial intelligence only possible in a system that can self-structure.

      • Nicely said. Limited self-modification capability is intrinsic to a mind.
      • This would be patent nonsense, if the statement itself had any real meaning. First of all, what is meant by "artificial intelligence", "successful", and "application", in this context.

        And what does "self-assembling" or "self-organizing" mean, really? The utexas link is pointing to a bunch of machine learning stuff (I research and publish in AI, sometimes in machine learning) that is frankly quite out of date (no kernel machines, SVMs, or any recent clustering techniques). Unsupervised learning can be se
        • I agree that we might use different meanings of self-assembly. Yes for me PDP (Parallel Distributed Processing) is some kind of a self-assembly. And so are support vector machines and graphical Bayesian networks.

          There are various conceptions of what Bayesian Models actually do. In some cases, e.g. Mixture models, you can easily interpret them in the sense that each model "tries to explain what it can" while at the same time interacting with other "agents" or models about which inputs it is responsible f

        • This would be patent nonsense, if the statement itself had any real meaning. First of all, what is meant by "artificial intelligence", "successful", and "application", in this context.
          That depends on what you mean by "is".
          (c) Bill Clinton
    • If a node came across somthing it didn't understand or was unsure about it could ask the other nodes for their opinion.

  • okument nicht gefunden

    Die von Ihnen gewählte URL ist auf unserem
    Server nicht bzw. nicht mehr vorhanden.

    Gehen Sie bitte zur AIS-Homepage oder
    benutzen Sie das Navigationsmenu links.

  • by Lolaine (262966)
    My girlfriend was going to study Networks Cabling and Construction (wires , switches ...). It will be funny to say her she will be messed with ants and spiders ... I think you will hear her scream from USA :D
  • i have nothing informative to say.. but hell, i think this is really great software-engineering development.
  • For now I stick to OSPF. And it is not centralized also. And so are BGP, RIP an ISIS.
  • by MeanE (469971) on Thursday March 27, 2003 @09:16AM (#5606156) Homepage
    "We are your network...ect..ect...we will adapt"
  • this isn't news (Score:2, Informative)

    by potaz (211754)
    Last Update: 18 March 1999
    The article was posted to his web site in 1999 and this is front-page stuff? And the article itself was published in 1997. Stop the presses!

    • Where did you find this? The main document linked to dates from March 26th, 2003, and the reports mostly have dates around November 2002. Maybe not exactly news, but not as outdated as you seem to suggest.
    • The method they talk about is some form of ant-system optimization, and it's most definitelty not new... I think there was a recent article that talked about that (swarm intelligence), but I don't feel like searching the link...
  • Now it seems that my time in college is wasted. No need for network admins.
    • That's what I thought, then I looked at the patch cabinets...this software'll have to get much better before it can sort out a cable that plugs into 3 different ports :S
  • by Neck_of_the_Woods (305788) on Thursday March 27, 2003 @09:20AM (#5606175) Journal

    HUB, "MALFORMED PACKET!!!! AHHH!!!! - HELP HELP HELP! I am lost!"

    Router "Calm down, this is nothing compared to the broadcast storm of 93. Everything will be alright."

    HUB, "Thank you,"

    Router "These simpletons, when will they ever learn just to ignore that packet."

    ala - bugs life.

  • "By emulating the behavior of ants and insects..."

    (I wonder who played the Queen...)

    I didn't know ants were this advanced! This must be the final proof that indeed insects are super-intelligent aliens come to earth to eat our... ehm... sugar-water... If only we can harness this power elsewhere! Maybe we should try milipede power-plants next... All that static electricity from all those legs must be harnessed!
    • Damn, that flies in the face of my years of belief that ants are really the tanks of a much smaller super-intelligent being or wait was it that ants were the only appendage we can see of another dimensional being. Nothing like a slashdot networking discussion to ruin my years of ant theory.
  • Wasn't Skynet [imdb.com] supposed to be self-aware like 5 years ago?

    I mean, it's 2003, and we don't even have systems that we can't leave alone over the weekend. Where's the AI that's supposed to do all of the thinking for us, so we can actually get some free time? [Okay, there's that little problem with it trying to kill off all humans, but well, I'm sure they'll fix that in release 2]
    • First of all, I bet there's a logical explanation for that, watch the new terminator movie (it's supposed to be released somewhere in june) and there's no such thing as skynet ofcourse, you are safe.

      Where's the AI that's supposed to do all of the thinking for us, so we can actually get some free time?

      the problem is that everytime ai comes up with a new findings it's quickly adopted in all kinds of automation processes, people don't consider it ai anymore when they know how it works: "hey, that's not ai
      • the problem is that everytime ai comes up with a new findings it's quickly adopted in all kinds of automation processes, people don't consider it ai anymore when they know how it works: "hey, that's not ai, it's just a mathematical formula that does things this way or that way". Many people don't realise how broadly ai is being used...

        This brings up an interesting point. When many people talk about some self-aware 'evil' computer systems they generally think of some large project having gone wrong (i.e.

    • Actually, no, they've decided they're not going to fix it http://slashdot.org/article.pl?sid=03/03/27/193025 6
  • ...and then all the system administrators disappear.. But then again, most sysadmins arnt even around their networks. instead they are answering calls from room 0139 because her computer can't boot up.
  • I am Captain Network!

    *theme song*

    Captain Network,
    He's our hero,
    gonna cut packet loss
    down to zero!
  • by henrod (661974)
    Anyone read Michael Crichton's new book, Prey? This kind of thing is a little scary after reading that one!
  • This article is beyond ridiculous. It is more like a pointless press release from Dr. Seuss than actual info.

    A node does this, then it does that, that somehow attracts other nodes doing something else, and POOF, the world is a great place to live in once again...

    Give me a break. I'd rather read about magic, self-healing, server pixie-dust.

    On a similar note, look for Dr. Seuss' latest book in stores soon: "One Node, Two Node, Red Node, Blue Node"
    • Take a look at the PDF the supplies the mathematical fomulae and diagrams. Rather convicing.
    • This article ... is more like a pointless press release from Dr. Seuss than actual info.

      Here's Dr. Seuss' explanation of how self-assembling networks handle errors:

      If a packet hits a pocket on a socket on a port,
      And the bus is interrupted as a very last resort,
      And the address of the memory makes your floppy disk abort,
      Then the socket packet pocket has an error to report!

      From here [sydney.net].

  • Potential (Score:4, Interesting)

    by perspex (635004) on Thursday March 27, 2003 @09:41AM (#5606288) Homepage
    This could be really cool for ad-hoc wireless networks.
    • How is this any better than existing ad-hoc network protocols?

      Here's a paper [uci.edu] that compares (in the context of wireless networks) DSDV-SQ, TORA, DSR, and AODV-LL protocols for how well they make use of shortest paths, number of packets successfully delievered, ability to deal with dropped nodes/connections, and routing overhead (as either packets or bytes).

  • Anyone who does knows this is just a step before the evil computer AI infecting all of the other computers in the world and setting about to destroy mankind. I will rise up to defeat this terrible menace right after I find a girl with blue hair and eyes the size of dinner plates.
  • A lot of the web service intermediaries have these kind of capabilities...
    You plug in their agents on the network and they slowly become aware of each other through message exchange. When one section of the network goes down the agents talk to each other to figure out which agent can be used to relay a message around the broken link.

    It's really wierd to be up in layer 7 and see the same modeling of behavior of lower layers in the stack...
  • If it was real ants there'd be none of this "You have the blue nodes and we'll have the red nodes" niceness.

    The blue ants would be killing the red ants and vice versa - and the scent given off by the dying ants would attract more ants to the area until there was one hell of a war going on for territory (nodes). With the winners getting better connectivity for their network. And the ants would quickly specialise into scouts, soldiers and queens (to reinforce the army).

    Come to think of it, that'd be much
  • by PHanT0 (148738)
    I've help test, write and measure the success of code that does exactly this. Our implementation uses the ants to collect routing latencies and update routing tables. It's actually surprisingly efficient and deals very well with downed nodes depending on your timeouts for downed nodes, etc.
  • Been there. [newscientist.com] Done that. [ulb.ac.be] These types of algorithms are not exactly new, and what this paper describes is no more "self-assembling" than any other distributed routing/discovery protocol - examples of which have existed for over twenty years. Of course, lots of things are new to the Slashdot editors that are old to the rest of us.

  • The somewhat self-assembling nature of P2P networks got me thinking about little swarms of tiny clean up robots. Instead of a hunanoid robot, it seems what would be more useful and simpler for things like household or even commercial maintenanc is a network of small robots relying on each other for various specialized functions sort of like cells in a larger organism.
    It seems like you almost have to forego the android approach and go this way to get automated maintenance workers financially feasible
  • by CAIMLAS (41445)
    People have been tlaking for years about how kludged together the current internet infrastructure is; my question is, might something like this make for a feasable replacement, or at least a suppliment to what is already out there? I can see this being very useful indeed. You'd be able to de-centralize the root servers, and have them be distributed from
  • by ajs (35943)
    I don't know a lot about the state-of-the-art in the area of network discovery/repair other than what I know as a socket-programmer and sysadmin, but I'm wondering if someone who does know can point out the differences between, say, this research and Apple's Rendezvous [apple.com] (not to be confused with Tibco's product by the same name [tibco.com])?

    It seems to me that the basic goals are similar, but with Apple focusing more on the engineering side of solving a user-problem rather than passing the point of diminishing returns o
    • And the answer is... RTFA ;-)

      My bad, wrong kind of "network". While the idea may be applicable to computer-interconnectivity, that's not what this is about, and I would have known that, had I read the article.

      Thanks anyway, all!
  • Why am I not surprised that the people in Humboldt found a way for themselves to do less work?

    I wonder if we'll see a press release from them later saying they've designed something to emulate a particularly famous local plant. ;-)
  • IT Guy: We're being nailed off our ABC uplink with a denial of service attack!
    Manager: Well, we still have our DEF uplink in reserve. Drop everything from ABC!
    IT Guy: Okay, much better now.... oh wait, the network reassembled to attack our DEF link!
    Manager: I think I'll be cavorting in Arizona for a while...
  • where agent-based architecture was able to quickly assemble itself into a network...


    Good god, didn't you people learn anything from The Matrix?! Agent-based architecture is the most dangerous type of computer system you can design!


  • I'm of the opinion that spammers represent an infection of the net and that we are watching how the network is adapting to fight it off.

  • Sun was right. Now, the network really is the computer.
  • by varjag (415848) on Thursday March 27, 2003 @12:19PM (#5607683)
    The paper is indeed very interesting and innovative, but keep in mind that it is very far from being suitable to embed into your next 802.11 adapter.

    While this approach is indeed appealing, it has still some drawbacks, e.g:
    - generally, you can't tell what your topoligy your network will end up having, so forget about architecting one
    - it does not guarantee that all your nodes will end up being networked within a fixed number of attempts (see the fig. 3 in the paper)
    - it tends to require significant redundancy of interchangeable nodes to function well

    Such approach can work well, say, for military field communications, but would be clearly suboptimal for building a corporate network.

    And of course, as most of agent research, this is still too far from established technology ready for production.
    • disclaimer: I have not (yet) read the paper, just the first linked article.

      But the problems you describe wouldnt be problems for say an ad-hoc p2p wireless network, with each node forwarding for others.

      - generally, you can't tell what your topoligy your network will end up having, so forget about architecting one

      It doesnt really matter, as long as it works and can get packets from a to b, to me it seems the whole point is that you don't need to architect the network.

      - it does not guarantee that all y
  • Perhaps I'm not understanding something, but how do the researchers intend on having the nodes in a real life network emit 'pheremones'? The only application I can see for that is self organizing wireless hubs. But for hard wired networks it does not seem to make sense, since there is no way to estimate the ease of connecting two points based on distance alone.

    Overall, this article only seems to apply to wireless networks. An interesting, but limited, usage.
  • Andrew Harvey wrote basically the same exact article about ten years ago. The only difference was that he attempted to relate this concept to spirituality. The guy who wrote this article, Frank Schweitzer, has been accused in the past of taking ideas from other Conihilimous (The study of self structuring autonomous networks) experts such as Andrew Harvey and the asian mathemetician Niyh B. Tihtzen. just my two cents
  • or does this sound remarkably like something leading up to Michael Crichton's Prey?
  • Admin: WHO in God's name sprayed for bugs in the office??? You just killed half of the network!!!

Biology is the only science in which multiplication means the same thing as division.

Working...