Slashdot Mirror
Ants in your P2Pants
Tim Finin writes: "Anthill is a framework being developed at University of Bologna to
support the design, implementation and evaluation of P2P applications,
viewing them as instances of Complex Adaptive Systems, typically found
in biological and social sciences. In Anthill, desired properties such
as resilience, adaptation and self-organization correspond to the
"emergent behavior" of the underlying CA system. An Anthill system
consists of a dynamic network of peer nodes; societies of adaptive
agents (ants) travel through this network, interacting with nodes and
cooperating with other agents in order to solve complex problems. The source code for Anthill v1.0 is available for downloading. MORE on this is at ebiquity.org."
Project Overview
Anthill is a framework aimed at supporting the design, development and analysis of peer-to-peer protocols and algorithms. The goals of Anthill are to: (i) provide an environment for simplifying the design and deployment of new P2P systems, and (ii) provide a "testbed'' for studying and experimenting with P2P systems in order to understand their properties and evaluate their performance. Anthill is based on the multi-agent systems (MAS) paradigm. A MAS is a collection of autonomous agents that can observe their environment and perform simple local computations leading to actions based on these observations. The behavior of an agent may be non-deterministic and its actions may modify the agent environment as well as the agent location within the environment. What distinguishes MAS from other agent models is that there is no central coordination of activity. MAS often exhibit a property called swarm intelligence whereby a collection of simple agents of very limited individual capabilities achieves ``intelligent'' collective behavior. In this manner, they are able to solve problems that are beyond the capabilities or knowledge of individual agents. For example, ant colonies, which are natural instances of MAS, are known to be capable of solving complex optimization problems including those arising in communication networks. In our opinion, MAS can be profitably adopted for the design of innovative peer-to-peer algorithms.
Anthill uses terminology derived from the ant colony metaphor. A P2P system based on Anthill is composed of a network of interconnected nests. Each nest is a peer entity capable of performing computations and hosting resources. Nests handle requests coming from users by generating one or more ants --- autonomous agents that travel across the nest network trying to satisfy the request. Ants interact indirectly with each other by modifying their environment by updating the information stored in the visited nests. This form of indirect communication, used also by real ants, is known as stigmergy.
Developers wishing to experiment with new P2P protocols need to focus only on writing appropriate ant algorithms using the Anthill API and defining the structure of the P2P system. The current version of Anthill includes a simulation environment to help developers analyze and evaluate the behavior of P2P systems. Ant algorithms are executed in simulated nest networks and various properties such as performance, fault-tolerance and availability are measured. Simulation parameters, such as the characteristics of the network, the type of ant algorithms deployed, the kind of workload presented to the system, and the properties to be measured, are easily defined using XML.
Ah, a fellow BITNAUT, eh?!
Nice to see a more practical application of ant systems. The past year I have been working with ant systems in a more academic setting: optimizing dynamic problems (problems that change over time) using ant systems. Travelling salesman to be more precise, but my salesmen (sales-ants?) encountered traffic jams. :)
The research can be found here
Next idea: ant based routing. Get rid off BGP, use ants
Sooner or later more of our Technological systems will emulate nature. For those who haven't read Kevin Kelly's masterpeice - Out of Control should do so. Our tecnhological future will become more and more alive as time progresses.
www.enthea.org
"Anthill is a frameworkDING! being developed at University of Bologna to support the designDING! , implementationDING! and evaluationDING! of P2PDING! applications, viewing them as instances of Complex Adaptive SystemsDING! , typically found in biologicalDING! and social sciencesDING! . In Anthill, desired properties such as resilienceDING! , adaptationDING! and self-organizationDING! correspond to the "emergent behavior"DING! of the underlying CA systemDING! . An Anthill system consists of a dynamic networkDING! of peer nodes; societies of adaptive agentsDING! (ants) travel through this network, interacting DING! with nodes and cooperating with other agents in order to solve complexDING! problems. The source code for Anthill v1.0 is available for downloading. MORE on this is at ebiquity.org."
Ladies and gentlemen, we haaaave a winner..
Bowie J. Poag
Draging a hazy memory out of my brain... I'm sure that I saw an article about 'ant-modeling' in New Scientist quite a lot of years ago (at least 2/3 years ago) where they were using ant's and their scent trails (modeled of course...) to find the optimum routes across a network.
I think it worked as followed:
Put a bunch of ants on the start node, all they know how to do is travel from node to node and they know when they reach the end node.
Each ant will go down a path to a connected node, except that they will not backtrack to their previous node. Each of these ants is releasing an electronic scent and they are more likly to go down the path with the strongest scent.
Repeat with several thousand ants and you should have your optimum path across the network.
I seem to recall that the article said that BT (British Telecom) was looking into this as a way to optimise the routes that a phone call takes from switching station to switching station... but it was a very long time since I read it.
[No ants were harming in the writing of this post]
Those words do have meaning. And while they can be used for empty hype, there existance alone is not an indicator of vapiedness.
And besides how can you consider "implementation" and "complex" buzzwords but not 'dynamic', 'peer nodes' or 'downloading'?
autopr0n is like, down and stuff.
...which is on a similar theme, LGPL and possibly more complete.
http://www.jtrix.org
Unlike Anthill we're addressing a wider variety of needs (in fact, any kind of app you want to scale up), so you can build all kinds of things on top of Jtrix, from traditional P2P (Freenet, Gnutella, etc) to other apps needing scaling and security (Web-based mail, Passport-a-like, and so on).
We've been at it 18 months now and have things like an HTTP server and servlet engine which work in this environment. It's great to see these kind of frameworks growing in popularity.
Nik
Anthill is full of bugs!
-
- - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
They have the right idea when it comes to studying large distributed networks. Network emergent properties is absolutely the most interesting thing in that field, and I strongly believe that it is a much better model for trying to handle highly dynamic networks than trying to route within a rigid structure as lot of object location networks have been designed to do.
That said, peer to peer network simulators and frameworks are a dime a dozen - anybody who was at the last O'Reilly "P2P" conference would have noticed that everybody, their mother, and their great-aunt Petrunella seems to have one. Setting up a bunch of java interfaces is not that hard, and calling a Node a "Nest" and giving it the following methods:
public interface Nest {
void request(Request request,
ReplyListener listener);
void addService(AntFactory factory);
void addNeighbor(NestId nid);
void removeNeighbor(NestId nid);
NestId[] getNeighbors();
}
is not exactly a radical new design.
What is really needed are more examples of analytical work regarding the emergence of and utility of routing properties in the known algorithms, as well as other radically new routing schemes besides Freenet. I'm glad that these people have their framework in place, but it really isn't news until they have something to show for it.
The BT Zeus project is an open source agent framework that includes lots of nice stuff (visual development / visualisation / intellegent agents etc.) See www.btexact.com/projects/agents/zeus/ for more info. I feel obliged to mention SoFAR an academic focusing on DIM (Distributed Information Managment) Agents (or at least they where last time I was there).
-- Paul
In the words of Terry Pratchett : "Anthill Inside" :)
MLT - simple and robust open source multimedia framework for Linux
This sounds a lot like JXTA.. For those who don't know about JXTA, it was originally a Sun Microsystems P2P framework project that has been released as open source (BSD license unless I'm mistaken badly). Judging by a quick scan, JXTA seems about 100x more advanced and mature as this project but I could be wrong. Can anyone who knows JXTA better confirm?
While reading their homepage, it struck me that their methodology is somewhat similar to what active networks do. So, does anybody see how active networks can be used for deploying P2P apps?
What? This is coming from the University of Baloney? Geez, how silly of a place is this? Next thing you know, there's going to be a Hacker University or something.
;)
(This is a Joke
-- Dan
Akk, that post did NOT explain what you're talking about. For the benefit of the confused I'm pretty sure he's trying to come up with a P2P network without using a central address to find the network. Looks like he wants to use a CODE-RED / NIMDA method and send "hello?" packets to random IP addresses at a rate of 1 per second looking for anyone else that happens to be running the P2P.
Randomly spamming the internet with packets is a very questionable plan. You WILL get complaints. That aside, if you *do* go forward with the random address search I'd suggest you start and several addresses per second with an exponentially decaying search speed (Multiply the delay by 1.01 for each address you test). It's a much more efficient systems. Early random tests are more valuable than later ones. If there's a significant network you'll find it faster, and if there isn't then the amount of spamming is self limiting.
In addition, his math is all wrong. First - "0.000118... probability of colliding per attempt" would requre a half million friendly P2P servers. Second he's assuming 255^4 IP addresses. He should be assuming 254^4 - and then checking approriate RFC for rurther reductions. Third - the formula for % per minute is wrong - maybe he calculated it right and wriote it wrong, because the number he got looks about right (but "right" is still "wrong" because is assumes a wrong % per test).
And a final observation - the time to "first circut" would also happen to be the time to spam every existing IP address at least once (and more likely several times).
-
- - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
"Ants in your P2Pants"
What the hell... I thought somebody has diabetes of something... Ants in your Pee to Pants.
Oh wait...
Take-off every
Will the ants be transporting electronical little pieces of leaves to their ant-nest? Perhaps they could carry any kind of food too, or files, or MP3s...
All in the name of research, of course.
This would be ideal for robot soccer competitions. "GO THE ANTS!"
Kevin
"It's not the cough that carries you off, it's the coffin they carry you off in" O. Nash
What you call bologna is called mortadella, with the difference that the latter contains chunks of fat and flavors. In comparison, bologna tastes as raw meat.
Hi,
sounds like these mad wizards from Pratchetts Disk-World series who are running a computer utilizing ants for bits (make that "electric signals") and punchcards for flow-control.
Does this count as bioinformatics? Is this already nonlinear computing?
Cheers, Lars
This is about telecommunication that might interest some of you:
1 537000/1537645.stm
/ 13ANTS.html
http://news.bbc.co.uk/hi/english/sci/tech/newsid_
http://www.nyt.com/2001/09/13/technology/circuits
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
apparently this counts as Autonomic Computing. IBM has it posted here:s earch.html
2 6
http://www.research.ibm.com/autonomic/academic/re
They also have eBiquity listed.
More autonomic on this old Slashdot thread:
http://slashdot.org/article.pl?sid=01/10/16/13142
pvcpie
(who thinks grid computing is a grail)