Re: Another interesting issue ....

[Mark P. Line]

Vladimir Jojic wrote:
> 
> I will describe a simple distributed Swarm system, just to start the
> discussion about this.
> 
> Description:
> ------------
> 
> Let's consider a system of ~1000 autonomous agents. For example. user has
> 11 PCs on a local net with i586 in each of those PCs (all machines are
> equal in the terms of the processing power).

Local net? How about if you define a TCP port for "Distributed Swarm"
and run the whole thing on an intranet or on the Internet?


> 1. Grouping agents (manually, or by intelligent load-balancer) that tend
> to communicate between them selves a lot (`friends') on the same machine.
> 2. Creating a map of agents on the other machines (based on information
> from coordinator from previous interation), so that the coordinator would
> in time become more-or-less idle.

As long as you take one or both of these approaches to keep the net load
down, I think it would be fun to try some distributed Swarming on the
Internet.


> Ideally, distributed Swarm, would be useful for the simulation of the
> large number of the complicated agents (for example based on neural
> nets), where more resources are spent on the calculating the next move,
> than talking to other agents and actually making the move (`thinkers' as
> opposed to the `talkers').

I would even go so far as to hope that the availability of distributed
Swarm would encourage modellers to go ahead and put more intelligence
into their agents -- especially when the agents are supposed to be
modelling entities they presume in fact to be intelligent.


> 1. Will I really get 10* performance?
> 
> Probably not, 5 times better is more likely (but this is still a lot
> better, hey even 2 times is better!). But in some special cases, you
> might

I think performance is just one of several relevant issues. The support
of certain kinds of collaborative work is another issue that, for me, is
more important than performance. I wonder how often the low-bandwidth
partitions might more-or-less (be made to) coincide with reasonable
divisions of labor within a team of modellers.


-- Mark

(Mark P. Line  --  Bellevue, Washington  --  <address@hidden>)


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.
                  ==================================