Re: ||ism

[manor]

> >Still, the promise of threads inside the language is powerful, as is
> >the paradigm of one thread-per-object. If anyone knows of simulation
> >systems that make use of these features, it'd be interesting to hear.
> >
>
> I have to double check, but I think the JavaSim simulation environment in
> Java (and his C++ "cousin"):
>
> http://marlish.ncl.ac.uk:8080/JavaSim/
> http://marlish.ncl.ac.uk:8080/C++SIM/
>
>
> supports this ability.
>    SWARM is agent based not only object based but there are definetely good
> ideas there.
>
> Also check:
>
> http://siesta.cs.wustl.edu/~schmidt/JACE.html
> http://siesta.cs.wustl.edu/~schmidt/research.html

Another good one is:

  http://www.dcs.ed.ac.uk/home/hase/simjava/simjava-1.0/

It has a short paper in it:

 "Using Java for Discrete Event Simulation" / R. McNab and F.W. Howell (1996)

     (http://www.dcs.ed.ac.uk/home/hase/simjava/UKPEWpaper.ps)

With regards to a true SMP version of the JVM apparently this does it:

  http://www.sun.com/solaris/products/javavm/

Having said that, I personally believe that in terms of shear speed,
Swarm is much more likely to get speed-up going the ObjC thread route
than through some sort of Java port (both in terms of port-time and in
terms of run-time). I was simply thinking that with an open implementation
of the ObjC runtime (namely Gnu ObjC) one could include some of what is
good about java into the ObjC language.

Manor.


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================