address@hidden: address@hidden: [lists.hpcoders.com.au: Re: [Swarm-Support] development priorities (was Re: Membership in Swarm Developmen Group)]]]

[Russell Standish]

Grrr... what have I got to do to change my email address around
here...


----- Forwarded message from address@hidden -----

X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on 
        hells-dell.boundaryrider.com.au
X-Spam-Level: 
X-Spam-Status: No, score=-1.6 required=5.0 tests=BAYES_00,NO_REAL_NAME 
        autolearn=no version=3.1.3
X-Original-To: address@hidden
Delivered-To: address@hidden
Envelope-to: address@hidden
Delivery-date: Thu, 16 Nov 2006 10:18:03 +1100
Subject: address@hidden: [lists.hpcoders.com.au: Re:
 [Swarm-Support] development priorities (was Re: Membership in Swarm
 Developmen Group)]]
From: address@hidden
To: address@hidden
Date: Wed, 15 Nov 2006 18:17:58 -0500
Precedence: bulk
X-BeenThere: address@hidden
X-Mailman-Version: 2.1.5
List-Id: Swarm Support <support.swarm.org>
X-List-Administrivia: yes
Errors-To: address@hidden

You are not allowed to post to this mailing list, and your message has
been automatically rejected.  If you think that your messages are
being rejected in error, contact the mailing list owner at
address@hidden


Date: Wed, 15 Nov 2006 23:23:51 +1100
From: Russell Standish <address@hidden>
To: address@hidden
Subject: address@hidden: [lists.hpcoders.com.au: Re: [Swarm-Support]
        development priorities (was Re: Membership in Swarm Developmen
        Group)]]
Mail-Followup-To: address@hidden
User-Agent: Mutt/1.4.2.1i

Grrrrr...


----- Forwarded message from address@hidden -----

X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on 
        hells-dell.boundaryrider.com.au
X-Spam-Level: 
X-Spam-Status: No, score=-1.6 required=5.0 tests=BAYES_00,NO_REAL_NAME 
        autolearn=no version=3.1.3
X-Original-To: address@hidden
Delivered-To: address@hidden
Envelope-to: address@hidden
Delivery-date: Thu, 16 Nov 2006 09:44:41 +1100
Subject: [lists.hpcoders.com.au: Re: [Swarm-Support] development priorities
 (was Re: Membership in Swarm Developmen Group)]
From: address@hidden
To: address@hidden
Date: Wed, 15 Nov 2006 17:44:32 -0500
Precedence: bulk
X-BeenThere: address@hidden
X-Mailman-Version: 2.1.5
List-Id: Swarm Support <support.swarm.org>
X-List-Administrivia: yes
Errors-To: address@hidden

You are not allowed to post to this mailing list, and your message has
been automatically rejected.  If you think that your messages are
being rejected in error, contact the mailing list owner at
address@hidden


Date: Wed, 15 Nov 2006 22:56:01 +1100
From: Russell Standish <address@hidden>
To: address@hidden
Subject: [lists.hpcoders.com.au: Re: [Swarm-Support] development priorities
        (was Re: Membership in Swarm Developmen Group)]
Mail-Followup-To: address@hidden
User-Agent: Mutt/1.4.2.1i

----- Forwarded message from Russell Standish <lists.hpcoders.com.au> -----

Date: Wed, 15 Nov 2006 22:54:08 +1100
From: Russell Standish <lists.hpcoders.com.au>
To: Swarm Support <address@hidden>
Subject: Re: [Swarm-Support] development priorities (was Re: Membership in 
Swarm Developmen Group)
Message-ID: <address@hidden>
Mail-Followup-To: Swarm Support <address@hidden>
References: <address@hidden> <address@hidden>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <address@hidden>
User-Agent: Mutt/1.4.2.1i
Status: RO
Content-Length: 3746
Lines: 81

On Tue, Nov 14, 2006 at 09:07:48AM -0800, glen e. p. ropella wrote:
> 
> Steve Railsback wrote:
> > Voluntary membership fees are the primary funding source for Swarm
> > Development Group---the way we finance services such as maintaining
> > Swarm and upgrading it to work on new operating systems and hardware,
> > and conducting SwarmFest.
> >
> > Among the projects we are considering supporting, if
> > we can afford to, are:
> > 
> >  o Updating libraries needed to run Swarm on 64-bit MacOX and Linux
> > hardware;

Porting to 64 bit should not be hard. EcoLab has always been 64 bit
clean due to a requirement to run on whatever supercomputer I had
access to. The biggest problem were the mixed 32/64 bit machines and
ensure that the correct libraries are linked. Currently Linux on EMT64
is in this boat, but I've been thought this cycle with Irix and
Solaris earlier.

I used to maintain an OSX Aqua port of EcoLab, which was never totally
successful due to poor support for BLT, but have dropped this with my
access to the console of a MaxOSX machine. Unfortunately, I can only
support OSX 10.2, as that is the latest release available to me
through SourceForge.

> >  o A reimplementation of Swarm in the C# language for use with
> > Microsoft's .NET environment and Visual Studio IDE, and the public
> > domain MONO environment;

Dunno - might be better to leave this one to the Repast guys. .NET is
weird, and Mono will never be a drop in replacement.

> >  o A reimplementation of Objective-C Swarm in the Cocoa and GNUStep
> > environments; and
> >  o Porting Swarm for use on the new Cell processor, which has a dual
> > core and 8 subprocessors.
> 

I'm sceptical of this. The subprocessors (as I understand it)
effectively act as a vector processor, and vector processors are
rarely useful in ABMs. A better target would be to ensure Swarm runs
on the dual core (or just one of the cores is threading is not
feasible - has OpenMP been implemented for Objective C yet?). This
shouldn't be hard, of the same magnitude as the 64 bit port, assuming
access to hardware. The ABMers with peculiar requirements (eg linear
algebra, or neural networks say) can then link to a Cell optimised
library that is outside of Swarm.

Perhaps a decent use of SDG money might be to donate a more modern Mac
to their compile farm, and perhaps a PS3 as well (if people are
genuinely interested in this) to SourceForge's compile farm.

Come to think of it, why can't we lobby Apple and Sony to donate some gear!


> I think the right path is to completely rewrite Swarm on commodity
> technology.  But, Swarm is mostly useless as it is, now, not because of
> the language and run-time but because it's out-dated and
> over-engineered.  It was fine for the "early days" when ad-hoc ABM tools
> were the best tools available.  At this point, there's no reason to
> adopt Swarm for any new ABM projects. But, we have a lot of new
> technology that could be exploited to bring ABM a few steps forward were
> we to rewrite it from scratch.
> 

I'm putting the finishing touches on a paper comparing EcoLab with
Swarm, Repast and Mason. Sadly Swarm didn't look too good compared with
the others, but I'm not a good enough ObjC programmer to really know
why. Nor am I a good Java programmer for that matter. I'll post a note
about this once the paper hits arXiv.

-- 

----------------------------------------------------------------------------
A/Prof Russell Standish                  Phone 0425 253119 (mobile)
Mathematics                              
UNSW SYDNEY 2052                         address@hidden
Australia                                http://www.hpcoders.com.au
----------------------------------------------------------------------------

----- End forwarded message -----

-- 

----------------------------------------------------------------------------
A/Prof Russell Standish                  Phone 0425 253119 (mobile)
Mathematics                              
UNSW SYDNEY 2052                         address@hidden
Australia                                http://www.hpcoders.com.au
----------------------------------------------------------------------------


----- End forwarded message -----

-- 

----------------------------------------------------------------------------
A/Prof Russell Standish                  Phone 0425 253119 (mobile)
Mathematics                              
UNSW SYDNEY 2052                         address@hidden
Australia                                http://www.hpcoders.com.au
----------------------------------------------------------------------------


----- End forwarded message -----

-- 

----------------------------------------------------------------------------
A/Prof Russell Standish                  Phone 0425 253119 (mobile)
Mathematics                              
UNSW SYDNEY 2052                         address@hidden
Australia                                http://www.hpcoders.com.au
----------------------------------------------------------------------------