|
|
|
|
Subject: Re: RE : running cormas on a remote computer
From: Paul Caplat (paul.caplat@gmail.com)
Date: Tue Sep 19 2006 - 15:35:56 CEST
That's a very bad new...furthermore, I read on the web that many
incompatibilities between VW and Unix systems have been found...What
do the specialists think of that ?
If it is a "GUI by network" issue, it could be fixed, but seems
extremely difficult nevertheless.
If it is a "UNIX incompatibility", it's dead. I just have to wait for
the new powerful Mac coming next month :-(
Paul
ps: too bad, shared memory computer brought the greatest hopes to me.
Imagine a world were a large scale (20km), long term (2000 years) but
still individual-based simulation would run in one hour !!!
haaaaaaahhhh...
2006/9/18, Mehdi Saqalli <msqalli@yahoo.com>:
>
> I have tried exactly the same trial by using a big computer. results were
> even poorer than my own laptop...
> I think that beyond the communication factor, in fact the calculation of
> each small step (for instance the growth of a crop in each "field" pixel) is
> not where time is loss but mainly during the "switch" from a step to an
> other (not a time step i would say but DURING the step)...
>
> Paul Caplat <paul.caplat@gmail.com> a écrit :
>
> Hi all,
> I have got the opportunity to run Cormas, remotely, on a BIG and
> POWERFUL computer. Unfortunately, the simulations appear to be quite
> slow. The reason I guess is that displaying the simulation windows by
> the network consumes more memory than the model itself.
>
> Do you think there would be a way of displaying the simulation windows
> only when asked ? I guess it'd need some changes in the kernel...???
> Paul
>
>
>
> ________________________________
> Découvrez un nouveau moyen de poser toutes vos questions quelque soit le
> sujet ! Yahoo! Questions/Réponses pour partager vos connaissances, vos
> opinions et vos expériences. Cliquez ici.
>
>