Hello Robert,
thank you very much for your answer.
The program is only Maxima codes consists of 3 loops (inside of the 3.
loop consists of a lot of if-conditions and equations).
The build_info() of the computer, in which the simulation has not been
finished is:
Maxima version: 5.23.2
Maxima build date: 15:35 2/4/2011
Host type: x86_64-redhat-linux-gnu
Lisp implementation type: SBCL
Lisp implementation version: 1.0.38-3.el6
and build_info() of another one, in which the simulation was finished
successful is:
Maxima version: 5.25.2
Maxima build date: 2:7 11/25/2011
Host type: x86_64-unknown-linux-gnu
Lisp implementation type: CLISP
Lisp implementation version: 2.49 (2010-07-07) (built on build05 [127.0.0.1]).
I will try , what you recommended me.
Thanks again.
Best Regards
Sara
2012/5/11 Robert Dodier <robert.dodier at gmail.com>:
> On 2012-05-10, Sara Pashmin <sarapashm at googlemail.com> wrote:
>
>> I run a maxima program parallel on two computers.
>> The program consists of 3 loops, which the data (with a lot of
>> if-conditions and equations) have to been run 76x101x101 times (I hope
>> that I explain it in English clearly).
>>
>> The simulation on one of PCs was finished yesterday after 12 hours. On
>> the other one the simulation has been run since 3 days but it seems to
>> me that the simulation has been stopped but not broken.
>
> Sara, you'll have to provide some additional information in order for
> others to understand the problem. What is the output of build_info() on
> each computer? What is the program which you are running?
>
> Without understanding anything, my first suggestion is to print the loop
> index and any other relevant data in the outer loop (the 76 x loop), so
> that you can see if you're making progress.
>
> Second suggestion is to break up the whole computation so that you don't
> have to run the whole thing in one pass. For example: load the current
> state for the outer loop, run the inner loops, save the current state,
> then exit Maxima. Then if you have to kill the program, you haven't lost
> much, and you can review the saved states.
>
> Those are just some ideas that can be helpful when making long
> computations.
>
> best
>
> Robert Dodier
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima