[Date Index] [Thread Index] [Author Index]

Re: "No more memory "

• To: mathgroup at smc.vnet.net
• Subject: [mg30617] Re: [mg30165] "No more memory "
• From: "Carlos Cabrera Zuniga" <carloscz at ift.unesp.br>
• Date: Fri, 31 Aug 2001 04:09:42 -0400 (EDT)
• Organization: Instituto de Fisica Teorica de Sao Paulo
• References: <200108010619.CAA03891@smc.vnet.net> <010e01c13038$bd3a4920$240610ac@ift.unesp.br> <3B8DDBA2.E35CF821@to.infn.it>
• Sender: owner-wri-mathgroup at wolfram.com

   Hi Gianfranco

 The use of the swap file and any other memory use is something
 controlled by the operative system, and it (S.O) uses as much memory as
 it needs. So you cant modify the use of the swap file,
 (well, exactly, you can do inmaterial modifications)
 really, the dangerous and critical point is that you could
 have more memory needs, in an incontrolable way, this is
 what you need to avoid. Well, i think so. But you have
 a more powerfull reason: If the swap file is increasing
 too much, it could damage your HD seriously, or alter
 data in it.

   Best regards,

   Carlos Cabrera Z.

 PHD Physics Student
 Instituto de Fisica Teorica de São Paulo, Brasil.


> Hi Carlos,
>
> Thanks for your comments.
> I agree that we could change strategy, but at present we
> have to understand WHY the Mathkernel exits BEFORE having
> used the allocated swap memory.
> We are waiting for a reply from Wolfram technical support.
>
> Best regards
>
> Gianfranco Zosi
>
> Carlos Cabrera Zuniga wrote:
> >
> >  Hi Sosi
> >
> >   I have seen your file in Mathematica, i do believe that you must
redesign
> > your
> >  attack, (well, it is what i would think about). Frequently If
Mathematica
> >  cant solve a problem "easily", then it takes lots and lots of memory,
so,
> > break your
> >  work in easy parts, consider that if not, if you work with more 1000 mb
ram
> > more,
> >  possible, it would be not enough, too. Obviously it demands that you
shall
> > no more
> > follow a rutinary scheme of work.
> >
> >  Althougth i didint understand your Mathematica file, i hope this can
help
> > you,
> > I am sure it is so as i would work if i were you.
> >
> >  Carlos Cabrera Z.
>
> --
> Gianfranco Zosi
> Associate professor
> Dip. Fisica Generale "A. Avogadro"
> Universita di Torino
> via P. Giuria 1 - 10125 Torino - Italy
> ph. + 39 011 670 7426(7425)
> fax + 39 011 658 444
>
>

• References:
  • "No more memory "
    • From: zosi <zosi@to.infn.it>