MathGroup Archive 1999

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

Search the Archive

Re: Avoid out of memory in Simplify

  • To: mathgroup at smc.vnet.net
  • Subject: [mg19552] Re: Avoid out of memory in Simplify
  • From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
  • Date: Wed, 1 Sep 1999 23:06:55 -0400
  • Organization: Universitaet Leipzig
  • References: <7qfmsl$al3@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Hi,

the best solution is to by new RAM.
The second best is to wrap your Simplify[] call
with MemoryConstrained[]

In[4]:=
?MemoryConstrained

"MemoryConstrained[expr, b] evaluates expr, stopping if more than b
bytes of \
memory are requested. MemoryConstrained[expr, b, failexpr] returns
failexpr \
if the memory constraint is not met."

The third best is to do some work by hand.
a) replace some common terms by a single variable
   say you have a^2*b in several terms, the replacement
   b-> tmp1/a^2
   will save some memory in your expression
b) if you have a sum you can simplify the summands first
   Simplify /@ yourSum
   and simplify the (smaller) result again  
c) do all the things you are doing then you work with 
   paper and pencil 

Hope that helps
  Jens

"Toshiyuki (Toshi) Meshii" wrote:
> 
> Hi
> 
> Please give me an advice on my problem.
> Version of my Mathematica is v3.0.1.
> 
> I was suffering from error message "out of memory" on my Win98 AMD
> K6-400Mhz, 128MB machine, every time I run Simplify on a complex function.
> I thought this might be solved if I tried it on Linux environment.
> 
> I tried it on Redhat 5.2, PenIII 550MHz, 512MB, HDD 9GB machine.
> Sure, I got the result faster.
> But the result was the same.
> 
> Can anybody give an advice on using my resource more effectively?
> 
> Toshi
>


  • Prev by Date: Re: Double evaluation in recursive function
  • Next by Date: Re: vector of abstract class
  • Previous by thread: Re: Avoid out of memory in Simplify
  • Next by thread: Re: Double evaluation in recursive function