Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
1997
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 1997

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

Search the Archive

Re: Memory Leak in 3.0 for students?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg8640] Re: Memory Leak in 3.0 for students?
  • From: weber at math.uni-bonn.de (Matthias Weber)
  • Date: Fri, 12 Sep 1997 04:10:53 -0400
  • Organization: RHRZ - University of Bonn (Germany)
  • Sender: owner-wri-mathgroup at wolfram.com

In article <5v2uhl$400 at smc.vnet.net>, Stephen Froehlich
<shorthnd at ccwf.cc.utexas.edu> wrote:

>         Why is it that as I do evulations, the Kernel takes up more and
> more memory?  Is there a leak?  If so, is there a fix in the works?

I don´t know whether that is the problem, but the kernel remembers
all the Out variables, and if you have lots of lengthy outputs
(even when suppressed), eventually the kernel will suffocate.

To get rid of that, use 

Unprotect[Out]; Clear[Out];

That helps a bit. Otherwise, I can assure you that also the non-student
version of Mathematica is extremely memory-consuming. I had to solve a
symbolic linear system of 10 equations in 10 variables, and the kernel
grew easily to 200MB (virtual memory) w/o producing a solution.
In this context I would be interested to have  less memory consuming
versions of some functions, even if they are much slower.

Matthias Weber


  • Prev by Date: Re: interpolating f(x,y) from evaluation
  • Next by Date: Re: interpolating f(x,y) from evaluation
  • Previous by thread: Re: Memory Leak in 3.0 for students?
  • Next by thread: Using MathLink for a remote Kernel over Telnet