MathGroup Archive 2013

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

Search the Archive

Re: Simple program leaking memory - why?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg130761] Re: Simple program leaking memory - why?
  • From: Joe Gwinn <joegwinn at comcast.net>
  • Date: Sat, 11 May 2013 02:26:57 -0400 (EDT)
  • Delivered-to: l-mathgroup@mail-archive0.wolfram.com
  • Delivered-to: l-mathgroup@wolfram.com
  • Delivered-to: mathgroup-outx@smc.vnet.net
  • Delivered-to: mathgroup-newsendx@smc.vnet.net
  • References: <kmabu3$qdj$1@smc.vnet.net> <kmd1ag$4km$1@smc.vnet.net> <kmfks5$bqt$1@smc.vnet.net>

Gabe,

In article <kmfks5$bqt$1 at smc.vnet.net>, Gabriel Redner
<gredner at gmail.com> wrote:

> Hi Joe,
> 
> I don't see how this could be the issue.  All of the statements have
> semicolons after them, and in any case I've tried changing $HistoryLength
> with no effect.

The semicolons prevent printing, not generation.

If you did $HistoryLength=0 early in the computations, with a clean
start (exit Mathematica and launch Mathematica) and the problem persists, then it is
probably some other cause.

The most general way to figure out what's going on is to divide the
notebook into two notebooks (communicating via files if necessary) and
see which half leaks.  Do this recursively.  If neither half leaks, the
cause is something in or near the cut point.  This approach is modeled
on binary search, and is very fast albeit manual.

Joe Gwinn



  • Prev by Date: Re: Import vs Macintosh Aliases
  • Next by Date: Optimization/Minimization of multiple argument function
  • Previous by thread: Re: Simple program leaking memory - why?
  • Next by thread: Re: Simple program leaking memory - why?