<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Prof Brian D Ripley wrote:
<blockquote TYPE=CITE>On Mon, 16 Jul 2001, Laurent Gautier wrote:
<p>>
<br>> So far this everything seemed to go this way (I saw the size of my
<br>> processes growing when I was using big objects and
<br>> so on). Howver recently I had trouble with the memory. It seems there
is
<br>> a limit of about 1,2 Go, beyond which R starts
<br>> to send memory allocation error messages... not consistent
with the
<br>> memory still available
<br>> (like 'Error: cannot allocate vector of size 125382 Kb', while there
<br>> still about 17Go free).
<p>Is this compiled as 32-bit or 64-bit process?</blockquote>
<p><br>This was compiled with the gnu compilers suite (gcc, g77, f77 and
gmake). If I remember well,
<br>gcc only compile 32 bits stuff (but I may be completely wrong)
<br>
<br>
<blockquote TYPE=CITE> And are there any
<br>per-process limits set? A 32-bit process will (by definition)
have a
<br>4Gb limit, and may well have 2Gb or less depending how the malloc is
<br>organised.</blockquote>
Mmmmh....
<br>I just tried a quick recursive perl hack and it seems the memory
limit for a process is around 1.7Go...
<p>I also talked to one of system administrators here, and he is not
aware of malloc specifc organisation, neither of a per-process limit (and
the man page for malloc does not tell me more)...
<br>
<br>
<br>
<br>
<p>Laurent
<br>
<br>
<br>
<br>
<pre>--
Laurent Gautier CBS, Building 208, DTU
PhD. Student D-2800 Lyngby,Denmark
tel: +45 45 25 24 85 <A HREF="http://www.cbs.dtu.dk/laurent">http://www.cbs.dtu.dk/laurent</A></pre>
</html>